-
PROPOSAl
THATWORK
5th
A Guidefor PlanningDissertations
and GrantProposals
EDITION
LawrenceF.Locke
University
of Massachusetts
at Amherst
WaneenWyrickSpirduso
TheUniversity
of Texasat Austin
StephenJ. Silverman
Teachers
College,ColumbiaUniversity
~SAGE Publications
~ Thousand Oaks
• London • New Delhi
Copyright © 2007 by Sage Publications, lnc.
All.righcs reserved. No part of this book may be reproduced or utilized in any form
by any means, electronic or mechanical, including photocopying, recording, or
by any information storage and retrieval system, without permission in writing from
or
the publi sher.
For information:
Sage Publications, Inc.
24:55 Teller Road
Thousand Oaks, California 91320
E-mail: order@sagcpub.com
Sage Publications Ltd.
I Oliver's Yard
55 Cir.yRoad
London ECl Y I SP
United Kingdom
Sage Publications India Pvt. Ltd.
R-42, Panchsh.cel Enclave
Post Rox 4109
New Delhi 110 01 7 india
Primed in the United States of Atncrica
Library of Congress Cataloging-in-Publicalion D,zta
Locke, Lawrence F.
Proposals that work: A guide for planning dissertations and grant proposals/
Lawrence F. I.ocke, Wanccn Wyrick Spirduso, Stephen J. Silverman. - Sch ed.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978-1-4129-2422-l (cloth)
lSBN-13: 978-1-4129-242.,-8 (pbk.)
1. Pl'oposal writing i11research-Handbooks , ma nua ls , etc. 2. Dissertations ,
Academic- Handbooks i man uals, ere. 3. Research grants- Handbooks,
manua ls, etc. 4. Fund raising - Handbooks, manuals , ere. f. Spirduso,
Wanecn Wyrick . IL Silver man, Step.hen J. m. Title.
Q180 .55.P7L6 3 200 7
001.4'4-dc22
2006029719
l'rinted on acid-free p(lper.
07
08
09
lO
Acquiring Editor:
Assistant Editor:
Editorial Assistant:
Production F.ditnr:
Marketing Manager:
Copy Edita,·:
Proofreader:
Typesetter:
Cover Designer:
ll
10
9
8
7
6
5
Lisa Cuevas Shaw
Margo Reth Crouppen
Karen Margretht: Greene
Sarah Quesenberry
Stephanie Adams
Halim Dunsky
Sue Irwin
C&M Digitals (P) Ltd.
Michelle Kenny
4
3
2
1
>
1
TheFunction
of theProposal
T
he dissertation process begins with the development of a proposal that
sets forth hoth the ~}(act _natu~eo( the mat.tcr co _be investigateq and
a derailed account of the methods to be employed. In addition, the proposal
usually contains material iupporting the importance of the topic selected and
the appropriateness of the research met h<)<lsto be employed.
Function
A proposal may function in at least three ways: as a means of communication, as a plan, and as a contract.
Communication
The proposal serves to communicate the investigator's research plans to
those who provide consultation, give consent, or disburse funds. The document is the primary resource on which the graduate student's thesis or
dissertation committee must base che functions of review, consultation, and,
more important, approval for implementation of the research project. lt also
serves a simi.lar function for persons holding the purse strings of foundations
or governm enta l funding agencies . The qual ity of assistance, the economy of
consultati .on, and the probability of financia l support will all depend directly
on the clarity and thoroughness of the p.i.-oposaL
3
4
Writing the Proposal
Plan
The proposal serves as a phln for action. All empirical research consists
of careful, systematic, and preplanned obJ ervations _0 some restricted _set of
phenomena. The acceptability of results is iudged exclusively in terms of the
adequacy of the methods employed in making, recording, and interpreting
the planned observations. Accordingly, the plan for observation, with its
supporting arguments and explications, is the basis on which the thesis, dissertation, or research report will be judged.
The research report can be no better than the plan of investigation.
Hence, an adequate proposal sets forth the plan in step-by-step detail. The
existence of a detailed plan that incorporates the most careful anticipation
of problems to he confronted and contingent courses of action is the most
powerful insurance against oversight or ill-considered choices during the
execution phase of the investigation. \Virh the exception of plans for some
qualitative research (see Chapter 5), the hallmark of a good proposal is a
level of thoroughness and detail sufficient to permit another investigator to
replicate the study, that is, to perform the same planned observations with
results not substantially different from those the author might obtain.
Contract
A completed proposal, approved for execution and signed by all members
of the sponsoring rnmmircee, constitutes a bond of agreement between the
student and the advisors. An approved grant proposal results in a contract
between the investigator (and often the university) and a funding source. The
approved proposal describes a study that, if condm:ted competently and
completely, should provide the basis for a report that would meet all standards for acceptability. Accordingly, once the contract has been made, all but
minor changes should occur only when arguments can be made for absolute
necessity or compelling desirability.
Proposals for theses and dissertations should be in final form prior to the
collection of data. Under most circumstances, substantial revisions should be
made only with the explicit consent of the full committee. Once the document is approved in final form, neither the student nor the sponsoring faculty members should be free to alter the fundamental terms of the contract
by unilateral decision.
Regulations Governing Proposals
All funding agem.:iesh:we their own guidelines for submissions, and these should
be followed exactly. In the university, however, no set of universal rules or
The Function of the Proposal
empirical research consists
ms of some restricted set of
:I exdusively in terms of the
recording, anc.l interpreting
n for observation, withit s
sis on which the thesis, disthe plan of investigation.
in srcp-by-step c.letail.The
.e most careful anticipation
mrses of action is the most
sidered choices during the
xceprion of plans for some
rk of a good proposal is a
·mit another investigator to
planned observations with
author might ohtain.
1
and signed by all members
I of agreement between the
oposal results in a contract
) anc.la funding source. The
onductcd competently and
t that would meet all stantract has been made, all bur
ts can be made for absolute
t
)e in final form prior to the
,stantial revisions should be
:ommittee. Once the docuent nor the sponsoring factental terms of the contract
.,hmissions, and these should
no set of universal rules or
S
guidelines presently exists co govern the form or conten t of the research proposal. There may be, however, several sources of regulatio n governing the form
and content of the finaJ rese_~rch report. The proposa l sets forth a plan of action
that must eventuate in a report conforming to these latter regulations; therefore, it is important to consider them in writing the propQsal. As we discuss later
in this chapter, understanding what the final report willlook like may help you
in completing the dissertation and submitting articles for publication.
Although it is evidenr that particular traditions have evolved within
individual university departments, any formal limitation on the selection of
either topic or method of investigation is rarely imposed. Normally, the
planning and execution of student research are circumscribed by existing
departmental policy on format for the final report, university regulations
concerning theses and dissertation reports, and informal standards exercised
by individual advisors or study committees.
Usually, departmental and university regulations regarding graduate
studcnt proposals are either so explicit as to be perfectly clear (e.g., "The
proposal may not exceed 25 typewritten pages" or "The proposal will conform to the style established in the Publication Manual of the American
Psychological Association") or so general as co impose no specific or useful
standard (e.g., "The research topic must be of suitahlc proportions" or "The
proposal must reflect a thorough knowledge of the prohlem area"}. The
student, therefore, should find no serious difficulty in developing a proposal
that conforms to departmental and university regulations.
Some universities now allow students to elect alternative dissertation
or thesis formats, such as a research paper (or series of papers) with an
expanded literature review and supporting materials in the appendix. We
discuss this in the last section of chis chapter and urge you to consider such
an option because the more compact research paper format can save considerable time in turning the completed dissertation into a publication.
Alternative formats for the final rcport, however, do not alter the need for a
complete proposal. A good study requires a sound plan, irrespective of the
format used for reporting the results.
Another potential source of regulation, the individual thesis or dissertation committee, constitutes an important variable in rhe development of the
thesis or dissertation proposal. Sponsoring committee memocrs may have
strong per~onal comrnitments concerning particular working procedures,
writing styles, or proposal format. The student must confront these as a
unique constellation of demands that will influence the form of the proposal.
It always is wise to anticipate conflicting demands and to attempt their resolution before the collection of data and the preparation of a final report.
Committei.:s are unlikely to make style and format demands that differ substantially from commonly accepted modes of research writing. As a general
6
Writing the Propo.sal
rule, most advisors subscribe to the broad guidelines outlined in this hook.
Where differences occur, they are likely to he matters of emphasis or largely
mechanical items (e.g., inclusion of particular subheadings within rhe
document).
General Considerations
Most problems in proposal preparation are straightforward and relatively
obvious. The comrnon difficulties <lo not involve the subtle and complex
problems of design and data management. They arise instead from the most
basic clements of the research process: What is rhe proper question to ask?
Where is the best place to look for the answer? What is the best way to standardize, quantify, and record observations? Properly determining the answers
to these questions remains the most common obstade to the development of
adeqL1ate proposals.
Simplicity, clarity, and parsimo ny a1·e the standards of writing that
reflect adequate thinking about the research problem . Comp licated
matters are best communicated wh en they are the objects 0£ simp le,
well-edited prose. Jn the early stag e of development, the on ly way ro
obtain prompt and helpful assistance is to provide advisors with a document that is easily and correctly understood. At the final stage, approval
of the study will hinge not' only on how carefully the plan has been
designed but also on how well that design has been communicated. In the
mass of detail that goes into the planning of a research study, the writer
must not forget tbat the proposal's most immediate function is to inform
readers quickly and accurately.
The problem in writing a proposal is essentially the same as in writing the
final report. \Vhen the task of preparing a proposal is well cxernted, the task
of preparing the final report is more than half done (an important consideration for the graduate studrnt with an eye on university deadlines). Under
ideal conditions, such minor changes as altering the tense of verbs will convert the proposal into the opening chapters of the thesis or dissertation, or
into initial sections of a rese;uch report.
Many proposals evolve through a series of steps. Preliminary discussion
with colleagues and fact1lty members may lead to a series of drafts that
evolve towarJ a final document presented at a formal meeting of the full
dissertation or thesis committee, or to a proposal submitted through the
university hi.erarchy to a funding source. This process of progressive revision
can be accelerated and made more productive by following these simple
rules:
"'
elines outlined in tl1is book.
atters of emphasis or largely
ar subheadings within the
aightforward and relatively
.ve the subtle and complex
arise instead from the most
the proper 4uestion to ask?
v'hat is the best way to stan:rly determining the answers
;tadt: to the development of
standards of writing that
h problem. Complicated
re the objects of simple,
opment, the only way to
1
ide advisors with a <locut the final stage, approval
·efully the plan has been
,een communicated. In rhe
research study, the writer
liate function is to inform
' the same as in writing the
ii is well executed, the task
1e (an important consider1iversity deadlines). Under
he tense of verbs will cone thesis or dissertation, or
ps. Preliminary discussion
to a series of drafts that
ormal meeting of the full
al submitted through the
ess of progressive revision
,y following these simple
111eFunction of the Proposal
7
them to
1. Prepare clean, updated copies of the evolvii,g proposal and ~uh111ic
advisors or colleague.~in advance of scheduled consultations.
2. Prepare an agt:nda of questions and problems to he discussed and submit
them in advance of scheduled consultations.
3. Keep a carefully written and dated record of all discussions and decisions that
occur with regard ro each item on the consultation agenda.
General Format
Guiddines for the format of proposals, even when intended only as general
ufren have an unfortunate influence on the writing process.
S uv"estions,
t>O
Once committed to paper, such guidelines quickly tend to acquire the status
of mandatory prescription. ln an attempt to conform to what rhey perceive
as an invariant format, students produce proposal documents that are
awkward and illogical as plans for action-as well as st.ilte<land tasteless as
prose.
Some universities and many funding agencies make very specifo.: demands
for the format of proposals. Others provide general guidelines for form and
content. Whatever the particular situation confronting the writer, it is vital
to remember that no universally applicable and correct format exists for the
research proposal. Each research plan requires that u:rtain communication
tasks be accomplished, some that are common to all proposals and others
that are unique to the specific form of inquiry. Taken together, however, the
tasks encompassed by all proposals demand that what is written fit the real
topic at hand, not some preconceived ideal. It is flexibility, not rigidity that
makes strong proposal documents.
Specific Tasks
The following paragraphs specify communication tasks that are present in
nearly all proposals for empirical research. Each proposal, however, will
demand its own unique arrnngement of these functions. Within a given
proposal, the tasks may or may not be identified by such traditional section
designations as "Background," "Importance of the Study," "Review of
Literature," "Methodology," "Definitions," or "I.imitations." Individual
proposals are sure ro demand changes in the order of presentation or attention ro other tasks not specified below. This particularly will be the case with
some of the tasks that are specific to grant proposals (see Part II). finally, it
is important to note that some of the adjacent tasks, shown as headings in
the following paragraphs, often may be merged into single sections.
8
Writing thc. Proposal
As you read each of the tasks below, an illustration can be found by turning to the first proposal in Part III of this guide. In rhat particular specimen,
we have edit ed the p.r:oposaJ so the sections correspond to the discussion of
each task. We have provided a critiq ue prece<ling each section of rhe specimen proposal to sumrnarize the suggestions presented in this chapter.
Introducing the Study
Proposals, like ether forms of written cornmunicacion, are best introduced by a shore, meticulously devised statement that e~tablishes the overall
a_£e,1 of concern, arouses interest, and communicates information essential to
the reader's comprehension of whar follows. The standarc.1here is a "gentle
introduction" that avoids both tediou~ length and the shock of technical
derail or abstruse argument. A careful introduction is the precursor of three
other casks {purpose statement, rationale, and background). In many cases,
it may be written simply as the firsr paragraph.(s) of an opening proposal section that includes all three.
For most proposah, the easiesr and most effective way to introduce the study
is to identify and define the central construct(s) involved. In the sense that constructs are concept!>that provide an abstract symbolization of some observable
attribute or phenomenon, all studies employ constructs. Constructs such as
"inrelligence" or "teacher entlmsiasm" are utilized in research by defining them
in terms of SDme observable event, that is, "intelligence" as defined by a test
score, or "teacher enthus.iasm" as defined by a set of classroom behaviors.
When the reader asks., "What is this stud y about?" the best answer is to present
the key constructs and explain how they will be represented in the iDvestigation.
The trick in these opening paragraphs of introduction is to sketch the study in
the bold strokes of major constructs without usurping the function of more
deta.iled sections that will follow.
Re1a6onships among constructs that will be of particular interest or
about which explicit hypotheses will be developed should be briefly noted.
Co11.structs with which the reader probably is familiar rnay be ignored in the
imro<luction, for rhcy are of less interest than the relationships proposed by
the author.
The most common error in int:ro<lucing research is failure to get to
the point-usually
a consequence of engaging in grand generaliz.arions.
For irntanee, in :1 proposed study of attributes contributing to balance ability, rhe opening paragraph might conrain a sentence such .:s "The child's
capacity to rnaintain balance is a factor of fundamental importance in the
design of elementary school curriculum." The significance of the construct
"balance'' in accomplishing motor t,1sks may make it an attribute of some
-:tration can be found by turn:. In that particular specimen,
rrcspond to the discussion of
ing each section of the speci~semed in this chapter.
1munication, are best introit that cs_E
ablishes the overalI
:ates information essential to
1e standard here is a "gentle
and the shock of technical
:ion is the precursor of three
,ackground). In many cases,
of an opening proposal secve way to introduce the study
,olve<l. Tnthe sense that conolization of some ohserv:ible
1structs. Constructs sud1 as
in research by defining them
igence" as defined by a test
set of classroom behaviors.
the best answer is to present
resented in the investigation.
ion is to sketch the study in
rp.ing the function of more
: of particular interest or
:d should be briefly noted.
iliar may be ignored in the
relationships proposed by
~arch is failure to get ro
in grand generalizations.
'.!tributing to balance ahil~nce sucl1 as "The child's
mental importance in the
nificance of the construct
<c it an attribute of some
The Function of the Proposal
9
. ortance I·n elenJentary education, but that point may be far from the
imp
f
dy invo lving balance . If, for example, the proposed study deals
I1 art o a sc:u
b
lat ionship of muscle strength to balance, observations about bal~
w 1tb r e re
· l
b1
if anyw here, in· a
factor in the desig n of schoo l curncu um e ong,
.
.
ch
h
.
a.nee as a
'
di ssion What belongs up front 1s a statement
at gets to t e pornt:
late r scu
·
·
·
h Id
.,..,ainraining static balance req mr es musc ula r act ion to o
"Thetas k Of '"'-'
. .
. .
. When
muscle strength 1s madequate co
. • 'in <a horizonta l position
•
•
•
.,
che pe Ivis
,nlis
h
chis,
performance
1s
1
mpatred.
ace Oll '.I'
• l
d
I
.
Some indication of the impo rtance of t 1e st u ~ to t 1~ory or p~act~ce may
be used to help capture che reader's interest, but m the mtroduct1on 1t 1s n<}t
explain completely all the study's significance. Present the basic
necess ary to
.
.
.
facts first and leave the detail of thorough discuss1'.m until a ~ore a_ppropnate point. Use of unnecessary technic~J lang~agc 1s another 1mped1men_t to
the reader's ability to grasp the mam idea. S1m1larly, the use of quotations
and extensive references are intrusions into what should be a dear, simple
preliminary statement. As a general rule, the_first ~aragraph of ~he introdu~tion should be free of citations. Dornmcntat1on of important pomts can wait
until a full discussion of the problem is launched.
Stating the Purpose
Early in the proposal, often in the introductory paragraph(si, it is wise to
set forth an explicit statement of your purpose in undertaking the study. We
are using the word "purpose" in its general sense as a statement of why you
want to do the study and what you intend to accomplish. Such statements
can be divided broa<lly into those related to the desire to improve something
and those reflecting a desire to understand something. In addition to such
practical and theoretical purposes, Maxwell (2005) has pointed out that, in
some instance s, it may be wise to be explicit about more personal purposes
as well, including interests related to simple curiosity, a sense of social responsibility, or career demands.
A statement of purpose need not be an exhaustive survey of your intentions, nor need it be written in the formal language of research questions
(which are much more specific expressions of what you want to learn).
An early and specific announcement of the primary target for the study,
and your purpose in aiming at it, will satisfy the reader's most pressing
questions-what
is this all about, and why is tbis study being proposed?
Succinct answers allow the reader to attend to your subsequent exposition
without the nagging sense that he or she still is waiting to discover the main
objective. Make your statement of purpose early, be forthright, keep it
simple, and be brief.
lO
Writing
I
the Proposal
Providing a Rationale
Once the reader understands the topic of the investigation an<l has at least
a general sense of your purpose, the next task is to address the question
"Why bother with that?" in terms that are rnore derailed and explicit. The
developrnent of a rationale that jusrifies the proposed srudy usually involves
both logical argument and documentation with factual evidence. The intention is to persuade the reader not onl y thar. the investigation (with its component questions or hypotheses) is worthy of attention, bur _also that the
problem has been correctly defined.
To that end, it often is helpful to diagram factors and relationships that
support your formular.ion of the proble m. Suppose that an assert ion proposed
for experimental testing is that older adul ts w ho had oxyge n therapy for six
monrhs woul d show superior cogniti ve function when com par ed to subjects
assigned to a contro l group . Th e implica tion of such an assert ion is that there
is a relationsh ip between the level of oxygen provided to the brain and cognitive capacity in older adults. The reasons for such a complex supposition
can be clarified by diagramming them in a simple form like the one shown in
Figure 1.1. Assuming rhat the consrrucrs have been defined, the rationale can
be developed by documenting the inform,Hion within each hox, and then
Cognitive
function is
better at sea
level than at
high altitude
,...._
Older adults who
are hypoxic also
exhibit cognitive
dysfunction
Oxygen is necessary for
glucose metabolism:
the brain's fuel
Oxygen therapy for six months
will increase cognitive function
in older adults
fii;ure l.l
Example of Diagram of Logic for Rarionale
Patients with
cerebral vascular
disease perform
better on cognitive
tests under high
levels of oxygen
p
'The Function of the Proposal
· investigation and has at least
;k is to address the question
>re derailed and explicit. The
:>posed study usually involves
1 factual evidence. The intene investigation (with its com: attention, but also that the
factors and relationships that
Jse that an assertion proposed
o had oxygen therapy f.or six
n when compared to subjects
such an assertion is that there
rov.ided to the brain and cog. such a complex supposition
,le form like the one shown in
,een defined, the rationale can
1 within each box, and then
Patients with
cerebral vascular
disease perform
better on cognitive
tests under high
levels of oxygen
J
~
:J
ationale
11
"n.ingthe enormous practical consequences that would attend a positive
exp laj
.
.
.
.
finding . A sound rationa le 1s one th at eoovrnces the reader that you are ra isin the right question-and that the answer is worth finding .
grn most cases, this ea rly attention to justifying the proposed st ud y should
be Jiiniced co the basic matters of defining what is to be studied and why
ic is worth so doing. These reasons may be practical, theoretical, or both
and should be presented economically. The detail of rationale for particular
choices in methods of darn collectio n and analysis can be deferred until such
matters ace discussed in subsequent parts of che proposal.
Formulating Questions or Hypotheses
All proposals must arrive at a formal statement of questions or hypotheses. These statements should be written in carefully constructed language
that specifies each variable in explicit terms. A statement such as "Studying
each day should result in improved learning" is better written as "Sixty minutes of studying each day will result in significantly increased scores on
a standardized test of achievement." These statements of questions or
hypotheses may be set aside as a separate section or simply included in the
course of other discmsion. Such statements differ from what was contained
in the statement of the purpose in that (a) they are normally stated in formal
terms appropriate to the design and analysis of data to be employed, and ·
(b) they display, in logical order, all subsections of the research topic.
The question form is most arpropriate when the research is exploratory,
when it is impossible and inappropriate to state hypotheses, or for qualitative studies where the question format is much more appropriate. The
~esearcher should i!,ldicatc by the specificity of questions, however, that the
problem has been subject to thorough analysis. By careful formulation of
questions, the proposed stu<ly should he directed toward outco1m:s that are
foreshadowed by the literatu re or pilot work, rather than toward a scanning
of potentially interesting findings.
The hypothesis form is employed when the state of existing knowledge
and theory permits formulation of reasonable predictions about the relationship of variables. Hypotheses ordinarily have their origin in theoretical
propositions already established in the review of literature. Because the proposal must ensure that the reader grasps how the relationships expressed in
theo"i-yhave been translated into the form of testable hypotheses, it often is
useful to provide a succinct restatement of the theoretical framework at a point
contiguous to the presentation of formal research hypotheses.
The most common difficulty in formulating a research question is the
problem of clarity. Students who have read and studied .in the area of their
i
12
Writing th e Proposal
topic for weeks or months often are distressed to discover how difficult it is
to reduce all they want to discover to a single, unambiguous quesrion.
Tbe clarity of a research question hinges on adequate specificity and the
correct degree of inclusiveness. The major elements of the investigation must
be identified in a way that permits no confusion with other elements. At the
same rime, the statement must maintain simplicity by including nothi_ng
beyond what is essential to identify the main variables and any relationships
that may be proposed among them. Questions for quantitative studies, for
examp le, must meet three tests of clarity and inclusiveness:
·1. Is the question free of ambiguity?
2. Is a rdarionship
:imon g variable s expccsse<l?
3. Do es rhe question imply an empiri~I
lc::st?
Applying these standards to the question "Does a relationship exist
between self-esteem and reading achievement in chilaren!" mfght appear to
identify the scudy's main eTements in reaso nab ly clear fashion. Self-esteem
and reading achievement are variables, and children are the subject popu lation. A relationship is suggested, and correlation of self-esteem and reading
scores clearly is implied as an appropriate empirica l test of the relationship.
The constructs of self-esteem and reading achievement, however, arc quite
broad and might be taken by some readers to indicate variables different
from chose intended. These potential sources of ambiguity might be resolved
without destroying the simplicity of the question by altering it to ask, "Does
a relationship exist between scores on the Children's Test of Self-Esteem and
scores on the reading portion of the Tri-State Achievement T est?" Whether
it also might be importanr to provide more specificity for the generic word
«children" would depend on whether tne intent was to examine self-esteem
and read ing in a particular type of child. rf not, the generic word would be
adequate, but if so, the importance of that variable calls for more careful
specification in rhe question.
In the case of qualitative research (discussed at length in Chapter 5),
because pee-established hypotheses arc seldom used, questions are the tool
most commonly employed to provide focus for thesis and dissertation studies. Although there is disagreement among scholars about the use of formal
questions in qua litative research, there is no escape from the need to have
a question (whecher explicit or implicit) rhat will serve to direct what is
observed or who is interviewed-at lease at the outset of the study .
The quesrion(s) .frequently are phrased in ways that make them appea r
very different from rhose used in che natural science model (and, thereby,
The Function of the Proposal
ssed ro <liscover how difficult it is
1gle, unambiguous question.
;s on adequate specificity and the
elements of the investigation must
:usion with other elements. At the
simplicity by including nothing
in variables and any relationships
tions for quantitative studies, for
nd inclusiveness:
J3
discrepant with some aspects of the advice given in this chapter). Some, for
paraexa nlp le, will sound highly generalized, as in the following examples
·
phrased from qualitative proposals.
1. What is going on in this urban school classroom?
2. How do professional wrestlers understand their work?
3. What does residence in a hospice mean to a parient?
Other question statements reflect the intention to use a particular theoretical framework in the study.
sed?
,t?
tion "Does a relationship cxisr
:nt in chi@ren?"n1ight afpear'
to
)nably clear ·fashion. Self-esteem
children are the subject populalation of self-esteem and reading
!mpirical test of the relationship.
achievement, however, arc quite
rs to indicate variables different
s of ambiguity might be resolved
stion by altering it to ask, "Does
hildren's Test of Sdf-Estcem and
te Achievement Test?" Whether
specificity for the generic word
ttent was to examine self-esteem
not, the generic word would be
: variable calls for more careful
1.
·ussed at length in Chapter 5),
om used, questions are the tool
for thesis and dissertation stndcholars about the w;e of formal
> escape from the need to have
,at will serve to dircct what is
the outset of the study.
1 ways that make them appear
al science model (and, thereby,
1,
What perspective do medical students adopt to make sense of their experience in medical school?
2. How do gay and lesbian soldiers manage the presentation of their sexual
preference within the social setting of their workplace?
J. I low do social roles influence the interaction between teachers and students
as they attempt to rcali:zc personal goals in the classroom?
In contrast with quantitative research, questions in a qualitative proposal
often arc treated as more tentative an<l contingent on the unfolding of the
study. Ncvcrthelcss, their careful formulation is no less important. They must
give initial direction to planning, bring the power of theoretical constructs
to the process of analysis, and reflect the degree of sophisticated thought
employed in determining the focus of inquiry.
Experienced qualitative researchers sometimcs do, in fact, elect not to
package their curiosity, interests, concerns, and foreshadowings into the
form of explicit research questions. Graduate students, however, embarking
on their first attempt within the qualitative paradigm, often find that their
advisors are greatly reassured when the proposal contains a careful accounting of what the data are expected to reveal that is not already known. In
other words, it is a good idea for the novice to explicate the questions that
motivate their interest, thereby firmly grounding the study in the conventions of scholarly inquiry. How a qualitative investigator's assumptions ahout
the world, and about research, serve to shape those questions will be addressed
in Chapter 5.
Research hypotheses differ from research questions in that hypotheses
both indicate the question in testable form and predict the nature of the
answer. A clear question is readily transformed into a hypothesis by casting
it in the form of a declarative statement that can be tested so as to show it
14
Writing d ,e Proposal.
to be e ither true or false. Get.ting pre cisely th e hy pothesi s tha t is want ed ,
ho we\'er, ofte n is more exac tin g than it app ears.
Unlike a ques tion, the hypothesis exerts a dire ct influen ce on each sub sequent step of rhe stL1dy, from design to pr epar at ion of th e final report.
By spe cif) ing a prediction about outc ome, the hypothe sis creates a bridge
between the theor.etical consiclerntions chat underlie th e question and the
cnsuing rcsearch proce ss designed to produce the answer. The investigator is
limited to p roce <llircs that wi ll tcst the truth of the proposed relationship,
and an y implications to be deduced from th e results will rest entire ly on th e
pa rriculax test selected. Because it exerts suc h powerful a priori influence, a
hyp oth esis denrnnds consid erabk at tention at the sta rt of a scud}' but makes
it easier to pres erve obj ectivit y in the Later stages of design and exec uti o n.
Aside from ~pet:ific imp ac t on design of the stud}', the gen eral advantage
of the hypothesis over the ques tio n for quantitative studies is that it permit s
more power ful a nd persu asive condusions . At the end of a study, a research
que stion nev er permits the investigator to sa}' more tha n " Herc is how th e
world .looked when Tobserved it.' ' ln. contrast, hypotheses permit th e inves tiga to r to say, "Ba sed on my particular exp lana tio n of how th e world works ,
this is what I exp ect ed to observe, and beho ld-that js exac rly how it loo ked!
For that reason my explanation of how the world works musr be given credibili ty ." Wh en a hypo thesi s is confirmed, the investig ator is empower ed to
mak e arguments about knowl edge that go far beyond what is ava ilabl e when
a question has been asked and answered.
We wo uld be rem iss herc if we did no t not e the curr ent debate among
researchers abou t the va lue of hypo theses and statisti ca l significanc e testing.
It has been argued (Schmidt, 1996; Thomp son , 1996, 1997; Thompson &
Kieffer, 2000) that statis tica l sig nificance testing (one step in the pr ocess of
testing hypotheses ) has certain technical Limita tions. For some studies, at
least, o ther types of analyses, suc h as examinin g effect sizes, might provide
grea ter benefits. That d ebate is beyond the scope of chis text. Wbat is certain, how ever, is that graduaw st udt:nts should discu ss the matter with a dvi sors an d co o1mitree members until a consens us em e rges th at meets bo th their
ex pectati ons and rhose (if any) of the graduate school. Whether hypo t heses
a re teste d o r question s are used to guide the research , they sho uld be written
with the grea test can : for precision and must be exactly appropriate to the
pm pos es of the stu d y.
A hypot hesis can be written eithe r as a nllll state ment (co nvenienrly ca lled
a null hypot hesis), such as "T here is no difference - berween ... , " or as a
directional statement indi cati n g the kind of relati o nship anric.ipated (called a
resea rch or directional hypothesis), such as "W hen this, also that" (positive)
o r "When this, not that" (negative) . Many arguments favor the use of
The function of the Proposal
e hypothesis that is wanted,
,.
rect influence on each subse1aration of the final report.
hypothesis creates a bridge
iderlie the question and the
e answer. The investigator is
f the proposed relationship,
sults will rest entirely on the
owerful a priori influence, a
1estart of a study but makes
s of design and execution.
;tudy, the general advantage
tive stlldies is that it permits
1e end of a study, a research
nore rhan "Here is how the
1ypotheses per.mit the inve.son of how the world works,
hat is exactly how it looked!
:I works must be given credvescigator is empowered to
1ond what is available when
· the current debate among
uistical signifirnnce testing.
1996, 1997; Thompson &
(one step in the process of
:ions. For some studies, at
: effect sizes, might provide
e of this text. What is ceriscuss the matter with advi1erges that meets both their
:hool. Whether hypotheses
rch, they should be written
exactly appropriate to the
:ement (conveniently called
1cc-between ... ," or as a
mship anticipated (called a
n this, also that" (positive)
~uments favor the use of
l5
.
.
lity because it permits more persuas ive logic a nd more statistica l
d1rect1ona
.
'd
. I£a pilot stud y has been comp 1eced or t.be I'1terature review
prov, es
power. ·easoning for a directional resuJt, then directional hypotheses are
.
. J l
I .
di
sn·ong r
·
appro pri ate . In some instances, part1cu ar y eva uauon Stu es, prac1
clear
y
'
·
f
d.
·
I
1
1
.
F
.
.
tters may dictate use o a irectiona 1ypot 1esJs. . or instance, 1'f
oca 1 ma
.
.
apy program is being evaluated and the only practical consequence
a cher
'd
. .
d .
h
uld be find ing that therapy prov1 es greater gams 111stress re uct 1on t an
;: program in cwTent use, a directional hypothesis would permit a direct
est of this singular outcome.
t Some of the technica l debate about the form of hypotheses is beyond the
scope of this guide, bm a good rule of thumb for the novice is to ~_rn(lloy
directional hypotheses when pilot data are available that clearly indicate a
direction, or_:'.V
hen the t~ e~ry from which ~he,hypotheses were drawn is sufficiently robust to indude some persuasive evidence -~or. directionality. If the·
invesdgation is a preliminary exploration in an area for which there is no
well established theory, and if it has been impossible ro gather enough pilot
data to provide modest confidence in a directional prediction, the format of
the null hypothesis is the better choice. Ultimately, as a researcher pursues a
line of questioning through several investigations, directional hypotheses
become more obvious and the null format less attractive.
Hypotheses can be evaluated by the same criteria used to examine
research questions (lack of ambiguity, expression of relationship, and implication of appropriate test). ln ad<lition, the statement must he formulated so
that the entire prediction can be dealt with in a single test. If the hypothesis
is so complex that one portion could he rejected without also rejecting the
remainder, it requires rewriting.
Several small, perfectly testable hypotheses always arc preferable to one
that is larger and amorphous. For example, in the following hypothesis the
word "but" signals trouble. "Males are significantly more anxious than
females, but male nurses are not significantly more anxious than female
teachers." The F test for the main effect of sex in the implied analysis of variance (ANOV A) will handily deal with males and females, but a separate test
as a part of a factorial i\NOVA would be required for professional status.
Should the tests yield opposite results, the hypothesis would point in two
directions at once.
Similarly, the presence of two discrete dependent variables foreshadows
difficulty in the following example: "Blood pressures on each of five days
will be significantly lower than the preceding day, whereas heart rate will not
decrease significantly after Day 3." The implied multivariate analysis of variance (MANOV A) could not rescue the hypothesis by indicating whether we
could accept or reject it. The required follow-up test might reject the blood
16
Writing the Proposal
pressure prediction while accepting it for. heart rate. In all such cases, division into smaller, unitary hypotheses is the ohvious cure.
When a number of hypotheses are necessary, as a result of interest i11
interaction effects or as a conse quence of employ ing more than one depe.ndcnr variable, the primary hypo theses shou ld be stated first. T hese primary
statements may even be separated from hypotheses that are secondary or
confirmatory, as a means of giving prom inence to the main intent of the
study.
Finally, hypotheses should be formu lated w ith an eye to the qua l itat ive
characteristics of available measure ment too.ls. ff, for example, rl1e hypothesis specifies the magnitude of rela tionship between two variab les, it is
essential that this be supportable by the reliabi lity of the scores for the proposed instrumentation. Returning to the ear lier examp l.e of self-estee m and
reading, the fact should be consider ed that the correlat ion between scores
from two tests cannot exceed the square root of the product for reliability in
each test. According ly, if reliabil ity of the self-esteem test is .68 anc.l that of
the reading test is. 76, then a hypothesis of a posit ive cor.re.lation greater th.an
.80 is doomed to failure ('/.68 x .76 = .72).
Delimitations
and I ,imitations
In some cases, a listing of delimitat ions and limi tations is .required to clarify the proposed study. Delimitarions describe the pop ulat ions to which s-~oeralizations may be safely made. The genera lizability of the study will be a
function of rhe subject sample and the analysis employed . Deli1:Ji: Gterall,.
means to define the limits inherent in rhe use of a particu lar co nstruct or
population.
Limitations, as used in the context of a research proposal, refer to limiting conditions or restrictive weaknesses . They occur, for example, when all
faccors ca nn ot be contro lled as a pan of study design, or when tl1e optimal
number of observations simply canno t be made because of problems involving ethics or feasibifoy. lf the investigat or has given careful thought to tlwse
prob lems and has determ ined that the information to be gained from the
compromised aspect of the study is nevertheless valid and useful, then the
invest igator proceeds but duly notes th e limitation.
All st udies have inherent delimitation s and limitations. Whether these are
listed in a separate section or simply discussed as they arise is an individual
decision. 1f they are few in number and perfectly obvious, the latter is desirable. Whatever format is used, howeve r, iris the investigator's responsibility
to understand these constraints and to assure the reader that they have been
considered during the formulation of the study.
,
The fun<.:tion of the Proposal
art rate. ln all such cases, <livibvious cure.
:sary, as a result of inrerest in
1ploying more than one depen1 he stated firsr. These primary
>othescs that arc secondary or
~nee ro the main intent of the
l7
·cl:ng
Definitions
prov1
.u
U oposa ls for .research usesyst~m atic languagt that may be spe(;ific to
A f-~~ of research or to chat propo sal. We discuss the use of definitions
rhat ie er detail in the section
·
· lec:I "('I, antv
· an <l Prec1s1on:
· ·
ot· Chapter 6· t1t
in grea t
,,
<in(r
i.n
System
Language.
Speal. t>
'
Discussing the Background of the Problem
with an eye to the qualitative
s. If, for example, the hypothebctwccn two variables it is
bility of the scores for th~ proier example of self-esteem and
:he correlation between scores
of the pro<luct for reliability in
f-esteem test is .68 and that of
ositive correlation greater than
AnY research prob lem must show its lineage from the backgro und of
e.Xiscing knowledge or previo us investigations, or, in rhe case of applied
research, from conte111p0rary practice.
The author
must
answer
three
questions:
J. What do we already know or do? (The purpose here, in one or two sentences,
is co support the legi.r.imacy and importance of the quesrion. Major discus-~ions of the importance and significance of the study will come under the
"rationale for the study" sectio11.)
2. How docs chis particular question i:elatc to what we already know or do?
(The purpose here is to explain and suppoi:t the exact form of questions or
hypotheses chat serve as the focus for the study.)
limitations is required to clarthe populations to which genizability of the study will be a
sis employed. Delimit literal ly
e of a particular c6 nsrruct or
-earch proposal, refer to limit. occur, for example, when all
y design, or when the optimal
le because of problems involvgiven careful thought to these
nation to be gained from the
ess vali<l and useful, then rhe
1tion.
limitations. Whether these are
I as they arise is an individual
dy obvious, the latter is dcsir\e investigator's responsibility
·he reader that they have been
J. Why select this particular 1nethod of investigation? (The purpose here is to
explain and support the sclec.cions made from ,\Jnong alternative
methods
of invcscigarion.)
In reviewing the research literature rhat often forms the background for the
study, the author's task is to indicate the main dircctions taken by workers in
the area and the ma.in issues of methodology and interpretation that have
arisen, Particular attention must be given to a critical analysis of previous
methodology and the exposition of the advantages and limitations inherent in
various alternatives. Close attention must be given to conceptual and theoretical fonnulations that arc explicit or implicit within the selected studies.
By devising, whe n appropr iate, a theo.rerica l basis fo.r rhe study chat
emerges from the scructme of existing know ledge, by ma kin g the questions
or hypotheses eme rge from the total matrix of answered aud unanswered
questions, and by making the selection of me th od contingent upon prev i.ous
resu lts, the aut hor inserts the proposed study into a line of inquiry and a
developing body of know ledge . Such carefu l attent ion to background is the
first step in entering the continuing conversation that is science.
The author should select only those studies that provide a foundation for
the proposed investigation, discuss these studies in sufficient detail to make
l8
Writin g the Proposal
their relevance entirely clear, note expJkicly the ways in which they contribute ro rhe proposed research, and give some indication of how the proposal is designed to move beyond earlier work. The second section of Chapter 4
provides guidelines for preparing che literature review.
ft is important for students and novice proposal writers co resist the
impulse to display both the extent of their personal labors in achieving what
they know and the volume of interesting, but presently irrelevant, information accumulated in the process. The rule in selecting studies for revi_ew is
exactly tbe same as that used throughout the proposal-funit
discussion co
what is essential to the main topic . A complete list of all references used in
developing the proposal (properly call.ed a bibUograpby as distinct from the
list of references) may be placed in an appendix, thereby providing both a
service ro the interested reader and some psychological relief to the writer.
We should note, however, that many dissertation committees will think the
references are all that is needed and including both a reference section and a
bibliography would be overkill.
Whenever possible, the author should be conceptually or theoretically
clear by creating 'organfaing frameworks that encompass both the reviewed
studies and the proposed research. This may take the form of something as
obviou s and practical as grouping studies according to certain methodological features (often for rhe purpose of examining divergent results) , or something as esoteric as identifying and grouping the implicit assumptions made
by various researchers in formulacing rheir statement of the problem (often
for the purpose of clar .ifying the problem selected in the present proposal).
[n many proposals, creating an organized conceptual framework represents the most important single opportunity for the application of original
thought. ln one sense, the organizing task is an extension of the need co
achieve clarity in communication. A category system that allows division of
diverse ideas or recond ite events into easily perceived and remembered
subsets is an organizational convenience for the author, as well as for the
reader. Beyond convenience, however, organizing frameworks idenrify_ distincrive threads of thougnt. The task is to isolate the parallel ways by which
researchers, working at different times and in varying degrees of intellectual
isolation, have conceived of reality . In creating a schema that deals meaningfully with sin1ilarities and dissimilarities in the work of others, the author nor
only contributes to the body of knowledge but also deals with the immediate needs of communicating this research to others .
Even relatively simple organizing or integrating systems de.mand the
development of underlying conceptual plans and, often, new ways of interpreting old results and presumed relationships. The sequence of variables
in the swdy may provide a simple and generally adequate place to begin
The Function of the Proposal
icirly the ways in which they conve some indication of how the proork. The secon d section of Chapter 4
rature review.
rice prop osa l writers to resist the
r personal labors in achiev ing wha t
, bur pres entl y irrelevant , ioforma e in se lecting studies for review is
: th e proposal-limit
discussion t~
nplet e list of all referenc es used in
1 bibliography as distinct from the
,pendix, thereby providin g both a
psych o l.og ical relief to the writer.
~nation committees will think the
ing both a reference section and a
be conceptually or theoreticallv
1at encompass both the.:reviewed
1y take the form ot something as
1ecording to certain methodologining divergent results), or some1g the implicit assumptions made
statement of rhe problem (ofte n
:leered in the present pr op osa l).
!d concept ual fram.ewock repre y for the application of orig ina l
is an extension of the.: need to
·y system that allows division of
-ily perceived and remembered
as for the
r the author, as wc.:11
~izing frameworks identi_fydis>latc the parallel ways bv which
.1 varying degrees of intellectual
1ga schema that deals meaning) work of others, the author not
1
ut also deals with the immedi~thers.
egrating systems demand the
and, often, new ways of inrerps. The sequence of variables
:rally adequate place to begin
19
· w Such questi ons as '' What is the relationship between
d
,,
.
. g the rev 1e ·
.. .
,1rra11
g,n
d choo l achievemen t wh en ab1luy 1s 11el consta nt? consist
·
soc1·a 1class anI s d within a convenient
sequentla· l <l'1agram. In tum, s ue h
.
prs pace
.
.
.
.
or conceual sc bema't·:i oft en contam usefu l asswnpa o ns about causal relat1on.
co .nceptnd' t hus can serve as effec tive precursors to exp l.anatory theory.
. The
ships a ·
. k" d of researc h proposa ls ac hieve exa ctly that sort of link age,
• '
.
..
.
st Jegant in
mo e f. ewo rk for organizing the review of hteratur e as a bridge con. g the ram
l spec 'f'
-'
usin
. ·
knowl edge a proposed theory , an d t1e
, 1c, ·u1eory
necring e>0snn g
. . '. .
o
theses
ro
be
empmq1
lly tested .
1
base d -I YP
Explaining Procedures
All proposals for em~irical research must embody a plan for the careful
. •
·itic observation . of events. The .methods
for such obseranuJ sysr<:m,
. selected
.
.
.
determine
the
quality
of
data
obta111ed.
i:or
this
reason,
the portion
vauons
·
.
dealing with procedures the.: researcher rntc.:n<lsto employ
f h ropos·,1
otcp
'
...
-..
,·t\ be.:subJ·ect to the closest cnt1cal scrutiny. Corrc.:spond111gly,the presenta~\ I
.
·1. 'J'he <j'1scuss10n
.
tion
of methodology requires great attc.:nt1on
to <fera,1
of
method must include sources of <lata, tht collection of data, and the analysis of data. In addition, the discussion must show i:hat the specific techniques
selected will not fall short of the claims established in previous sections of
the proposal.
The section(s) dealing with methodology must be freely adapted to the
purpose of the study. Whatever the format, however, the proposal must provide a step-by-stc.:p set of instructions for conducting the investigation. For
example, most studies demand explication of the following items:
J. Identification an<l description of the target population and sampling methods
to be used
2. Presentation of instrumenrs and techniques for measurement
3. Presentation of a design for the collel.:tion of data
4. Presentation of procedures for collecting and recording <lata
5. Explanation of data analysis procedures to be used
6. Development of plans for contingencies such as subject mortality
Many justifications for particular method selections will c.:mcrge in the
development of background for the problem. The rationale for some choices,
however , will most convenientlv , be [)rcscnte<l when the.: method is introduccd as part of the investigation plan.
20
Writing the Prnpo.sal
fn describing such elements, prop osals can includ e pages of description
that fatigue and frustrate the reader wit hou t yielding a dear picture of the
overa lJpattern . In ma11ycases,this probl em can be avoi ded by the use of diagrams. Although Figure 1.2 disp lays a co unte rbalanced treatment design of
moderate comp lexfry, it would ,require no more than a brief paragrap h of
accompanying texr to prov ide a clear acco unc of the p roced ure.
Diagrams are he lpful when presenting statistical mode ls that will be tested
later, once the data are coUecred. No'te how dearly the interre lations of a
hypothetical statistical model appear in Figure 1.3. In the figure, the ovals
represent clusters of variables, the boxes show the variab les in each cluster,
and rhe various arrows represent interrelatio nships. Imag ine how many
words it woul d rake to describe all of those relationships ! Given a brief
exposure co these figures, however, most readers wou ld find fu rther explanation unnecessary.
Provi<ling Supplementary
Material
Fox the purpose of clarity and econo mical pxesentation, many items may
be placed in appendices keyed to appropriate references in the main text .
So placed, such materials become options avai lab le to rhe r:eadex as needed,
rather rhan distractions or impediments to unde rsta nding rbe main themes of
the proposal. focluded in the appendices may be such items as the following:
L Spc:cificarions
for cquip111er1t
2. ln.struc:tions to subjects
3. Lctte.rs a11dother rel.evam documents
4, Sl1hject consc.ut forms
S. Raw data or tabular rnarerial fror:n pilot studies
6. T ::ibular m:nerials from related research
7
Copies of paper i~ncipc11cillnstr.un1erus
8, Questions for sr.rucrurec.1interviews
9. Crcdenrials of experts, judges, or other special personncJ r.o be cmployc<l in
rhe .study
JO. Dingr<1mmaricmodels of the research design
I l. Diagramma1ic models of the stati.~ticill ::inalysis
12. Schemar.ic.~for cousrrncred eqll ipmenr
I J. Chapter <>ntli11efor rhc final 1·eporr
H. Propo.~cd time schcdllle for executing the stL1dy
l 5. Supp!erne11tary bibliographics
The Function of the Proposal
,ls can include pages of description
·hour yielding a clear picture of tl1e
::m can he avoided by the use of diaJunterbalanc.:ed treatment design of
10 more than a hrief paragraph of
:ount of the procedure.
statistical models that will be tested
how clearly the interrelations of a
~·igurc 1.3. ln the figure, the ovals
show the variables in each cluster,
dationsl1ips. Jrnagine how many
hose relationships! Given a brief
readers would find furtl1er expla-
GROUP INSTRUCTION
FORM COMPLETION
RANDOM ASSIGNMENT
INTERVIEW
FILM VIEWING
FORM COMPLETION
:al p.resentarion, rnany items may
·iate references in th e main text.
1vaila ble co the read er as needed,
in.derstandi.ng the main themes of
.y be such items as the following :
Figure 1.2
Example of Method Flow Chart
Figure 1.3
Example of Statistical Model
studies
ecial personnel ro he employed in
gn
ilysis
cuc.ly
21
22
Writing the Proposal
Completing the Tasks: The Proposal and the Report
As we have indicated in earlier sections of this chapter, different universities
and funding sources have different requirements for completing the tasks
disn 1sscd above. Some universities requi re a short pre-proposal, a prospectus, prior to complcring a full proposal; others require a full proposal; and
still others leave it up to the college or department co determine what is
accepcablc. We strongly urge you to get the documents that govern the
requirements to which you will be held-and read them as early as possible
in this process. Ther e is no need to create an initial structure for your proposal and then convert to the one for which you will be held accountable.
We also think it is to yom advantage co think ahead about how you
will-once rhe data are collected and analyzed-turn your proposal into che
<lisserration an<l then into one or more research reporrs that can be submitted to acadcmi<.:journals . Nor coo many years ago, on most campuses, there
was lictle latitud e in how the dissertation was organized . Typically, the completed d1~enarion had five or six chapters organized as follows: (a) an introductory chapter with introduction to the study, the purpose statement,
the rationale, the questions or hypotheses, and, in some cases, limit.atious,
delimitations, and definitions; (b) the review of the lirerature (i.e., the
background); (c) method; (d) results; and (e) discusi,ion and concl usions (the
latter may be divided into separate cbaptcrs). Ideally, when this format is
used, the first three chap ters of the proposal-introducti on, literature
review, and method-would be updaced and then the results, discussion, aod
conclusions simply would be appended to cornplerc the disserrarion or thesis document.
One of the problems inherent in organizing a dissertation in che five- or
six-chapter format, however, is char substamiaJ revisions will have to occur
to rurn the dissertation into one or more research articles that can then be
submitted to journals for consideration as a published article Uensen, Marrin,
& Mann, 2003; O'Brien, 1995; West, 1992). We have seen many scudencs,
Ollr O\\c1, included, who have take n monchs or years co tum a dissertation into
an article. Others, faced hy the substantial task of preparing an entirely new
document for publication , have nor been willing to revisit their dissertation co
rake this next step. lu some cases the result of that reticence must be counted
as a genuine loss to both the author and the body of knowled ge.
The purpose of this hook is co help you navigate the tasks of planning and
executing a dissertation. Beyond rhar, however, we have co me to believe that
ear ly planning can both expedite and encourage the vital process of sharing
whar is learned. Getti ng rhc task done and graduating with a degree is the
first priority, but co stop rhere is co leave yourself unfini shed .
The Function of the Proposal
posal and the Report
· this chapter, different universities
rements for completing the tasks
! a short pre-proposal, a prospec•thers require a full proposal; and
department to determine what is
· the documents that govern the
tnd read them as early as possible
an initial structure for your pro:h you will be held accountable.
to think ahead about how you
'zed-tum your proposal into the
earch reports that can be submitars ago, on most campuses, there
as organized. Typically, the com)rganizcd as follows: (a) an intro! study, the purpose statement,
and, in some cases, limitations,
view of the literature (i.e., the
·) discussion and conclusions (the
·rs). fdeally, when this format is
oposal-introduction,
literature
1 then the results, discussion, and
::omplere the dissertation or the.ing a dissertation in the five- or
1tial revisions will have to occur
=Search articles that can then be
n1blished article (Jensen, Martin,
:). We have seen many students,
r years to turn a dissertation into
:1.skof preparing an entirely new
ing to revisit their dissertation co
,f rhat reticence must be counted
body of knowledge.
vigate the tasks of planning and
er, we have come to believe that
:age the vital process of sharing
graduating with a degree is the
urself unfinished.
23
·versities now permit-and some even en.courage-dissertations
·
• . e from the traditio nal format . As with proposaJ regu lat ions, we
·
nI ar. devlat
. should understand what options
are ava i·1a ble ro you an d d'1scuss
· a f orrnat ava1·1a blet hat
beheve you
·
··
irh you advisor early m the process. If· t her:e LS
fV[,111y uni
th~rnw dire wrning your dissertation into resea rch articles, we urge you
t0
will expe
.
.
. .t
1 close co ns1deranon.
g,v~ ble 1.1 shows how che proposal can be converred to a traditional disa·on and then ro a format that makes it easier co revise the disserra rion
sertat1 •
.
.
.
researc h articles . (The pamcular for mat used here 1s on ly one among a
· ) l rr _,_:
_ ·1
·
f'guinto,ber now in grow111g
· use in
· h'1gber eclucat1on.
w_u_:, a ternatlve con 1
nun
.
.
I
.
d
.
l
I
tion the fast chapter or sect ion ts a genera intro uct10n t1at secs up t1e
ramp lete scudy. The next chapter or chapters are potential articles represent~og individual part s of the study, in this case repor ts conta ining the review of
;~eracure, the design and methodology , and the findings. Th e decision as to
the acrual number of publishable ar ticles would be a function of the particu lar disserradon or thesis, as well as the ava ilabi lity of appropriate venues for
dissemination.
following the chapters that appear in the form of individual articles, the
dissertation would present a general discussion and condusions, followed by
referenc.:esand appendices. Where the content of articles does not include all
of the material that would be essential to the proposal, as, for example, might
be the case for a review of literature, a complete version simply can be placed
in an appendix.
The artide chapters, with one last edit and the addition of references, can
be quickly converted to article for.mat and submitted for publication .
Table 1.1
J'0
~
.'Vfovingfrom the Proposa.l to the Dissertation
Dissertation Proposal
Traditional Disser/.ation
Inrroducrion
Iutroducrion
Purpose
Rationale
Questions or hypotheses
Limirations/dclimirarions/
definitions (if included as
separate sections)
Introduction
Introduction
Purpose
Rationale
Quesrions or hypocheses
Limitarions/ddimitarions/
definitions (if included as
separare sections}
Dissertation l'repared to
Facilitate J'ulJlishing
Chapter or Section
2
3
Review of Lireracure (complete)
Review of Literature {complete)
Method
Merhod
4
Results
Discussion and Conclusions
7
Article 11
Arricle 2
Article 3
5
6
Incroducrion
Inrrnducrion
Purpose
Rationale
Questions or hypotheses
Limitations/delirnitarions/
definitions (ii included as separ.ate
~sections)
~
References
References
Appendices-including some or all
of the information found in the
list on page 20.
Appendices-including some or all Appendices-induding a complete
of the information found in rhe
review of literature and some.-or all
list on page 20.
of rhe informllrion found in rhe list
on pase 20.
References
1
Complex designs may involve a set of related inqu ires, each of which r.eprescnts .i legitimate and reportable r,'search invesrigarion. Thus, there rnay be 011e
or. several anic.:les, the numher depen<ling on rhe nar.ure of the srudy. Eacli anide consisrs of m;He1·i:ildrawn from rl,e set"cn c/)aprer.; o,· scaious and each
consrim1cs a cha plcr in the dissenatio,, document. In som~ cases, the review of literature also may be included as on.e of rhe puhlish.,hle reports.
discover or have strong reason to
colle ague or faculty member? Our
~lly, about how to protect yourself
ler the terrib le consequences of an
1ot you take action vtill be a matter
d tell you that cheating in research
course of action you choose, howrudence, and courag e.
1e course of action. If you judge that
)thcr options. In most uni versities,
~gain the omb ud spcrso n will kn ow
~ feder al r.esearch funds will h ave
research ethics. ln some instanc es,
e-blowe r " (Miceli & Nea r, 1992),
arion has a pr op er claim to d ue
s of eth ical viola ti ons that can not
und to the discomfort (an d often
:sting account see Spra gue , 1998).
o f Responsible Science (Panel o n
t o f Resear ch, 1992) before you
;y, but at least it can he done the
1sibility. Every stud y is part o f a
). When anyone docs not do wha t
3
Developing
theThesisor
Dissertation
Proposal
Some Common Problems
T
he gener a l purp o ses and broad format of the pr oposa l document have
now been pr esented . There .remain, howev er, a numb er o f particular
points thar cause a disproportionat e amount of diffic ulry in preparing
proposals fo r stud ent -co nduct ed resea1·c h. In some cases, rbe problems ar ise
because of rea l diffic uJty in the subcle and complex natur e of the writin g
task. In oth er cases, however , rhe problem s are a consequence of confusion,
con flictin g o pinions , and ambiguous standards among research workers
themselves and, mor e partic u larly, among univ e rsity research advisors.
As with many tasks invo lving an eleme nt of art, it is possible to establish
a few general rule s .co whi ch most practitioners sub scri be. Success in te rms
of real mastery, how ever, lies nor in knowing, or even fo llowing, rhe rules
but in what the student learns co do wirhin th e rul es.
Each student will discover h is or her own set of special problems. Some
will be so lved only throu g h pra ctice and th e accumulation of experie nce.
While wrestling with the frustrations o f preparing a proposaJ, you sho uld
rry to rememb er that t!1e ~~?I fascination of resea rc h lies in its problematic
nature, in the searc h for serviceable hypothe ses, in se lecti ng sensitive means
of a nalyzing data , and in the creative ta sks of study d esign.
Some o f the pr o blem s gra duat e students face can not be so lved simp ly by
reading about che m. What follows, ho wev er , is an effort to alert yo u to the
41
42
Writing the Proposal
most comm on pitfalls, to provide some genera l sugges tions for reso luti on of
the prob lems, and to sound one encourag ing note : consu ltation with col.
leagues and advisors, patience with the ofte n slow process o_f"figuri ng our,"
and sc rupul ous care in writing w ill overcome or circum venr most of the
problems encoumered in preparing a research proposal. In the midst of dif.
ficulty, i.t is usefu l to remember that problems are better encounte red when
developing the proposa l rban when facing a deadline for a fina l copy of the
report.
The problems have been grOl1ped into two broad sections : "Before
the Proposal: First Th ings First" and "T he Sequence of Pcoposing : Froin
Selecting a Topic to Forming a Committee . '' Each sectio ,n conta ins a numb er
of specific issues that may confrom che student researcher an d provides some
mi es of thumb for use in avo iding or reso lving the atte ndant difficulties. You
shou ld skim through the two sections se!ecrive ly, because not all th e discussions will be relevant to your needs. Chapter 4 ("Content oI the Proposal:
Important Considerations"), Chapter 6 ("Sry le and Fo rm in Writing the
Proposal"), a nd Chap ter 7 ("T he Ora l Presentation"} deal with specific technical problems and shou ld be consu lted after completing a review of what
fo llows here.
Before the Proposal: First Things First
Making Your Decision: Do You Really Want to Do It?
The following idealized sequence of events leads ro a thesis or dissertation
proposal.
I. ln the process of cornpleting umlcrgrn<luore or rnaster's level preparation, the
student identifies an area of particular interest in which he or she proposes to
corn:enrrate advanced stu<ly.
2. The student selects a graduate in.;titution rhat has a strong reputation
re.~c,\rch and teaching in the area of interest.
for
3. ·rhe stude11t iclcmifies an advisor who h:1spublishe<l extensively and regularly
chairs graduate student research in rhe area of interest.
4. Based on forrher srudy and imcraction wirh the advisor, the student select.~and
formulates a question or hypothesis as the bas.isior ;:i rhesis or dissertarion.
Because we do not live in the best of all possible worlds, few students are
able to pursue the steps of this happy and logical sequence. For a variety of
reasons, most students have to take at least one of the srcps in reverse. Some
Oe.ve.lopingthe Thesis or Dissertation Proposal
some general suggesrions for resolu .
.
encourag,ng
note·' const iltat·
. tlon Of
. ·
· 100 With
l the often slow pr ocess of "f · .·
CoJ,
·'jj .
.
tgu1ing Ot.t ~
r overcome or c1rcumvenc ti
t,
I ost of th
: a researd1 proposal. In the midst of d' e
it_p~oblems arc bett er encountered Whif.
tacmg a deadline for a final copy oft~;
Ded into two broad sections : "13 ,
d "Th ·
e~ore
n
c.; .Sequence of Proposing· F
· , ,, r
.
· .ro111
m1ttce. ,.:,;icJ1section conra ins a nu b
,
<l
•
m e.r
1e -~tu ent resear cher and provides
·
some
.
reso Ivmg t11c atrendan r difficu lties y
s . J • ) b
. OLJ
. se ect1ve Y, ccaus e not a'J the d'
a
~
.
i~~
. ,a~;er 4 {"Comenr of the l)roposal:
i . 6 ( Style and Form in Writing th
I Presentation"). deal with specific tech~
ted after completing a review of what
ings :First
Really Want to Do ft?
~venrs leads to a thesis or dissertation
l~trate or_masr~r'slevel preparario11, the
l1eor sl1cp . .
r rntere~t in wh1cJ1
·
roposcs to
Jtion that has a strong reputation for
Herest.
1as puhJi~hc<l extcmivdy and regularly
· :irca of •merest.
vith the advisor, the student sckcr.~and
he hasis for a thesis or dissertation.
possihlc worlds, few -~tudents arc
logical sequence. For a variecv of
'. one of the steps in reverse
s),
..• -< me
i
43
elves at rhe end of severa l semesters of study just beginning to
·
·· · ·
1
b l
f
f.111c1 rhems
ven ·Fy
, . ·y area of inte.rest, 1nan mst1tutton t '.lat may e ess t 1an pei-pnmar •
.
.
.
11
idcnn a
. te co their needs, and assigned to an adv isor who has little or
d
.
h'
.
I appropria
(t-ct Y ' .·
e in that particular omam. For t is unfortu11ate state of affaii:s,
110 expel ienc asy solution. We do believe that one significant decision is, or
..
Of(er no e
we
1 . vailable ro the srudenr-t he decision to do, or not co do, a
shouldd
,e, ady· Faced with conditions sucb as those described above, if the
1r-, sru
rese,
. av
. ailabJe, the more rational and educationa lly profitable course may
rion is
t
op
I not to undertake a research study. You can detennfoe whether this
be
co
e
before the school is selected, or at least before the program
option ~\vailable
,s
f study is selected.
0
There are sound reasons to believe _that experience in the conduct of
and subrtscarc.-11 contributes to graduate educat10n.
.
.There .also are good
.
.
t
'al
reasons
co
believe
that
other
kinds
ot
expencnces
are
immeasurably
•
•
•
•
sran 1
and profitable for some students. The question ts,· "Which
,nor·e appropriate
.
· 1l t fo1·}'<)11'"
experience 1s· ng
•
If you are, or think you might be, headed for a career in scholarship and
higher e<lucation, then the decision is dear. The sooner you begin accumulating experience in research activities, the better. If you are genui nely curious about the workings of the research process, interested in combining
inquiry witl1 a career of professional service, or fascinated by the problems
associated with a particular app licaci,on of knowledge to practice, again the
decision is dear. An experience in research presents at least a viable alternative in your cJucat.ional plans.
Lacking one of these rnotives, rhe decision shou ld swing the other
way, toward an option more suited to your needs. Inadequa tely motivated
research tend s not to be comp leted or, worse, is finished in a pedestrian fashion far below the student's real capaciry. Even a well-executed thesis or dissertation may exert a powerful negative influence on the graduate experience
when it has not been accepted by the student as a reasonable and desirable
task.
O.ne problem touches everyone in graduate educatio n, facLLlty and
stud ent:s alike-the hard co nstraint s of time. Students want to finish their
degree programs in a reasonab le period of time. The disposition or circumstances of some, however, may define reasonable time as "t he shortest possible time." Others £ind the thought of any extension beyond the standard
number of semesters a serjous threat to their sense of adequacy. For students
such as these, a thesis or dissertation is a risky venture.
Relatively few research studies finish on schedu le, 31\d time requirements
invariably are underestimated. Freq uent setbacks are almost inevitable.
This is one aspect of the research process that is lea1med during the research
44
Writing th~ Proposal
experience: Haste in research is lethal to both qL1alityof the product and worth
of the experience. If you cannot spend the time, deciding to initiate a research
project endangers the area of inquiry, your advisor, your institution, your education, your reputarion, and any satisfaction you might take in completing the
task. In short, if you can't afford the time, then don't do it at all.
Choosing Your Turf: Advisors and Areas
Once a firm decision has been made to write a thesis or
choice of an advisor presents a less difficult problem. Here,
dictates selection because it is essential to have an advisor
edgeable. rurther, it always is preferable to have one who
lishing in the domain of interest.
dissertation, the
area ot interest
who is knowlis actively puh-
Competent advisement is so important that a degree of studem flexibility
may be required. Ir is far better for students to adjust their long-range goals
than co attempt research on a topic with which their advisor is completely
un familiar. It may be necessary for the thesis or dissertation to be part of the
advisor's own research program. As long as the topic remains wirbin the
broad areas of student interest, however, it is possible to gain vital experiem;e in formulating 4ucstions, designing studies, and applying the technology and methods of inquiry that are generic to the domain.
It is desirable for student and advisor to interact throughout the development of the proposal, beginning with the initial sclecti.on and formulation of
the question. On occasion, however, the student may bring an early stage
proposal to a prospective advisor as a test ot his or her interest or to encourage acceptance of forrna l appointment as adv isor . Experience suggests that
this strategy is most likely co produce imm ediate results if tbe proposal is in
rhe primary interest area of the advisor . If the proposal involves repl ication
of some aspect of the advisor's previous research, the student may be amazed
av the inte nsity of attention this attracts .
Your Question:
What Don't We Know That Matters?
Finding
Before launc hin g imo the prnccss of identi fying a suitable topic for inquiry,
we suggest a short co'tu-se of semantic and conceptua l hygiene. The purpose
of this sma ll therapy is ro establish a simple and reliable sec of terms for thinking through what can sometimes be a difficult and lengthy problem-what
do
I study?
All research emerges from a perceived problem, some unsatisfa<.:tocy sitllation in the world that we wanr w confront. Sometimes the difficulty rests
Developing the Thesis or Dissertation Proposal
:>ot~1quali~ of the product and worth
1e tlln~, dec1ding to initiate a research
ir advisor, Your institution your,.. I
'
"t Uion you might take in completing th
,
.
e
I t do 1t at all.
, t hen Lon
l Areas
o write a thesis or
:ult problem. Here,
to have an advisor
. to have one who
dissertation, the
area of interest
who is know]is actively puh-
rhat a degree of student flexibility
its to adjust their long-range goals
which thefr advisor is completely
sis or dissertation to be part of the
: as the topil: remains wid1in the
it is possible to gain viral cxperitudies, and applying the technolic to the domain.
interact throughout the develop:itial selection and formulation of
tudenc may bring an early stage
,f l1is or her interest or to cncourdvisor. Experience suggests that
·diare results if the proposal i.s in
·he proposal involves replication
irch, the student may he amazed
)
ying a suitable ropic for inquirv
•nceptuaJ hygiene. The purpose''
d reliable set of terms for thinkand lengthy problem-what
do
,lcm, some unsatisfactory situSometimes the difticulty rests
45
t.act th·tt
, we don't understand bow . things. work and
. . have the
~·ifllPIY
. ·cc
h to l<11ow • At other times
.
' we are confronted by dec1s1onsor the
1
ht1n1an
when the alternatives or consequences a.re unclear. Such per•
for acnon
d'
·1·b .
d.
.
need
bl 5 are experienced as a 1sequ11 num, a 1ssonance m our cog.
d
pro
em
·
·
l
Id
b·
·
ccive
. however, they do not eXJstout 111t 1e wor , ut tn our min· ds.
· · n· Nonce,' olind at first like one o.f t hose "mce
· points'
· ' o f w h'1.ch aca de111no
Thar may s
·
h l
' somea·mes fond, but for the purposes. of a novice researc er, . ocat-f
.
are
1111
cs
. blem in the right place and semng up your understandmg o
10
·ing che Ph is unsatisfactory may represent muc h more th an an ar b'm a ry
eel)' w at
ti
d
c)(a .
Thinking clearly about problems, questions, hypo 1eses, an
c:xerc'.se
h. urposes can prevent mental logjams that somerimes block or delay
.
b .
.
I
researc p
·
·d tification of what 1s ro e invest 1gatec.
clear t en ·
JI
vice will encounter research reports, proposals, and even some we h
T e no
.
"
hi " d "
d J rextbooks that frcdy mterchangc the words pro em an quesre"a r e
.
I
f
.
(
.
"Tl ,e quest10n
. in.
. " ,, ·u1 W<.
avs that create all sorts of logica con us10n
non
. . as 111
.
.
. i:he
111
. Stll
( I\IS
problem m this stuc.ly
dv1 is· ro investigate the problem of ... " or "J he
•
•
.IS to InV
· estr'gate
the
question
of
...
").
The
problem
1s
locate<l
alternately
111
•
.
•
the world or in the study, the distinction betw~en ~rohlerns and quesnons 1s
unclear, and what is unsatisfactory in the situation 1s nor set up as a clear target for inquiry.
.
.
.
We suggest that you be more careful as you tlunk through the question of
what co study. Define your terms from the start and stick with them, at least
until they prove not to be helpful. The definitions we prefer are arbitrary, but
it has been our experience that making such distinctions is a useful habit of
mind. Accordingly, we suggest that you use the following lexicon as you think
and begin to write about your problem.
ProfJlem.-the experience we have when an un:;arisfacrory situation is encountered. Once carefully defined, it is rhat .~ituation, with all the attendant questions
it may raise, that can become the target for a proposed stu<ly. Your proposal,
then, will nor lay om a plan to study the prnblem hut will address one or several
of the questions that explicate whar you have foun<l "problematic" ahour the situation. Note chat in this context neither situation nor problem is limited co a
pragmatic definition. The ohservarinn that two theories contradict each other
can be experienced as a prohlein, and a research question may be posed to addre.~s
the conflict.
Question-a .~tatement of what you wi.~h ro know about some unsatisfactory situation, as in the following: What is the relation between . , . ? Which is the
quickest way ro ... ? What would happen if ... ? What is the location of ... ?
"Wh::it is the perspective of ... ? As exph1incd below, when cast in a precise,
answerable form, one or several of these questions will become the mainspring
for your .mrdy-rhe formal research question.
46
Writing the Proposal
l'ur{Jose-the explicit intention of rhe invescigaror to accumulate data in such .i
way as ro answer the research quesr.ion posec.1as the focus for the study . The
word "ohjective" is a reasonable synonym here . Although only people can have
imencions, it is common to invest our research design wirh purpose (as in "The
purpose of rhis study is to determine the mechani sm through whic h ... ").
Hy{Jothesis-an affirmar.ion about the narure of some situation in the world.
1\ tentarive proposition scr up as a convenient rarget for an invesrigation, a state-
ment to be confirmed or denied in terms of the evidence,
Given this lexirnn, the search for a topic hecomes the quest for a situation
that is sufficiently unsatisfactory to be experienced as a problem. The pro.
posal has as its purpose the .setting up of a research question and the establishment of exactly how (and why) the investigator intends co find the
answer, thereby eliminating or reducing the experience of finding something
problematic about the world. Prohlerns lead to questions, which in turn lead
to the purpose of the study and> in some instances, to hypotheses. Table }.1
shows the question, purpose, and hypotheses for a study. Note that
rhe hypotheses meet the criteria established in Chapter 1 and arc the most
specific.
The research process, and thus rhe proposal, begins with a question.
Committed to performing a srudy within a given area of inquiry and allied
with an appropriate advisor, students must identify a question that matd1es
their interests as welI as the resources and constraints of their situation. Given
a theoretically infinite set of possible problems that might be researched, it is
no small wonder that many students at first arc overwhelmed and frozen into
indecision. The "f can'r find a problem" syndrome is a common malady
among graduate students, bur fortunately one that can be cured by time and
knowledge.
Research questions emerge from thr ee broad sources: logic, practicality,
and accident. In some cases, the inves tigator's cruiosity isd i.recred to a gap
in the logical structure of what alread y is known in the area. In other cases,
the invesrigaror responds ro the demand for information about the application of knowledge to some practical service. In yet othtr cases, serendipity
operates and the investigator is stimulated by an nnexpectcd observation,
often in the context of another study. lr is common for several of these factors to operate simultaneously to direct attention to a particular question.
Personal circumstance and individual style also tend to dictate the most
common source of questions for each researcher. Finally, all the sources
knowledge of
depend on a more fundamental and prior factor-thorough
the area.
Developing the The sis or Dis.serta tio n Proposal
invesrigator to accum ulate dara in such a
)n posed as the focus for the study. the
1ym here. Although o nly people ca n have
resea rch design with purp ose (as in "T l
.
l(!
1e mec hamsm rhr oug h which .. . ").
n~rure of some situation in the world.
enient target for an investigation, a star e.
is of th e evidence.
.
·fa ble 3. 1
~s
f'roblum-
47
proble m , Question, Purpose, and H ypot heses
ive reacher planni ng of lessons rcqui n:s large investm ents of time
·in<loften mu st compet e with ocher important responsibi lities-both
d eocrgy, '
I
an
f . n:1l a d person a .
pro es,5 10 , 11
..
1<
. n-fs r.heam ount o r kind of lesson planning don~ by tea chers positi vely
st10d stud ent in-class learnmg beh avior.~such as time-on -task?
fcJare co
Q1 1e
The purpose of this srudy is to ex amine the rela tio nsh ips between several
)pie becomes the quest for a situation
experienced as a problem. The pro::,fa resea rch question and the estabhe investigator intend s to find tbe
. the ex perience of finding some thing
lead co questions, which in tuJ:n lead
=insta nces, to hypo theses. Tabl e 3.I
ypothcs es for a study . No te that
;hed in Chapt er 1 and ar e che most
proposal, begins with a questio n.
n a given area of inqu iry and aHied
use identify a questi on that matches
constra ints of their situa tion. Given
, /ems th at might be researched, it is
·st are overwhelmed and froze n into
" syndrome is a common malady
one tha t can be cured by time a11d
broad sources: l~ gic, practicality,
ttor's curiosity is directe d to a gap
known in the area . In other cases
.
'
·o r in formatio n about the ap plicaice. In yet o ther cases, sere ndipity
·d by a n unexpected observation,
common fo r seve ral of these fac ttention to a particular question.
le a lso tend to dictate the most
searcher. Fin ally, all che so urces
facto r-th orough kno wledge of
furpose--·es (types) of reacher lesson planning and student time-on-t a sk in a high
~~00
sc Iioo I a
.
.
ucomobile mechani cs class.
.
H-.,potie
I ses
. (Note char. di rectio. nal hypo theses are used fo r Hypoth eses .1-3 and .tha r
.,,c:veoHvpothesis
4, state<l m th.cnull form, could he ba sed on data tro111a p1lor
,
scudy.)
1. The uumbe r of teacher lesson planning decisions chat relate ro <lesign
and use o f active learnin g strate gies will be positively rela ted to stud ent
rime-on -cask when those lessons nee implem ented.
2. The number of dass managem ent planning decisions relat ed to partiu ,lar
lessou co mponents will be positively related to student
time-on -task wh en those comp onents arc im plc:mcnreJ.
3. Teacher lesson planning decisions chnt require ~rudems to w ait for the
::ivailability of tool s or work sites will be negativel y relat ed to student
rirne-on -task when those lessons are implem ente d.
4. T he tora l numb er of teac her p launing decisions (irrespective of category )
will not be relate d to stu dent time-on -task wh en those le.•;sons are
implemented.
It is thi.s latter factor that account s for the "graduate stud ent syndrom e."
Only as one grasps the general framework -a nd the specifi c.:detail s of a
particular ar ea ca n unknowns be revealed, fortuit ous observati ons raise ques tions, and possible applications of knowledge become apparent. Traditional
library study is the firsr step toward the maturity that permits confident selection of a resear ch question. Such study , how ever, is necessa ry but not sufficient. In any active area of inquir y, rhe current knowledge base is nor in the
library-it is in the invisible college of informal associarions a mong researc h
workers .
The worki ng knowled ge base of an area tak es the form of unpubfo d1ed
paper s, coofere.nce sp eeches, sem inar rranscripts, memoranda, dissertations
in progress, gra nt applications, per sonal rnrr espondcnce, telephone ca lls,
48
Writing the Proposal
and electronic mail communication s, as well as conversations iu tl,e corr idors of conference centers, restaura nts, hot el rooms, and bars. To obtai n
access to this ephemeral resource, the student must be where the action is.
The best introduction co the curr ent stah~ of a research area is close association with advisors who know the terr itory and are busy formulating and
pursuing their own questions. Conv ersing with peers, listening to professor.
ial discussions, assisting in research projecrs , attending lect ures and con.
ferrnccs, exchanging papers, and wrr esponding with facu lty or students at
other insritutions arc all ways of capturing the elusive state of the arr. In
all of these, however, the benefits derived often depend on knowing enough
about rhe area co join the dialogue by asking questions, offer ing a tangible
po1nc for discussion, or raising a point of cr iticism . In research, as elsewhere,
tli e more you know, the more you can learn .
Although establishi ng a network of excha nge may seem impossib le to
young students who view themselves as novices and outsideJs, it is a happy
fact that new recruits genera lly find a warm welcome within any well-defined
area of intensive study. Everyone depends on informal relati onships among
research colleagues, and thjs rapport is one so·urce of susta ining excitement
and pleasure in the resea1·ch enterprise . As soon as you can aJ:ticulate weUformu lated ideas about possible problems, your colleagues will be eager to
provide com ment, critical ques tions, suggestions, and encou ragement .
In rhe final process of selecting the thesis o.r dissertation problem, mere is
one exercise that can serve to clarify the relative significance of competing
questions. Most questions can be placed within a gene.ral modeJ that displays
a sequence of related questions-often
in an order determined by logic or
practical considerations. _Smaller questions are seen to lead to larger and
more general quesrions, methodological questions are seen necessarily ro
precede substantive questions, and theoretical questions may be found interspersed among purely empirica l questions . The following .is a much simplified but entire ly realistic exa mple of such a sequenria l mode l. It begins with
an everyday obse r vation and leads throu gh a series of specific and interrelated problems to a high-order question of great sign ificance .
OBSERVJ\TION: Older adulrs generally lake longer than you11g a<lults to complete cognitive tasks, but those whu arc physically accivc seem to be quicker
mentally, e.~pecially in casks that demand behavioral .~peed.
J. What cypes of cognitive function might be related to cxercisc?
2. I-low can these cognitive fuuctions he measured?
3. What are the effects of habitual exercise on one of these types of cognitive function-reaction
timc?
Developing th~ Thesis or Dissertation Proposal
; well as co11versations in the corribote J rooms , and bars · To obt a1n
.
udenr:_must be w_here the action is.
-~cus of a research area is close asso·1tor'. and are busy formulating and
g_with peers, listening to professor.
J 1ects, attending lectures and con.
,_onding with faculty or students at
·mg the elusive stare of the art. Cn
l often depend on knowing enougl 1
1ki_n~_qllestio.ns, offering a tangible
criticism . fn research, as elsewhere,
arn.
~xchangc may seem impossible to
ioviccs and outsiders, it is a happy
n welcome within any well-defined
' on informal relationships among
1c source of sustaining excitement
s soon as you can articulate well;, tiur colleagues will he eager to
st1ons, and encouragement.
is or <li.sserration problem, there is
:el~tivc significance of competing
,chm a general model that displays
an order determined by logic or
s are seen to lead co larger and
1ue~rions are seen necessarilv to
cal questions may be found ;,;terThe following is a much simpli. sequential model. It begins with
1 a series of specific and interregreat significance.
longer than young adults to cornsically active seem ro he quicker
,avioral speed.
1t be related to exercise?
measured?
se on one of these types of cog-
49
A.re active older. adults faster on a simple reaction time cask than sc<lenrary older adults?
5. Are active older adults faster on a more complex reaction ciinc task,
such as choice reaction rime, than older sedentary adults?
QVESTJON:What effect does habitual exercise have on choice reaction rime
in older adults?
By making the rwists and turns of specu lation visible in the concrete
rocess of sequent ial listing, pceviously unnoticed possibilities may be
;.evealed or tentativ e impressions confirmed . [n tbe simple examp le given
above, rhe reader may immediately see ocher questions that could have been
inserted or alrernative chains of inquiry that branch off from the main track
of logic. Ocher diagrammatic lists of questions about exercise and cogn itive
function might be consrructed from different but related starting points.
One might begin, for example, with the well-established observation that
circulation is superior in older individuals who exerci~e regularly. This
might lead through a series of proximal experiments toward the ultimate
question, "What is the mechanism hy which exercise maintains cognitive
function?"
Building such diagrams will be useful for the student in several ocher
ways. It is a way of controlling the instinct to grab the first researchable
question that becomes apparent in an area. Often such questions arc inferior
co what might be selected after more careful contemplation of the alternatives. A logical sequence can be followed for most questions, beginning with
"What has robe asked first?" Once these serial relatiooships become clear
.
'
it is easier to assign priorities.
In addition to identifying the correct ordering and relative importance of
questions, such conceptual models also encourage students to think in terms
(if a series of studies that lru ild cum11latively toward more significanr conclusions than can he a~hieved in a one-shot thesis or dissertation. The faculty member who has clear dedication to a personal research program can
be a key factor in attracting studenrs into the long-term commitments that
give life to an area of inquiry.
Researchable questions occur daily to the active researcher. The problem
is not finding them but maintaining some sense of whether, and where, they
might fit inro an overall plan. Although this condition may seem remote
to the novice struggling to define a first research topic, formulating even a
modest research agenda can be a helpful process. The guidance of a sequential display of questions can allow the student ro settle confidently on the target for a proposal.
50
Writing the Proposal
The Sequence of Proposing:
From Selecting a Topic to Forming a Committee
A Plan of Action: \!\/hat Follows What?
Figure 3.1 presents a plan of ani on for deve loping a proposal. ft can
be useful for the novice if one cen tra l point is understood. A ridy, linear
sequence of steps .is not an accurate p icture of wha t happens in the developmenr of most research proposals. T he pecu liar qua lities of human thought
processes and the serendipity of retri eving knowledge serve to guarantee that
development of a proposal will be a nyth ing but tidy. Dizzying leaps, periods
of no progn:ss, and agonizing backt racking are more typ ical than is a continuous, unidirectional flow of events. The diagram may be used to obtain
an overview of the task, to establish a rough time schedule, or to check rer.rospcctivcly for possible ornissions, but it is nor to be taken as a literal representation of what should or will happen.
To say that devclopmenr of a proposal is not a perfectly predictable
sequence is not co say 1 however, chat it is entirely devoid of order. Starting
at the beginning and following a logica l sequence of thought and work has
some clear advantages . When the prop osal has been completed, a backward
glance often indica tes that a more or derly pxogression through the development steps would have saved time and effort.
for instance, although the mind may skip ahead and visualize a specific
type of measure to be used, Step l1 ("Consider alternative methods of data
collection") should not be undertaken until Step 6 ("Survey relevant literature") is completed. M.any methods of measurement may be revealed and
noted while perusing the literature. Sometimes suggestions for instrumentation materialize in unlikely places or in studies that have been initially cuegorize<l as unlikely to yield informat ion concerning measurement.
Additionally, repo1ted evidence of the reliabrlity and validity of the scores
from alternative procedures will be needed before any final selection can be
made. Thus, a large commitment of effort to consideration of alternative
methods can be a waste of time if it precedes a careful survey of the literature.
for simplicity, ma ny important elements h.ave been omitted from hgure
3.1.. No reference is made to such pivotal p rocesses as developing a theoretical framework, categorizing lirerat ure, or stating hypotl1eses. fourthcr, the
detailed demands that arc intrinsic to the writing process itself, such as esrablishing a systematic language, receive no mention. What are presented are
the obvious steps of logic and procedure-the
operations and questions that
mark development toward a plan for acr1or1. f-inally, the reader who begins
to make actual use of tl1c diagram will find that the sequence of steps at
Developing tl1eThesis ur Dissertation Proposal
SI
. ·unctures leads into what appear to be circular paths. For examp le,
1g a Committee
1
s~ve,;
Jest.ion F a s ingle form of inquir y does
not present itself as most
:iate, the exi t line designated "NO" leads back co the previous pronPP1
iopJseep of considering alternarive forms of inquiry. The intention in this
cecura ment is not to 111cl
· 1·,cate a trap ·1n w I11c
· I1 b eg11111111
· · g ,esearchers are
arrang e
ed forever to chase their tails. In eac h case, the closed loop suggests
doomthat when questions
.
. J •mput 1s
• req uire
• d
can not be answerec,:I a dc11·t1ona
00 1
Y_e st udy thought, or advice), or that the questi.on itself is inappropriate
(mOI
>
·
the case and must be altered.
co for the most part, Figure 3.1 is self-exp lanatoty. We have assumed that
students will be working with, and obtaining advice from, their advisor as
they navigate the various steps. In the pages that follow, however, we have
sclecce<la few of the steps and questions for comment, either because they
represent critical junctures in th~ proposal process or bec~wse they have
roven particularly troublesome tor our own advisees. It will be helpful to
iocate in the diagram sequence each of the items selecte<l for discussion so
that the previous and succeeding steps and questions provide a frame for our
comments.
it
at?
· developing a proposal. It can
it is understood. A tic.ly, linear
>i what happens in the developiar qualities of human thought
ow ledge serve to guarantee that
>Ut tiJy. Di,:zying leaps, periods
are more typica I than is a conliagram may be used to obtain
time schedule, or to check ret'lOt to be taken as a literal repis not a perfectly predicrn ble
tirely devoid of order. Starting
1ence of thought and work has
ts been completed, a backward
>gression through the devdopahead and visualize a specific
er alternative metnods of data
tep 6 ("Survey relevant literaurement may be revealed and
s suggestions for instrumentas that have been initially catc1 concerning
measurement.
lity and valic.liry of the scores
fore any final selection can be
) consideration of alternative
:areful survey of the literature.
we been omitted from Figure
esses as developing a theoretting hypotheses. Further, the
g process itself, such as estabtion. What are presented are
)perations and questions that
inally, the reader who begins
·hat the sequence of steps at
.J~
Step J: Narrow down. ·~what do I want to know?" Moving from general to
.~pecificis always more difficult for the beginner than is anticipated. It is he.re
that the student first encounters two of the har<l facts of scientific Iife: logistic practicalir.y and the perverse inscrutability of seemingly simple events.
Inevitably, the novice must learn to take one small step, one manageable
question, at a time. In other words, the proposal must confori11 Tr·~-scope to
the realistic limitations of the research process itself. At their best, research
tools can encompass only limited bits of reality; stretched too far, they produce illusion rather than understanding.
It may be important to think big at first, to puzzle without considering
practicality, and to allow speculation to soar beyond the confines of the sure
knowledge base. rrom such creative conceptual exercises, however, the
researcher must return to the question, "\Vhere, given my resources and the
nature of the problem, can I begin?" Delimiting questions such as "In which,
people?" "Under what conditions?" "At what time?" "In what location?" "By
observing which events?" and "Ry manipulating which variables?" serve the
necessary pruning function.
Step 5: Identify reasons answer is important. This step places the proposed
research in scientific-societal perspective. The study shoul<l contribute to th~
generation or validation of a theoretical structure or Sllbcomponent or relate
to one of the several processes hy which knowledge is used to enhance
52.
Writing the Proposal
TWENTY STEPS TO A PROPOSAL
BEG/NHEAE
(1)
Browse .
Converse.
Think .
.
- - - - --- ..- ..- - -- ....- - - ..-
,.
-~
(2j
(7)
Idea Sprouts .
"What if. . • ?"
"Why does . .. ?"
Write article .
Submit to journal .
(3)
Narrow down.
"What do J want
lo know?"
NO
(4)
YES ,
Formulate a clear,
specific question .
(5)
' NO
Identity reasons
answer is
impo rtanl.
YES
iG)
'(~s-- ··
-·
Survey
relevant
literature .
(8)
NO
'
·--------------- ---
lritensive review
of literature .
~ _•• • - - •• - - - -
NO
(9)
\~-
--
- -,
forms of inquiry.
-,,
YES
Consider alternative
'•,,
--Y~,S'
NO
(10)
Y~~
----..l.,-..R_o_l_Jg
_ll_o_u_t_d_es
-ig
- n-,
of study .
(11)
Consider alternative
methods of
data collection .
r-- ---NO
~,
YES
• - -- >- STEP
12
Developing the Thesis or Dissertation Proposal
•SAL
51
(12)
fAOM
Consider alternative
methods of analysis .
---- - - - -----
ouESTiON (G)
NO
-----------·----- -- -- ....- ..- ..- - . - - . -
,,--~~~
------
(13)
,_
j
Review and
refine design .
·-...
"€"s··.
(14)
Specify all
rrocedures in detail.
Return to appropriate
step between 8 and
14 if remediable .
--
--,
I
I
I
I
I
I
I
(15)
I
I
I
Prepare first
full draft of proposal.
YES
'I'
'I
'
'I
I
(16)
I
Share and discuss
with colleagues
and advisors .
' NO
(17)
Conduct pilot study,
analy1e data, and
roview all procedures .
(18)
,,
,
,,
-{'<-;~
---,
Prepare revised
draft proposal.
YES
(19)
Present to
committee . Explain
and support.
.-
YES
'llo
NO ,
--- .
-...-
-
(20)
Gather d<.1ta
. Process
and interpret as
planned in proposal .
EX/THERE
Figure 3.l
YES
·· ··· > STEP
12
NO
Twenty Steps to a Proposal
NOTE: "!foxesrcprcscnr major p,:o(:e<lurnl Sll~ps,and unhrnkcn lines crace the main scquet,ce of
rhose sreps. Circles represent r.he major questions to bt' confronted, and broken li11esleaJ to che
pr.o.:edural conseqtJences of the alternative YES or NO answer.,.
54
Writing the.: Proposal
professional pi:.acti~e.The trick here ·is to jus tify the question .in terms appro.
priace to its nature . Inquiry that is directed toward filling a gap in th e struc.
ture of knowledge need not be supported by appea .ls co practical app licat ion
(even though later events may yield just such a return} . Inquiry that ar ises
direct ly from problems in the world of practice need not be support ed by
appea ls to improve tutderstanding of basic phenomena (even though later
events may lead ro thi s). Each kind of question has its own correct m easure
0£ imporrance. The task of distingu ishin g t.be trivial from th e substantive is
nor aJw.ays easy; do not make it even more difficult by attempting to app ly
the wrong standard.
Question A: Reasons justify continuing? In examini ng a list of reasons that
support th e importance of a questio n, th e issue of worth may be viewed
from severa l dimensions: worth to the individual contemp lating the answer
and worch to a professio n, co the academic com muni ty, and ultimately to
society . Question A, "Reasons justify contiuujng?" is the ques tio n that the
resea1·cher must answer in terms of personal in terests and needs. Tbe world
is full of clearly formulated and specific questions that may nor, once
seen in the ir forma l d1·ess, seem wo.rrh the effort of answering . Because
researchers are hum an, perfectly legitimate questions may seem du ll, inreresring veins of inquiry may pe ter our into rriviaUty, and well.-defined issues
may fail ro sui r for no better reason than a clash with pe_rsonal sty le. On the
ot her hand, some questions are supported by the resea1·cher's immediate
need co en hance teac hing in a vita l subject area or co quenc h cu riosity about
a long-held hunch.
The basic rule i.s to be honest before proceeding. If you really don't care
about answering the question, it may be better to st,lrt again while the rnvcscmcnt still is relatively small.
Step 6: Survey relevant literature. A preliminary scanning of the most obvious, pert inent resources, particularly reviews of the titeratt11:e, is a way of
husbanding time. It is far better to abandon a line of thought after severa l
weeks of selective skimming than to work one's way vi.a slow, thorough
digescion of each document to the same conclus ion after several months of
effort.
Conscientious students sometimes ieel vaguely guilty about such quick
surveys. Keeping in mind the real purpose, which is to identify questions rhat
already have sntisfactory answers, is one way of casing stKh discomfort.
Question E: Reason fo1' 1w answe1' remediable? In some cases, the literature conta.ins an empry area because the state of technol.ogy, the available
Developing die Thesis or Dissertation Proposal
stify the question in rerms approtoward filling a gap in the srrucy appeals to practical application
JCh a return) . Inquiry that arises
act ice need not be supported by
: phenomena (even d1ough later
tio n has its own correct measure
he trivial from the sub sranrive is
difficult by attempring to apply
examining a li~t of reasons that
i.ssue of wonh may be viewed
idual contemplating the answer
· community, and ulriinarely to
llling?" is the quesrion that the
interests and nee<ls. The worl<l
questions that may not, once
· effort of answering. Because
Juestions may see m dull, interviality , and well-defined issues
1sh with persona l style. On the
by the researcher's imm.ediate
~aor ro quench curiosity ahout
iediug. If you really don't care
to start again while the invest-
ry scanning of the most obviof the literature, is a wav of
line of thought after se:eral
ne's way via slow, thorough
usion after several months of
1
.1ely guilty ahout such quick
:h is ro identify 4uestions that
>fe;:ising such discomfort.
e? Cnsome cases, the literaof technology,
the available
55
framework, ethica l considerations in completing the study, or
knl.!wle_d~e demands peculiar co rhe question have ma de it impossible or
1ogisuc
.
f orms o £ rnquiry
. . . So Iong as tne
t
t I,e
ble co conduct appropnare
gap
~easona ·
h
d
f" d h
· or
un · '
dge
seems
ro
exist
because
no
one
as
yet
e
me
t
e
quesuon
1
· know ·e reresced in pursumg
· the answer, lt
· 1s
· reasona bl e to procee d. Th ere
in
b orne rn
.
.
.
ec h r reasons for empty or ambiguous areas m the l1te.rarure , however ,
c1·
·ire or e
. . bf
:,nd chey signal caunon e ore p.rocee mg.
t . n I: Meaning of all possible results clear? The tighter th e log ic, the
Ques ro
.1.
l I l cl · · ·
elegant the theoreti cal framewor k , tne more c ose y n e es1gn 1s ta1morde
lore to produ ce clarity a lo ng one dim ension-in short, the better the quality
the risk that the proposer . wi ll be lured
o.f rI1e Proposa l-th e greater
.
. into
c
rtunatc
presumptr.on:
char
the
result
of
the
study
1s
known
before
the
ll 0
an u t •
·
d b L
J.
are in hand. That scudenr researchers sometimes arc contronte
y tne
0
,1ta
.
··
. . ·
- ·I
v news that their treatment produced a reverse effect 1s m 1tsc1t nelt 1cr
ll· 11
sruni ,.,
.
.
.
.
.
of
surpr.·isin<>
,., nor harmful. Be111g
. unahle co make .an mrelltgenr mterpretat1011
.
such a situation, however, 1s unfortunate and m most cases avoidable.
Unanticipated results raise a fundamental question that the investigator
muse confront. Does the finding truly reflect what is resident in the data, or
is ir only an artifact of che analysis? If there is any doubt about rhe appropriateness of rhe analysis, particularly if the procedures were not perfectly
aligned with the research question, the latter possibility must he considered.
Jf reexamination of the analysis provides no accounting for findings that are
sharply incongruent with expectations, another explanation must be sought.
All of chis is made more difficult i.fthe possibility of discrepant findings has
never been contemplated. A strong proposal, constructed iu an orderly, scepby-step sequence> will enhance rhe likelihood that you can manage the unexpected with at least a degree of dignity.
Through serious consideration of alternative outcomes at the time of constructing the proposal, it may be possible to includ e elements in the study
that will eliminate ambiguity in some of the most likely resuJcs. One method
of antici.pacing the unexpected is to fo llow through the conseq uences of
rejecting or failing to .reject each hypothesis of the stud y. If rhe hypothesis
was reject ed, what is the exp lanation? How is the exp lanation just ified by
the rati onale for the stud y ? What finding s would s upport the explanation?
Conversely, if the findings of the study fail co p.rovide a basis for rejection,
what ex planati ons are to be proposed? At the least, some careful pr eliminary
th.ought about alternative ex planation s for eac h possib le result will serve as
a sh ield against d1e panic that produces such awkward post hoc interpr etations as "no significant differences were observed becaus e the instrnmen rs
employed were inadequarc."
56
Wr iting the Proposal
Step 16: Share and discuss with colleagues and advisors. There is a well.
known S)'ndrome displayed by some who attempt research, characterized b
the inclin ation to prolong the period of writing the finaJ report-indefinite!:
Some people simply cannot face what they perceive co be the personal threa~
implied in open ing their wo rk to challenge in the public arena. These individual s ar c terribl y handicapp ed and only rarely can become man 1re, pro.
ductiv e schol a rs. An early sign of this is seen in students who cannot bring
themselve s co solicit ad vice an<l criticism for their proposals .
Someti m es st udent s experience severe criticism because they present their
ideas befo re the y have b een sufficiently developed into a conceptual framework chat represents careful pr epara tion. Many professors avoid speculative
conversations about "half-baked" ideas that have just arrived in a blinding
flash of revelatio n ro the studen t. Few professors, however, refuse a requ est
for advice concerni ng a proposal that has been drafted as the cuJminarion of
several weeks of hard thougbt, research, and developmenr. Even at that,
having one's best effort devastated by poinred criticism can be an agonizing
exp erience. Nev erthele ss, the only alternative is to persist in error or ignor ance,
an<l that is unten a ble in resea rch.
lf yo u are fortunate enoug h to be in a department that contains a vigorous
com munity of inquiring minds, with the constant give and take of inteUectual
disputation, the rough and tumble soon will be regarded as a functional part
of producing good research. Th e novice will solicit, if not al way s enjoy, the
best criticism that can be found.
The n otion that it is vaguely immoral to see k assistance in p rep aring a
proposal is at best a parody of real science and ar wo rst., as in the form of an
institutional rule, it is a serious perversion arising from ignorance. Res earch
may ha ve some game- like qualities, but a system of handicaps is not one of
them. The object of every inqu iry is to get the best possible answer und er the
circum stances, a nd th at presumes obtaining the best advice ava ilable. lt is
hoped that the student will not be held to any lesse r standard.
It should be obvious that students, after digesting and weighing all the criticism received, must still make their own choices . Not all advice is good, and
nor a ll criticism is valid. There is only one way to find out) however, and that
is to share the proposal with colleagues whose judgments one can respect, if
not always accept.
T he process of p rop osal development is enhan ced if you o btain advice at
various steps and do not wait until the end to solicit feedback. We str ongly
recommend working with your advisor and committee in ways that help you
move s teadily forward. Fo r example, at Step 4 co n sulti ng your advis<.>rabout
possible research 9.uestions may help you refin e c:hem and may assi1;t yo u in
finding relevant literature. At Steps 9 through 12, short, focused meetings
pr
IHI
inc
fO(
str
an
Wl
fUL
de1
tiv
Mi
pu
Developing the Thesis or Dissertatiun Proposal
57
?!leagues and advis ors Th ere .
~ h
.
is a W LI
- w o attempt resea rch cl1a. t . e ,
i f
..
'
me enzed b
. o. writing the fina l 1:epon -ind efin1 }'
it they P:rc eive to be tile p ersonal t t~ly,
!allenge in the publi c arena 1·1
~teat
1
. 1ese Jnct·
d
on y rar ely can become m
i. .
anire Pi·
is Is seen in student s who can not 'brio,
118
:ism for their propo sals .
'ere criticism be<.:auscthey present th .
rly developed into a conceptual '
eir
.
M
irarne,
1 .
um. any professor s avo'd
.
h
.
I sp ecu at1ve
.as t at have Just arrived in a bLindin
' professors, howev er refuse
.
g
has hecn drafted as
culm . a ':que ,5t
h
the
marton of
r<.:.' and development. Even at that
pomred criticism <.:anbe an.
. . '
. . .
.
, agonizing
iat1ve ts to persist in error
.
'
or ignorance,
~ ~epartme~t
that contains a vigorous
c~nstant give and take of intelleccual
. w1~!be ~e~ar?ed as a functional part
. will solicit, rf not always enjoy, the
al ro seek assistan<.:e in preparing .
a
at worst, as in the form of
..
f
an
>n ansmg rom ignoran<.:e. Research
ei system of handicaps is not one of
t_the best possihle answer under the
ung the best advi<.:eavailable. fr is
>any lesser standard.
. digesting and weighino all th
.
i:,
e cnt.
:1101<.:es.
Not aJJ advice is good, and
way to find out, however and that
I1osc JU
· dgmenrs one can respe<.:t,
·'
if
~a~
enhanc ed if yo u obtain advice at
1 to solicit feedback ~e str
1commit. tee in ways ·tl1at h Iongy1
.e p you
) ~ consulting your advisor about
in
efme them and may assist Y<>ll
ugh 12> short, focused meetings
arc prepared to discuss specifics may be particularly
beneficial.
where yo~ coiiscructive_<.:riticismare best when re<.:eivedin ~mall doses and
.i.,,jce an .
,.\u "Jtcd throughout
--:sa.l
1nregt,
h
t e proposa
I
process.
~ . 9 . Present to committee . Explain and support.
Presencation of yow·
Stefl
n,ay take ~ lace before a thesis or disserta tion con:nutcee on an occa~roP,1 u,allY sanct10ned by the grad uate school, or at an mforma J gathe ring
11
s10
·
·
. heoradvisor's o ff'tee. In e ·tt her ·msrance,
t I1e purpose serveeiJ an d the irnporin c assumed will depend on borh local traditions and the relationships cbat
ranee
th
.
~L
I .
.
evolved to at pomt amon g u1e c 1au·pei·son, com mitt ee members and
have
'
student.
.
.
If for example, the chairperson has closely monitored the developing
ro;osal and is satisfied that it is ready for final review and approval, the
~arnre of the: meer.ing is shaped accordingly. In addition, if other committee
members have consulted on the proposal at various stages of writing, the
meeting may serve primarily as an occasion for final review and a demonstratio n of p resentation ski lls, rather than evaluation, extensive feedback,
and judgment. When these con ditions do not apply, the meeting assumes far
greater significance, in itself, and the length and nature of the presentation
will be affected .
Whatever the cir<.:urnstances, both a prudent respect for the important
function the committee members nwst perform and a proper desire to
demonstrate the extent to which the efforts of your advisors have been effective make careful preparation and a good presentation absolutely ne<.:essary.
Much of our advice about thar is contained in Chaptet 7. For the present
purpose, we want to underscore the following points.
1. The more you can work with cummittec members hcfore an official meeting,
the more that meeting can focus on impruving (and appreciating) your
proposal-rarher than just on understanding it .
2. As committee members ta lk with you and with eac h other at the meeting, it
is natural that new insights and concerns will surface . So long as those are
accurarely recorded, and so long as tbere is clear prov ision for how the committee will manage subsequent revisions in the proposal, chat process is all co
your advanrage. The objec t is nor simply to ger rhe proposa l (as it stands)
accepted; it is to create the best possible plan for your disserrarion or thesis.
3. Where you have had co make difficult choices, accept comp romises i.n merhod
for pragmatic reasons, or leave some fina l decision(s) for a later poinr in rime,
.it is best ro bring such matters directly to rhe attent ion of your committee.
Don't wait co be questioned. Take the initiative and lay out d1e problematic
aspects for your adv isors as you go through the presenrario n. You need not
58
Writing the Proposal
make the proposed srudy appear to he mired in difficulty. Propose solutions
and give your rationale, but never ignore or gloss over what yot1 know
rcqu.ires more attention-and
the help of your committee.
4. Tf the proposal is approved with the undersranding rhat certain revisions or
additions will be made, chc best procedure is co obtain signatures on documents while at the meering. The signed forms can rhen be held by your chair.
person until he or she has approved the fi11aldrafr.
Step 20: Gather data. Process and interpret as planned in proposal. This is
rhc payoff. A good proposal is more than a guide co action, it is a framework
for intelli~ent interpretation of results and the heart of a sound final report.
The proposal cannot guarantee meaningful results, but it will provide some
assurance that, whatever the result, the student can wind up the project with
reasonable dispatch and at least a minirnurn of intellectual grace. If that
sounds too small a recompense for all the effort, con.sider the alternative of
having to write a report about an inconsequential question, pursued through
inadequate rnethods of inquiry, and resulting in a heap of unanalyzable data.
Originality and Replication:
What Is a Contribution to Knowledge?
Some attenrion already has been given to considerations rhat precede
the proposal, the uitical and difficult steps of identifying and delimiting a
research topic. One other preliminary problem, the question of originality,
has important ramifications for the proposal.
Some advisors regard student-conducted research primarily as an arena for
training, like woodchopping that is expected to produce muscles in the person
who holds the axe, but not much real fuel for the fire. Whatever may be the
logic of such an assumption, students generally do not take the same attitude.
Their expectations are more likely to resemble the classic dictum for scholarly
research, to nrnke an original contribution to the body of knowledge.
An all-too-common problem in selecting topics for research proposals
occurs when either. the student or an advisor gives literal interpretation to
the word "original," defining it as "initial, first, never having existed or
occurred before." This is a serious misinterpretation of the word as it is used
in science. In research, the word "original" clearly includes all studies deliherately employed to test the accuracy of results or the applicability of
conclusions developed in previous studies. What is not included under that
rubric are studies that proceed mindlessly to repeat an existing work either
in ignorance of its existence or without appropriate atrention to it~ defects
or limitations.
Develo ping the T hesis or Dissertati on Proposal
be mired in difficulty. Propose solutions
ignore or gloss over what you know
Ip of your committee.
understanding that certain revisions or
cedure is to obtain signatures on docu:d forms can rhea be held by your chair:he final drafr.
pret as planned in proposal. This is
1 a guide to action, it is a framework
1d the heart of a sound final. report.
ful results, but it will provide some
~udent can wind up the project with
imum of intellectual grace. 1f that
e effort, consider the alternative of
quential question, pur sued through
:ing in a heap of unanalyzable data .
:!dge?
:n co considerations that precede
ps of identifying and delimiting a
,blern, the question of originality,
sal.
research primarily as an arena for
d to pr.oJu~e muscles in the perso n
for the fire. Whatever may be the
ally do not take the same attitude.
Jlc the classic diL'.tum for scholarly
o the body of knowledge.
1g topics for research proposals
sor gives literal interp ret ation to
1I, fir.st, never having exist ed or
)retation of the word as it is used
' clearly include s all studies delif resulrs or the appli cabilit y of
What is not included under. that
o repeat au existing work eirher
Jropr.iate at tenti on to its defects
59
nsequence of the confusiou surrou nding the phrase "original conOn_ec?, is chat misguided students and adviso rs are led to ignore one
rribunonost important areas of research activity and on e of the mo st useful
of rile rnf training for the novice researcher- replication. Tha t replica tion
forrns.:es is regarded simply as rote imitation, lacking sufficient op por.tus~rne: srudents co apply and develop their ow n skills, is an indicatio n. of
n,cy bordly some students misunderstand both the operation of a research
how a
rise and the concept of a body of knowl edge.
enrerp
· m
· researc h h as I,een cogent ly argue d
The essential roe[ o f rep 1·
1cat1on
Gall, Gall, & Borg, 2007). ~hat has not be~n made sufficiently dear , how~vcr is that replicati on can mvol ve challenging problems that deman d ere resolution. fourther , some advisors do not app rcciaw the degrel'. co wh ich
1
for rc1,licativc studies can co nst itu te an ideal learn ing
awn . tl·..,g
u !Jroposals
'
portuni ty for research trainees.
op Jn Jirc ct rl'.plication, students must not only cor rectly identify all the critical variables in the original study but also create equivalent condi tions for the
conduct of their own study. Anyone who thin ks th at the critical variahles will
immediate.ly be appar ent from a reading of the original report has not read
very widely in the research literatu re. Similarly, an individual who tbinks that
truly equivalent conditions can he created simply by "doing it the same w ay"
just has no t cried to perfor m a replicati ve study. Thoro ugh understandi ng of
thl'.probJ.em and, frequently, a great deal of technical ingenuity are demanded
in develop ing an adequate p roposal for direct replicatio n.
As an alternative to direct replication, th l'.stutknt may rep eat an interesting st udy cons iderl'.d to have been dl'.fective in sample, method, analysis, or
interpreta tion. Here the student introduces deliberate ch anges to impro ve the
power of a previous investigation. Ir wo uld be difficu lt to imagine a more
ch.alknging or useful activity for anyonl'. interested in both learning ab <>ut
research and contributing to the accumulation of reliable knowledge.
In writing a pro posal for dthe r kind of replicativl'. stud y, dirl'.ct or revised,
the student should introduc e the original with appropria te citat ion, make the
comments that are needed, and proceed witho ut equivocation or apology to the
proposed study. Replicative research is nor, as unfortunate tradition has it in
some depart ments, slightly impro per or someth ing less than genuine research.
(;iven the limitations of research reports, it oft en is useful to discuss thl'.
source study for the r.eplication with the original aut hor. Most research
workers are happy to provide grl'.ater detail and in soml'. instan ces raw data
for inspection or reanalys is. In a healthy science, replication is the most sincere form of flattery. A proposal appendix co ntaining co rrespondl'.nce with
the author of the original rl'.port , or data no t prov ided in that repo rt , often
can serve to interest and reassur e a hesitant ad visor.
,t'v;
60
Writing the Proposal
Getting Started: Producing the First Draft
The student who has never written a research proposa l commonly sits in
front of a desk and stare s at a blank piece of paper or an empty video m 011•
itor for hours. The mind is brimmin g with knowledge gleaned from the
literature , but how does one actuall y get start ed ? The concept of "a research
proposal" conjur es up ideas of accuracy, precision, meticulous form, and use
of a langllage system that is new and unpracticed by the neophyte researcher .
The demands can sudd enly seem overwhe lming . Th e student should realize
that this feeling of panic is experienced by nearly everyone, not only th ose
who are new to the writing endeavor but those who are skilled as well.
Fanger (1985) exp ressed it beautifully: " I have co me tO regard panic as the
inevitabl e conc.:omitanco f any kin d of serious academic writing" (p. 28). For
anyone temporarily incapacitated by a blank page or empt y monitor, the
followin g suggestions may be helpful.
Make an outline that is compatible witb the format selected co present the
comm unication tasks listed in Chapter 1. An initial approva l of rhe outline
by the advisor may save revision time later . Gather the resource materials,
notes, and references, and organize tbem .in.to gro ups that cor respond to the
outline topics. For instance, notes supporting the rationale for the srndy
would be in one group, and notes suppo rtin g the reliabiliry of an instrum ent
co be used wou ld be in anothe r group .
Once the ourline is made and the mate rials gathered, ta ckle one of the
topics in the oucline {not necessarily the first) and stare writing. If the section
to be written is labeled "Th e Purpose," try imagining that someone has
asked, "W hat is the purpose of this st udy? " You r task is to an swer that question. Start writ ing. Do not worry about gra mmar, syntax, or writing within
the language system. Ju st write. In chis way you can avoid one of the greatest inhibitions to creati vity- self-criticism so severe thar each idea is rejected
before it becom es reality. Rememb er, it is easier to corr ect than to create . If
all the essentia l parts of a topic are displayed in some fashion, they ca n lacer
be rearrang ed, edited, and couc hed within the language system. With experience, the novic e wi ll begin chinking in the language system and forms of
the proposal. Until that time , the essential prob l.em is to begin. Awkward or
elegant, labo rious or swift, there is 110 substin1te for writing the first dr afr .
One way to approac h writing is to use the o ntline feature on your word
procesS£r, which allows you to develop your outlin e an.<lrhen go back and
progressively fill in the detail uJ1der each heading . Learn t o use this featur e.
The effort needed to learn its use wili be repaid many time s over. Word processing programs provide the opportunity for writers to edit , rearr ang e text
tas ily, an<l stor e ma nus cript cop y for futur e revision s. Th ere is a significant
Devffoping the Thesis or Dissertatiun Proposal
·st Draft
1
ul
ical advantage in the ease with which revised draft s ca n be pro-
ps)'chO;t s encourages the au th or to make revisions that m ight oth er wise
esear ch prop osal co mmonl y sit, .
s In
: of polper or an empty video JU
.h k
on.
v1t no wledge gleaned from the
ta rte ~? The concept of "a rcst:arch
>rec
1s1on, mericulo us• form
an d
.
. ' '' u~
K need by th e neoph yte researche
!lming. Th e stude nt shou ld real iz:
·y nearly everyo ne, not onl y those
ut th ose wh o are skilled as wdl,
have com e to regard panic as the
J us acad emic writin g '' (p. 2 8). For
lank page or em pty rnonitot, the
L the
format selected ro prese nt the
An initial approval of the outline
er. Gather th e resour ce mat erials
nto groups that correspond to tlie'
·ning the rati on ale for the st udv
ng the reliability of an insrrumen ,t
erials gath ered, tackl e one of the
st) and sta rt writing. Jf th e section
try imagining chat someon e 1rns
' Your task is ro answer th at ques ammar , synta x, or writi ng wit hin
Y you can avoid one of the great;o severe that eac h idea is rejected
easier to correct th an to create. If
::d in sorne fashion, thq can later
the lan guage system. 'With exp e1e language system and forms of
problem is to begin. Awkward or
;titute for writ ing the first draft.
the outli ne feature on vour word
·llr outl ine and then g~ back and
cadi ng. l.earn ro use th is fcanJre.
:aid ma ny times over. Word pro:or writers to edit, rearr ang e rext
e revisions. Th ere is a significant
duc.:cd, ide under the press of limited time, and has greatly enhanccJ the
bc_~cr a~proposal writers ro revise aod po lish the.ir w or k.
,1b1htY
o
,.;ng Your Thesis or Dissertation Committee
SeIec...
Master's chesis committ ees vary in number from one professor to a co m. e of five or six faculry membe rs. A doctoral dissertation committee
in1cce
. mem bers. ln some .instances, a II comm .ittee
. lly consists o f four to six
cyp1ca
bers are from within the depa rtm ent of the stud ent 's major. 1n ocher
~temnces the committee is multidisciplinary, with faculty representing other
,nsta
,
d partments on campu s.
e Ac most universities, students have some opportunity to request specific
faculty members for their comm ittee. If the stu dent does have some freedom
co exercise cho ice, committee membership should be designed to max imize
che support and assistance available. A student interested in the study of
behavioral treatment of dru g abuse in young upward ly mobil e wom en could
rap the value of different facu lty perspectiv es and skills by blending members
from several depa rtm ents. For this purpo se, individua ls with multipl e interesrs are particula rly useful. For examp le, a faculty member in the psycho logy
depnrtment might be selected for both statistical com petence and interest
in behavior modification, someo ne in the school of soc ial work might bri ng
epidemiological expertise regarding drug usage, and a faculty member in the
schoo l of public hea lth might be a part of the co mmitt ee beca use of expertise in both experimental design and therapeutic co mpli ance techniques.
Because st udents know from the begi1111ing
of the grad uate program that
faculty eventua lly will have to be selected for s uch a comm ittee, it behooves
chem to be chinking abo ut these matters durin g th e selection of elective
courses rlu-oug hout the program. If a choice has to be made betwe en two
professors for an elective course, and one of them is more interested in the
student's prob able area of research, that may ca rr)' th e day in determining
which course to rake. Although it is not essentia l chat students have taken
their commit tee membe rs' cl.asses, it is easier to ask a kn ow n facu lty memb er
to serve on you r committee. That person is likely to ta ke a greater int erest in
your work, and you have a good idea of his or her standards an d metho ds
of scholarshi p.
4
Contentof the Proposal
Important Considerations
T
he topi<.:scovered in this chapter are designed to assist the true beginner. Most experien<.:ed proposal writers may want to go directly to
Chapter 7 or simply skim this chapter for review.
Reviewing the Literature: Finding it First
Areas of inquiry within the disci.plines exist as ongo ing conversations among
those who do rhe work of scho larship. The pub lished Llterature of an area
constitutes the archival record of those conversations : research regorts, research
reviews, theoret~-il speculation , and scholarly discq_mse.o.ulJ kin~ You join
tiielo ng
of
science as you join any other, by first listening to
-·- convers.e_tion
---_,,___
-- ~hat is beiJ!_Ss~i.d, and-on ly-rh~_p
_.fr>m1~1lating
<!_S
omme_p.tdesigned to advance
the dialo~t~ .
-
The metaphor of scholarship as an extended conversation works wetl at
a variety of levels-be<.:ausc at hea rt it is an acqJ.rate...representation , The
process of locating the voices of individu al conversants, for example, is called
retrieval.That involves sea rching through the accumula ted ru:d1ive of li terature to find ouc what has been said (when, by whom, and on the basis of what
evidence). The process of listening car efully to the ongoing discourse about
a topic of inqu iry is called review. That involves studying items previously
ret1·ieved until hoth the history and rhe current state of rhe conversation arc
63
64
Writin~ the: Proposal
understood. It does nor srretd1 the metaphor too far to observe that writin,
the proposal is a step in preparing yourself to have your own voice he heard:
to do research and enter what you learn into the long conversation.
Recrieval, review, proposing and conducting reseai;ch, and even report Wtit.
· ing are rasks that have their own sets of requisite technical skills. Each also has
a place for art and instinct as well as intelligence and accumu lated knowledge.
This chapte r deals with what goes into the proposa l (content) once the top ic
and terms of discourse have been defined . It follows, then, that it must begin
with what yo u have lea rned by listening in on the conversation-a
review of
the literamre . In yoLu· review, you will..§tablish what has b~lLS.aid. to this
sfo r proposing yonr own co n tribmion ._Rerrieva l, however,
point as the bas1'
coine s first in the order oflhfogs. You can't review what you have n 't found.
That brings m; to the technical skill. and fine art of searching the literature.
We will not bur<len you with the specifics of a particular search procedure.'. The demands ma<lc by proposals differ widely, as do the backgrounJ
and skills of each proposal writer. further, the facilities for retrieval vacy
enormously at different institutions anJ, of course, each discipline and suhspecialty has its own peculiar mechanisms for searching the literature. What
we can do here i~ set forth the small number of general rules that, if observed
from the outset, have the power to make any retrieval effort more efiicient.
Knowing what you need to know is the ohvious first step in formulating
a retrieval strategy. Knowing how much you really neeJ to know, however,
rs a vital secon<l step-and
one not always properly apprccrared by the
novice in research. Discussion with your advisor, consultation with colleagues who have written proposals, inspection of proposals previously
accepted by the graduate school, and the preliminary reading already done
during the process of identifying the topic of your proposed study will all
thereby the literature yoLIwill
serve to identify what you need to know-and
seek. Normally that includes research reports an<l reviews related to your
questions or hypotheses. This literature provides information about research
methodology in the area and items dealing with both rheory and application
as they are related to your study.
Deciding how much you need to know is a more complex decision-in
part because you often cannot answer that question until after some retrieval
and review already have been accomplished. This is a matter of defining the
purpose to be served by what you retrieve. Acquiring a hroad overview of
previous work in an area leads to one kind of retrieval strategy. (f your purpose is to know what a small set of senior scholars have reported in the last
two years, rhe strategy will be different, as will be rhe case if your purpose
is to do an exhaustive search in which every scrap of fugitive literature is
doggedly pursued until acquired.
Content of the Proposal
hor too far co observe that wr,·t·
·
rng
to have your own voice be heard_
1to tl1e Jong conversation.
:ting research, and even report Writuisite technical skills. Each also has
~ence and accumulated knowledge,
~ proposal (content) once the topic
It follows, then, that it must be~
. o 1-ll
on the conversation-a review of
:ab.fu;bwb~Lh.~s.b_~e.1uahl
to this
. ~9.titriliu.tio.n_.
__Retricval, however
't review wl1at you haven't found'.
~ art of searching tl1e litcrnnm:.
fies of a particular search prncefer widely, as do the ba<.:kground
r, tl1e facilities for retrieval vary
f course, each discipline and sub-
:or scar<.:hing the litc.:rattrre.\Xi'hac
· of general rules that, if ohservcd
:1y retrieval effort more efficient.
ohvious first step in formulating
u really need to know, however,
ys properly appreciated by the
advisor, consu!tar.ion with coJ•ection of proposals previously
reli.minary reading already done
of your proposed stu<ly ~viii all
1d thereby the literature you will
rts and reviews related to your
ides information about research
·ith both theory and application
, a more complex dcci:-ion-in
1c.:stionuntil after some retrieval
Thi:- is a matter of defining the
'\.cquiring a broad overview of
f retrieval strategy. If your pur1olars have reported in the last
1
ill be the case if your purpose
1
scrap of fugitive literature is
GS
ggest is chat yon call<with your adv isor. abo ut the question of
Whitr ~\~, uaod tha t you stay in touch with him or he.r o n that top ic as you
111
..1,ow L~l~~not ion that every seaJch of the literat ur~ ne~ds to be exha ustive
procee\ b more destructive mythologies that persists m gradua .te St1Jdenr
0
is on~ .;h:n
you have eno ugh sense o f rhe _conversation to argue persuac.:ulnu.e
- cle rarget for your proposed study 1s sound, and char the methods
. ly chat 1
sive . r·e correc t you kn.ow enough for the purpose of the proposal.
'
. ,
f · wry la naer period
O ,nq
of time, you may have plenty of mottvanon co read
Over a o ~
.d ly an.cl deep ly than anyth ing demanded by your pr oposal, but that
11ore wi e
· · d1·scuss10n
· of t I1e .PIace o f
'. d.1fferent marreJ. (For a recent and provocanve
. ce-viewsin p reparat ion for dtssertarions, see Boote & Beile, 2005 .)
i.sa
lireracu.re
· .
.
1·esent it is wise to acq uire some sense of how much you need to
ch
P
For e .
'
•
•
,
.
char yo u can shape th e s1ze of yo ur review task acco rd111
gly- and
kn OW SO
.
.
iJ
h
d
b know what and how much to retrieve. Unt you ave one some
rhe(e
Y
·
·h
d ·
'LI b bl
.·
.,1and review' it is likely
retJtev..
. thar ne1t er you nor your, a . vtsor w1 ea . e
that target with precision , but some careful prelimmary thought will
w•
. .
.
.~erveyou well.
Retrieval Rule I. Do not hegin hy going to the lihrnry or your computer and
starting to search for literature. Talk first co your advisor (or entire thesis or
dissertation commitree) and research colleagues who have some familiarity
with the area of your proposal. Make a list of what they think you should
read. Locate the items, skim them, and record full citations for all that appear
to be appropriate. These form the core of your retrieved literarnre base. Go
over the reference lists in the items that appear ro be most <lircctly relevant to
your needs and make chose citations the priority for retrieval when you
return to the library.
Retrieval Rule 2. When you go to the library, do not begin hy starting to
search for literature. Talk first to r.he reference specialists who can identify
the retrieval systems that arc most likely to be productive for your topic.
Then, we urge you to take advantage of sc,ninats and classes on using spe·
cific retrieval systems that arc offered at many academic libraries. Retrieval
is one pare sweat and two pans knowing where co look.
For every scholar, tbe marvels of computerized rerrieval are available at
your university library, through the university compurer network, or on the
lnterner. The databases that any given system cao access, however, <liffer in
imporrant and suhtle ways. Attending introductory seminars and asking for
expert help can save hours. Swift, accurate, flexible, and powerful beyon<l
anything we could have <lreame<l when the first edition of rhis book
appeared, computer retrieval systems are, nevertheles.~, only as good as rheir
search stmctu.res, and chose are just as full of limitations and idiosyncratic
quirks as any of rhc primed indexes that scholars used in the past.
66
Writing the Proposal
That observation lea<ls to two irems of advice. Firsr, p lan to devote
considerable amount of retrieval time ro learning how eac h sysrem Worfcsa
Second, by all means use compureri zed sysrems, bur do not automar icali ·
assume that manual search is without value. In other words, all bur a ve/
small number of highly technical topics can profit by a visit to whatever is th:
equivalent to the J{eader's Guide to Periodical Literature-on CD-ROM Or
in rhe primed version-in your partic ular subjecr area.
Retrieval Ru.le 3. From the our.set, think of your retrieval effort l\S consistiog
ofa series of stages. It i.sunlikely that you will (or sh01dd) m·arch thrnugh thcn
in perfect sequence; it is more :~matter of moving back and forth among the1
stages in ways thar. will make best use 0£your time.
Stage 1: Jdentific.ztion-Find and record cirntio11sthat seem potentially
rc:levanr. This is work <lone with in<lcxe$,bibliographies, reference lisrs,
and, most often, rl1e computer.
Stage 2: Confirmation -D etermine that the iterns identified can be
obtaine<l for use. This is work done with the library holdings of serials
{clecrror1ic and hardcopy) and hooks, reprint .~ei-viccs,interlibrary loan
pcrsoni,el, microfiche files, and the relcphonc.
Stage 3: Skimming mid Screening-Assess each irem to confirm that it
actually contains content ro be reviewed (to be read and studied with
care). This is work thnr. demands enough masrery of the system language
a11dthe constructs related to your topic to recognize what is and is not of
potential use. M uch of this (though nor all) can be accomplished without
raking the resource item into your physical possession or downloading it
ro your hard drive. This means time at the comp uter or i.n.the sracks and
time at the microfiche reader. The most important rccricval skill here is rht"
abilir.y to resisr the lemptation ro stop the work of ski1nming and screeni11ga11dimmerse yoursc:lfin rhc conversation.
Sta1;e 4: Retrie11al-J\cquire the literature . T his is wo.ck do ne by checking out hooks, downloading or copying articles from journals, ordering
microfiche an<lreprints, and iniriating requests for incedibrary loans . Not
everything musr he (or shollld be) retrieved. There is a strong argument
for nor having every article immediately ar hand when you ar,e drafring
rhe review of litcrarnre, :rn<lone way ro ensure rhat is t0 ta ke notes from
reforeucr.s that theri stay in the stacks.or in a folder on your computer.
St.age S: Review-Read and srudy the literature rh:cttrecords thr. co11versatioo about your topic. Subsequent secrions of this chapter will deal
with how ro use what you learn to build your proposal.
Retrieval Rule 4. From the first moment of your search, keeQ_~g of all the
_t.0._
learn about . These will. ~e
words used to ~ hfille.
keywords userj_gY.i11dexi11g--5¥srem.s.
(of.ten r.h,ey-~ specificaUy desigruu:ed
Building a keyword
as keywords by the authors ) for acce~i!!g their hoJdiiJ~_,_
Us
_t is-Ii~ acquiring a Ser of master key~ co a large~Y.ilding. They can open
doors iJL.a. vacie~ f locati2.!!§..~nd withol!I chem you can-wande r for
hours wirh9ut _gaining_e~r. to anyw here you want to be. Although most
Content of the Proposal
67
s of a<lvit·c. hrsr, plan tu devote a
to learning how each system works
:I systems, hut <lo not amomaticall~
,ralue. I1~other words, all but a ver~
:an profit hy a visit to wharever. is the
·iodiw/ Literature-on CD-ROM or
u subject area.
d:icabasesnow provide the oprion of searching titles, authocs, abstcacrs, au d
k _ words-or a ll of them-knowing what phrases produce rhe greatest yield
eyll help now and in the future. In almost every case, the novice will be aston-
of yom retrieval effort as consisting
will (or should) march rhrough them
Jf moving back and forth among the
your time.
est prioriry in your search _P,lan,as shou ld annotared bibliogra_ehies and the
reference list ar the back of every article and book you retrieve . For the s~me
reason, your fast stop inrhe library shou ld be~ roQuest Dissertations and
Thes.~ ~hat could be a better search strategy than reading the reviews of
literature crafte<l by studcnrs who have worked on similar problems?
Dissertatioi:i_s~re the _Y
.§l/.Q1uJ~ i:es of research retrieval. From the sta11:, let
y·our fingers do a lot of the walking.
:cord citations that seem potentially
foxes, bibliographies, reference lists,
that the items i<lemified can be
· with the library holdings of serials
.s, reprint services, inre1·library loan
elephone.
Asses.5 each item to confirm that it
cwed (to be read and studied with
,ugh mastery of che system language
1ic to recognize what is and is not of
1ocall) can be accompl.ished without
1ysical possession or downloading it
at che computer. or in the stacks and
sr important retrieval skill here is the
) the work of skiinming and .~crecn:-rsation.
arure. This is work done by checking articles from journals, ordering
\ requests for interlibrary loans. Nor
tr.ieved. There is a strong argument
rely at hand when you arc drafting
to ensure that is to take notes from
or ir1a folder on your computer.
: literature that records the conversections of rhis chapter will deal
.1iltlyour proposal.
.>fyour search, ~ecp_a__)
og of all the
~am aho_ut. These will -b~c~ ~;~he
::n they a_re specifi~ally
Jh.~ir_hol<lings,_Buil~ °i a· k~yword
to a large building. They can open
ithom chem you can - wander for
: you .yam to be. Although most
i~Jgn'ii~
w.1
ished at the variety of words and phrases emp loyed to categorize items that
appear to be identical.
Retrieval Rule 5. Always take maximum advantage of other people's work.
for char reason, reseaxch ceviews in your area al~fil'.lt sl~ould h~v_e _E.
Q_ehigh-
Retrieval Rule 6. Record a complete. citation for every item you identify.
Whether with index cards or a computer program rhar alphabetizes and sorts
by keywords, keep a complete running record of what you find-whether
immediarcly reviewed or not. No frustration can match that of having to
backrrack to rhc library for a missing volume or page number. We strongly
recommend you use a computerized bibliographic nOl'e raking and retrieval
program {e.g., EndNote, ProCicc, or Reference Manager, among others).
Employing such a program will aid in retrieving citations and developing
your own searchable database. That, in turn, will greatly reduce the time later
required to prepare the p.roposal's reference list. Many universities now have
site licenses that provide for downloading such bibliographic programs without charge or ar a very small cost.
Retrie11alRufe 7. Whatever noces you may take during the .~tagesof Skimming
and Screening or Review, never write anything down in which there could be
the slighrc.srconfusion at a lacer date as to whether the words are your ownor those of another auchor. [f you take down a qnocation, take it verbatim and
attach the proper page citation. If you write anything other than a direct quotation, make absolutely sure it is a paraphrase in your own words. There is no
place for anyrbing in between those two species of notes.
}{etrieval Rule 8. l.lccaucious about computerized systems thar seem roo good
co be true. There are reference retrieval systems, for example, wirh which you
can simply highlight a phrase and the computer then auromatically imports
rhe phrase co a designated poim in your own document, complete with citations in the text and a reference at the back. Wrirers using such a system don't
have to read rhe full arriclc or do the imelleccual work required to truly
understand how it fies into the wider conversarion among investigators in the
a.rea. What seems quick and efficiem may serve to underCllt the ability to
make informed decisions ahout source material and, ultimately, to wrire a
sound proposal.
G8
Writing the Proposal
If you follow thes e eight rules, if you build a reaso nabl e sense of h
much you really need to know , and if you persist, you may have one of
most wonderful epiphani es a scholar can exper ience. As you look dowt t
page of references, you w ill recognize all the names, and the voices of
_a
conversation will fill your ears. 1\t tbat moment, you ar e current and you
ready t o take part. Given the pace of work in many areas of science, thtc
mome nr.will he brief , but savor it! That is the sweet fruit of retrieval.
at
oh
th
:'t
Reviewing th e Literature: Writing the Right Stuff
By much deserved reputation, the reviews of literature in stud ent research
proposals are generaUy regarded as consisting of clumsy and turgid proSe
written as proforma responses to a purely ceremonial obligation in the plan'.
ning format. Even when carefully crafted with regard to basic mechanics
they make duU reading, and when not so prepared they are exc ruciating tor-'
ture for most readers. Much of th is problem arises from a misunderstanding
of rhe task served by reviewing the literature, and none of it need be true.
To begin, the commo n designation used in proposals, "review of rbe
literatur e,'' is a misleading if not completely inappropriate title. A research
proposal is nor the place to review the body of literature that bears on a
problematic area, or even the place co examine a ll the research that relates
to the spec ific question raised in the proposal. A variety of methods for
"reviewing th e literature" do exist, such as best evidence synthesis, critical
reviews, and even mera-analysis, but they are rarely appropr iate for proposals. Analyses of that kind may be useful documents publishable in their own
right. Ind ee~ some journals such as th e Review of Educational Research are
exclusive ly devoted ro such critical retros pectives on scholarship . The task to
be performed in the proposal, however, is different. It is not inferior to the
true review, it simply is different.
ln writing a research r~osal, the author is obliga ted to place the ques-\....J tio;-oi: hypQthesis in the conrext of previous ~Ork in su;h - a ;ay as to
y·, " Jixp laii: and justify the de~iilims t~ade. That alo~ is reqJ~
N;thing_ _1:n ore
is appropriate, an.d tlOthing m~h2..u.ld.k.a.n.e.rnp recf
Although the author may wish ro persuade the reader on many different
kinds of poincs, ranging from rhe significance of the question to the app ropriateness of a particular form of data analys is, ~.nd.p.c_opo.sals devote most
how ~ nd WE}'. the research
~eJite.cacure
review co ~ai nin g (a) <:_x:~E.
question Qf hy.e_othesis was formulated in the proposed form and (b} exactly
_, \ whY.rh.e proposed- r-;;searcb stra~y was selected. What is required t0 "iccomplish thete mks is a ste p-by:Step exp lanation of decisions, punctuated by
-
~(
~~
-- -------
------
Content of the Proposal
·ou build a reaso na ble sense of 1
.
l o\\,
you persist, you may have one of
:an experience . As you look do t~e
di the names, and the voices of ;n _a
·nomenr, you are current and yo eit
.
Ua~
Nork 1n many areas of science th
· t1e
l sweet fnut- of retrieval. ' at
· 1.s
:ing the Right Stuff
,vs
,L
. .of l.iterature in-srudenr researc,
1
,1st10g of clumsy and turgid pi·os
V ce1:emonialobligation in the pla::
d with regar d ro basi<::mechanic
s,
prepare d they are excruciating tormi ar ises from a misunders tand ing
ure, and r~one of ir need be true.
sed in proposals, "review of the
:!ly inappropriate title. A research
ody of literatur e that bears on a
irnine all the research that relates
posal. A variety of merhods for
s best evidence synthes is, critical
re rarely appropriate for propos:uments pub lisha ble i11their ewn
•iew of Educationa l Research are
:rives on scholarship. The task to
liiferent. It is not inferior to the
~ i~ o~ligate9 ~ lace the ques~u~~ ·kin sucl~r_as
to
:Lloneis requir ed. Nothing_mo.re
~
-
le the reader on many diffon:nt
e of the question to the appros, ~ound.pro_
ls devore most
:dy how anJ wh~
a~c
£:9posed form an d (b) exactly
~d. What is required ro a~C()fll.1 of decisions, punctuated bv
-----
.
69
tudles chat suppon che....9ng9ing_filgJJ,Q1~!).J
._In this, the w.rirer
·I<:often some critiqu e of J?tevious work, and some times
re.:
.ous wor '
- - 7
r: .
.
.
I
,5 preVI ;,_
-f thebroad
f<
now1edge
as lt eX IStS
lll t le a rea CO
11si:
- .i:;natt ern_OI
- • d
51non o
01ne expo
. der's acceptance of the logic represented m the pr,opose
s , I for the cea
. --------:ippeil
srudY. articular arguments must be susta ined in the review of the
Whnrevei p is no p lace for the ''Smith says this .. . " and "Jones says
re rhere
I
·
I
lirer.iiru '
raph-by-parag raph recita l that ma <es nov ice proposa s
" parag
.. . .
. tJ ·
d ,
rhat · · ·
f dulling the senses. [ 111
s 1s cl1eplace to answet 1e rea er s
k
d h
rnents or
iosrru.
d~->
te nuescinns: What is .it the author wanes to now 1 an w Y
1
~
=---.
d
.
I 1·
1110. ,mme _
~£.hi;plan been
_g~yi~~d_t~fmd the answer? In a~
rev1e,:, t ~e 1terah::1Sc_ . d to serve the reade r's Quecy by sup_poi;t
1n_g,__
_~ ltcarm~ an d
re is ma e
. . .
d .
. .
tLI • .
rJ1e logic now 1mphc1t m rhe propose mvest1gat1on.
- . I ,'
1·ttumrnatu1g
I
d .
JI
then that where there is little rele..Ya.JluJJ;
~rat ure, or w 1ere ec1It fo ows,
' .
.
. .
.
.
b.f
_
re clear-cut and with<ll1ts_~1bsrnn_tia_l_
1_s
_:,m~~,.{h.G.J:~Y~~~
sho uld be ne .
_..,.---:-----.
f
.
1·
b
b
stons a
ses
the
cxaminat1on
o.
supportmg
1terature
may
est . e
In some ca ,
.
.
proposal. To write a revtew
Ied o r woven into another. secnon. of the
,1ppen<
.
i
.
k .
.
.
for
the
sake
of
having
a
review
m
the
c ocument 1s to ma e 1t a
ol hrcraturc
arudy and not a proposal.
_
P Remember, the writer's task is to e~~l~loyth~ .rts~:ir!.=h_l1cer,.1ture
~rt[ully
.th~cl10ices mad~ f9r this study, n~t- to e~ucatc the
ro support anJ e?(~l_ain_
. der c;mcerning
theI pur1ea
_,_the_ State
_ _ of science
-- in the problem area. t",;e1.ther
. lS h'
I
,e
section
to
displav
the
cncrgv
and
thoroughness
wit
l
w
1c1 t 1e
1
pose o f Cl
.
·
·
. .
·
author has pursued a comprehensive understanding of the literature.:. If the
author can explain and support the q uestion,, ~e§ign.,__g_
nd procedures wi.tlu
minimum demand on the reader' s time and intellect.,,thc1!__L
hat readcr will
<
m-ore than s_uffi~iensly
_imnre:,;i,_cd
with the applicant's capa!_,ilities~ nd serio~1s
purpose.
None.:of this is intended to undervalue the.:task that every researcher must
face, chat of locating and thoroughly assimilating what is already known.
To do this, the student must experience what Fanger ( 1985) describe_d as
"immersion in the subject" by reading extcnsively in the areas that are either
directly or inclircctly related to the topic of study. This may lead at firsr to a
sense of frusrrarion and confusion, but perseverance usually leads out of the
wilderness to the point at which what is known about the topic can be seen
in the light of what is not known. The goals of the proposed swdy can be
projected against that backdrop.
The proposal is the place to display the refined end products of that long
and difficult proccss. It is nQtunc.Qmmon, for example, for the study's best
support t~ 1ergc fro~ ;·s~phisricar~~i-~~1Z1~;st,~~icffng of gap~ in~ ne l5oay of
~S!!.~ 2 l!_!;
1_9~quate
knowledge, lirnrtatw ns in_pf eVi~ s__
toriniJfat1ons of ~-h~_9t~
5
.,(crerice_~
~f
70
Writing the Propo8a!
methods of data c~l~tion or inaQW'QJ?.!iat~
incer_eretatiQ.n
~ results. th
reviewof - fhe lireracure section then E_e~o~esa vehicle for illustrating Wh
e
ne better. What readers need, however, is nor a f ii
....
an~
it atl c~1 becfo"
tour retracing each step the author took in arriving at the better mouserrall
but a concise summary of the main argumentS_f>r.21?erlrjuxtaposed to
new and better plan for action.
Most scudents will agonize over the many studies discovered that, alrhough
fascinaring and perhaps even inspiring during the immersion process, in the
final stages of writing mrn out to fail the test of critical relevance and there.
fore merit exclusion from the proposal. It is cempcingto see discarded studies
unused note cards, and bibliographicentries on your compurer as wasted tirne'
but that misses the long view of learning. The knowledge gained through syn:
thesis and evaluation of research results builds a knowledge base for rbe furure.
The process of immersion in the literature provides nor only the information
that will support the proposal bur also the intellectual framework for future
expertise. What may apperu· in the crush of deadlines and overload stress to
have been pursuit down blind alleys, ultimately may provide insights that will
support new lines of choughrand future proposals.
Writing the section o
e literature often is no more complex than
first descn ing the major concepts that ledyou 00 yoi1r cesear<j,_quesrionor
hypothesis and then describing the supporting research findings3.!ready in
the literat~~~ybe
as simple as bJa:?_othesizing
thar..A..is_gi:.eateuhanC.
Why do you hypothesize that A is g_:EaCE_r
t~n C? Beca~den~ggests
and B is greater than
therefore, it is rea~onable
char A is greater than s-;to hypothesize that A mus-t be greater thl!!! k_.
In the review of tl"u:
rclated lit<mi~ou
would present those conceptual
relationships in an organized fashion and then document each with previously reported studies. For example, the first section would include the most
important studies indicating that A is greater than B, and the second section
would present similar evidence supporting the proposition rbar B is greater
than C. The literature section would then conclude with the argument that
given such information, iris reasonable co hypothesize that A is greater than
C. In addition, either incenvovcn throughout or in separate sections, material from rhe literature would be presented in support of decisions about
design and measurement in the proposed study.
Look ar the example in Table 4.1, which also is represented diagrammatically in Figure 4.1. In this table, the general research question is posed,
followed by the specific hypothesis through which rhe question will be
answered. They are shown here merely to establish the frame of reference for
the outline. In this example of tne development of the related literature, rhree
major concepts are necessary to support the legitimacy of this hypothesis.
l;
5;
Content of the Proposal
l'
Rl2!QI2!!~retation
of results ""I
b
~. Whe
:.. -~es
a vehicle for iUustrati~g
~bat readers need h.owever is not • >'
.
.
. .
,
'
a till!
.ook m arnvmg at the better mouset
tap
1
~ rgumen!Le::9_p~
_t:_fr_iuxcaposed to th;
c many studies disc overed that, althou h
ig during the inunc .rsi.on process, in t~
I the test of critiGtl relevance and the .e
re...
Ir .
· -~ts tempting to see discarded stud ies
ntn,es on you r;computer as wasted time:
ng. ~he knowl edge gained thro .ugh syn ...
; builds a knowl edge base for the futu
re.
ture provides not onl.y the information
) tl1e intdlectual framework for future
,s_hof deadlines and overload stress to
timately may provide insights tliat will
Preparing the Relat ed Literarure Section
~
4
~1--• 1Jfc' •.1 : 1s physi.ca l fitness re Iate d to cog nm
· ·on m
. OId. er ~.d uIrs.> ;v '1or~
QUES1•f'tON
•nJfy. can an aerobic exercise program increase cognitive p.rocessmg speed
"Cl lt:i• · '
sp~
I ?
.1110 Ider
adu ts
·
1
_
·t<;· Mainten:mce of physical fitness th.rough a physical training
PO HlES · .
. time
. mo
. Ider
f/Yprogr,
-1·.,nificantlv. decrease (ma ke f aster ) reactwn
am w1·11s,,
individuals.
.
Outline: Develop the Concepts
d
, le the Rationale for the Stu y
r..:· r Scage
< ,rs
Tfol(
{'rOVI(
.
.
Reaction time is related to physical f1tne.~slevel.
[. ,v.
''1a1·11renancc of cognitive function is dependent on maintenance of
II.
aerobic capacity of rhe brain.
The aerohic.;capacity of brain tissue is affected by physical activity-related
regional cerebrovasrnlar changes.
m.
: proposals.
!ttJJe often is no more complex than
to your research_ question or
Ied ~ou
--------:e_ort111g
research findings aj1·ea dy in
Second Stage Outline: Development
of Subtopics for Each Major Concept
r.
Reaction rime is related to physical fitness level.
A. Comparisons of the reaction time of physically active and inactive
subjects.
B. Training effect.~on reaction rime.
C. Reaction. time of those in poor physical condition (cardiovascular
disease, hypertension).
II.
Maintenance of cognitive function is dependent on maintenance of
aerobic capacity of the brain.
~t...A..
is...~~ceu.l:ian c.
~½_an C? Because e_Yidence....fillggests
:.!:_
_than C; therefore, it is reasonable
1an C.
you would present those conceptu;i]
1_~ d1en document
each wirh previfirst section would include the most
:atcr than R, and rhc second section
1
g d1e proposition that B is greater
n conclude with the argument that
J l1ypothesize that A is greater than
·out or in separate sections, macer·ed in support of decisions about
study.
·h also is represented diagrammatneraJ research question is posed,
ugh which the qut:stion will be
stablish the frame of reference for
lent of the related literature, three
he legitimacy of this hypothesis.
A. Relationship of cognitive function and hrain aerobic capacity in aging
individuals.
R. Relationship of the neurological measure of brain function,
electroencephalography (EEC;),to cerebral blood flow and cerebral
oxygen uptake in older subjects.
IH. The aerobic capacity of br..1intissue is affccte<l by physical activity-related
regional ccrebrovascular changes.
A. Tncreased metabolism in specific regions leads to ccrehrnl blood flow
shifts to those regions.
B. Regional blood flow shifrs in motor areas of the hrain are related co
physical movement.
C. Exercise is related ro changes in brain capillarization.
(Conti11ued)
71
72
Writing the Proposal
Table 4.1 (Continued)
Third .Stage Oulline: Add the Mo.st Important References That Support Each
Suhtopic
I.
Reacr.ion tirne is related to physical fitness level.
i\. Physically active imlividll:ils have faster reacr.ion rimes th~n do
sedentary individuals (Clark & Addison, 2003; Cohen, 1993, J995.
Jones, 1998, 1999;.Jones &Johnson, 1991; Lloyd, 1994).
'
B. Reaaion time is foster after a physical training p.rogram (Black,
l 992, 1997; Douglierry, 1987; 1'vlorg:
rn & Ramirez , 2006; Ram irr.z
2003; Richard~, I 99S, 1997; Richards & Cohen, 1989; Roe &
'
Williams, 1995; \Xialte,·s, 199!).
C. Car<liovascular-discased pmienrs have slower reaction time than
normal individuals (Brown, 199l; Brown, Mathews, & Srnith, 1998,
!Vfillcr,l9 91, 1992; Miller & Roe, 2005; Smith, llrown, & Rodg~rs,'
1999; Smith & Rodgers, 1998).
II. Maimc11ancc of cognitive function is dependent on niainrenancc: of
aerobic capacity of the braiu.
A. 80th cognirivc function and aerobic capaciry decrease with age
(Gray, 1988; Petty, 2006).
13. EEG, ccrebr,d bloocl flow, and cerebral oxidative capacity dec,·ease
with age anci arc relate<l (Doc & Smirh, 1999; Doe, Smith, & Snyder,
1997; Goldberg, l 998; Smith & Doc, :r99]; Waters, 1989, 1993;
Waters & Crosby, 1992).
HI. The aerobic capacity of br;iin tissue is affected by physical a..:tiviry-rdate<l
regional cercbrovascular changes.
A. Increased rc:gio,1almet:~bolism leads ro blood flow shifts (Green &
Neil, 1966; Lewis, 1979; Thomas, 2001).
13. Regional hlood flow shifts to motor areas of the brain arr. related to
physical movemcms being comrolled (Caplan, Myerson, & Morr.is,
1991; Goldsmith, 1993, 1994;.Johnson, Goldsmith, & Rodriguez,
1990).
C. Exercise is related to changes in brni11capillarization (i\ileyers &
Templeron, l99 l ; Parrick, 199~; Patrick & Stone, l995; Rohi.nson &
Sper1eer, l 997i.
Jn Table 4.l, the gucsrion suggesrs that rhe way physical fitness and cogni·
tive function are related is through a change in brain acrohic ca pa city as a
result of I raining. Jf this is a reasonable quescion to ask, one would have ro
show thac there have been some prior studies in which physical fitness lcvef
has been related to some measure of cognitive function (Concept I). Second,
some evidence that cognitive function might be al.tcred by aerobic functional
Content of the Proposal
>rranc Referem:es That Support Ea~
ica) fitnc:sslevel.
; have fa.qer rcacrion times rhan Jo
: &hAddison, 2003; Cohen , J993
. , l t)9
. .),
.1o nson, l 991; Lloyd, .t994 }.
,
,ri.physical training progr.arnil3lack,
,7; Morgan & Ramirez 2006· R .
.
· '
, a111Jrei
: R1chanls & Cohen, 1989· Roe &
,
) l}.
'
~nts have slower re·icrr
·on r·.1111et ha11
'
991; Bro~n, Marhcws, & Srnirh, l<J9&;
Roe, 200); Smirh, Ilrown & Rodg ..
n
, .
~~
is <lependenr on maintenance of
:robic capacity decrease with a~e
cerebra! oxidative , ..,,,. 1r 1·...,, de
.
'""r." '- ,.,
crea:sc
'.k Smith, 1999; Doe, Smirh, & Snyder
-<.Doc, 1991; Watc1·s,1989, 1993;
'
.i:saffected by physical activity-related
:ad1;to blood flow shifts (Green &
1s, 2001 ).
>tor areas of the brain arc related to
>lied (CapJ:rn, Myerson, & Morris
>hnson, Goldsmith , & n''-0 u.,r·g
·'
I llez,
•rain Cflpillarization (Meyer.s&
Patrick & Stone, 1995; Robinson &
~ way physical fitness and cooni~e _in hrain aerobic capacity
a
~st10n to ask, om: would have to
es in which physical fitness level
ve function (Concept I). Second
he altered hy aerobic functional
:s
73
f I e bra in sho uld be shown (Concept Il). Fina lly, some evi dence
.3p,1cirY~ r ~,at p hysical moveme nt can a lter bJood flow shifts in the brain,
'
Id o:<ISt r
.
.
,i,ou 1;;, blood flow shifts a re rela ted to aero bic capac ity (Concept III).
,and cliM U the major concep ts ar:e sup porte d by two or thr ee subt opics,
(;ener~ :i cad ro the form a lization of the ma in concept . For exan, ple, the
:ill of whi~at react io n time a nd physica l fitness level are re late d (J) can be
,on cepr ~ i·n thi-ee differe nt ways, by show ing (a) tha t react ion time is faster
0
s11
PP r~e ally fie perso ns th an in se de nrary pe rso ns, (b) tlrnt physica l tr a ining
in physic reacdo n time , an d (c) that reaction time in chose on the lowes t en d
1
l! ~hiuic~:ysicaJ ficness con tin uum is che slowes t of all. Eac h of these s ubtop ics
h
of r e P·rted by the fin dings from severa l stud ies, as shown i-n the t hird stage
is suppo
I
• 1. e section of the tab e.
1 ·
ourWriting
in
· n ·1s mu cll easier
. I'f a n o ut 1me
· .is deve Ithe relate c, 1itera
cure secuo
ope d ·111 ·srages of increasing detail, as shown in Table 4.1, prior to the actual
section of the •proposal can be.
.
.
wn-t·11I",.,. Once the outline is developed, this
wriccen in a straightforward manner, with little backtrackmg necessary. If
mericuJous care is taken in selecting each reference, an enormous amount of
rit11t:
will be saved in the long run.
Another easy way to conceptualize the organization of the related literacure is to diagram it, as shown in Figure 4.1. In this figure, the question (Q)
i.~shown in the first box, and then the major components of the rationale arc
shown as they relate to one another. Component I refers to the behavioral
observations that have hecn reported in the research literature. Component
IT refers to the literature in which the effects of aging on cerebral aerobic
capacity and cog nitive funct ion are desc rib ed . Beca use these relat ion ship s
have negative o utcomes, the)' are sh ow n w ith a nega tive sign . Co mp onent ill
refers to all litera tur e tha t provi des support for a re lat ionshi p betw een physical activity and cerebral aero bic act ivity. In this case, these a re aJJpos itive
relationships, which a re sh own with pins signs. Fina lly, the last b0x dep icts
how all d1ese relationsh ips lead to the hypot hesis (H) of the stu dy .
Both the o utline a nd diagra m format can be very helpfu l in co ncep tu a lizing the relate d litera rure . Th e entire process ca n be summ ar ize d in th e 15
steps in Table 4.2. Table 4.3 contains guidelines for evaluating rhe related
literature section.
Spadework: The Proper Use of Pilot Studies
The pilot study is a n espe cially usefu l for m of ant icipa tion, a nd one roo ofte n
neglected in stud ent proposa ls. When it co mes to co nvincing the schola rly
skep tic (some times your own adv i·sor), no argum ent ca n be so effect ive as to
w rite: " I trie d it a nd here is how it wo.cked. "
Q:
---J
Is physical fitness relaled to cognition?
Can an aerobic exercise program increase cognitive speed in older adults?
...:,.
l
I
I
IA
1B
The RTs of physically active
older men are faster than
those of sedentary older men
Assumption
ofhigh
I
IC
RT is faster after a physical
trainir.g program in young
men
Persons with CVD
have slower RTs
I
fitness
Assumption of increased litness
Assumption of low fitness
J
Ill
Physical activity is related to
cerebral aerobic activity of the brain
IIIA Increased regional metabolism
_.. CB Fr shifts
111BCBFr changes occur in brain
motor areas related to movement
(+)
I~
(-)
IIA
(-)
118 Decreases in EEG, CBFr, and
II
~
(+)
Effective cognition is
dependent on maintenance
of aerobic capacity
Cerebral aerobic capacity and
cognitive function both
decrease with aging
cerebral oxidative capacity
with aging are related
IIIC Exercise is related to increased
brain capillary density
H:
Figur~ 4.1
Example
Maintenance of fitness through a physical training program will
significantly decrease (make faster) reaction time in older individuals
of a Diagrammatic
Overview
of rhe Rdared
Lirr,mrure.
Content of the Proposal
75
Seeps in Writing the Related Literature Section
T blC4.2
-
:t
·ne the major concepts (generally 110 more than two or three) rhat arc
1. oec~rnu co the proposed researt:h question. That is, what are the concepts that
. to he appropnate
.
pc,rc1nent
b ·tie for vour question
or hypot I1eses tenable?
muse e tt
.
•
ccptS either in descending order of importance or in terms of logical
:!,. List con ri'on Th'<ltis, does one concept have to he underscood before another
prcsenca ·
e introduced?
con b
an oucline with these major concepts as the major headings {such as the
J p,cpar C
·
.
· Table 4.1, Concepts l, H, and lll).
one
in
Underea<:h111ajorheading, list t.hc articles that arc mo~t directly related (authors
4
co
an£1dares only).
If rhe articles under a major heading cluster themselves and suggest a
5· subhead ing, then arrange rhe clttsters under the major topics in logical order .
For example, you might note that of the nine studies pertaining to the notion of
a relationship between reaction time and physical fitness, in five of these .reaction
cimes of animals were reported, whereas in the ocher four sruclies, the reaction
times were from humans. The inrerpretation of these studies, when clustered in
rerms of type of subject, might be different and have substantia l bearing on the
potential outcome of the proposed research.
-
6. Withour referrin·g to the derails in the articles, summa rize in one paragraph the
combined findings of cad1 clusrer of studies . For exampl e, in Concepr I.A of
Table 4.1, the summary might be char reaction rime of physically active men and
women is faster than that of sedentary individuals, as long as the subjects are
over 60 years ol<l.The summary of Concept I.B might be char aerobic train.ing
improves reaction time, bur strength training in older individuals does not
improve reaction time. At some point, you will have to discuss the interpretation
co he made from different results of physical fitness on reaction time, depending
on the way physical fitnes:; as an independent variahle is measured.
I~
:i:::
7. Write an introductory paragraph explaining what rhc rwo or cluee major areas
are and in what order they will he disrnssed. Explain why the order used was
selected, if that is important. Explain why sorne literature may be omitred if it
might seem logical to the reader that it would he included.
8. Write a statement at the end of each sectio11sun1marizing the findings within
each duster of studies. Show how this summary of findings relates to those in
the cluster of studies described in the following paragraphs.
-
9. Wrire a paragraph at the end of each major topic (I, H, and 1II in Table 4.1 ),
with a subheading if appropriate, that summarizes the major points, supporrs
the cohesiveness of the subtopics, and ~stahlishes the relevance of these concepts
to the proposed research question.
(Continued)
76
Writing l11c Proposal
Table 4.2 (Conrinued)
10. Write a paragrnph or shorr sl'Ction (wirh the appropriare heading) ar the
condti.~ion rhar drnws rogether all rhe major summari;,:iug p:~ragraphs.
11. Read rhc paragraphs and subject rhem ro Sreps l .7 in rhe "Guidelines for.
Evaluaring rhe Related Literature Section" ('fable 4.3j .
12, Afrcr al\ these conceprs and suhropics have heen carefully inrroduccd,
dcscrihcd, and summMizcd, rcrnm ro the beginning and inserr rhe
doeulllenration ior each of chc concepts in the prop er location. Thar is,
document the sraternenrs made in each of che parag raphs by describing
the :;tudics leading to them or verifying them.
13. F.ad1 lime a reference is inserrcd, phce rlie cornplerc cirarion in a special file for
evcntual compilation of a reference list.
1.4. 1\fter a week has p;isscd, reread the reJared liternrurc scctiori a11duse the
co111plcte"Cuidclirrc5 for Evaluating the Related Literature Section" that are
prnvidcd ir1Table 4.:~.?vlake whatever rcvisi.onsseem nceessary and w:~ir.one
more week.
l S. Read the entire related li.lcrature seetio11for l'.oheren.::e,concimrity, and
s111oothnes.~
of transition from one concept to anorher. Check carefully for
acct1racy of all citmions and, again, edir for rnechanics.
Jr.is difficult w imagine auy proposal that could not be improved by rlw
reporting of actual preliminary work. Whether it is t0 demonstrate the reiiabilit-y of scores produced by the proposed instrumentation, the practicality
of procedures, rl1e availabiJir.y of volunteers, the variability of observed
events as a basis for power tests, subjects' capabilities, or the investigator's
skills, the mo<lest pilot study is the best possible basis for making wise dccisions in designi.ng research.
-:,..c:-
\
\
The pilot study, for examplc, is an excellent means by whi.ch to determine
rhe sample size necessary to discover siguificam differences among cxpcri1
meut;1J treatmems. Sample size estimation or "power analvsis" recentlv ha.~
become commonplacc in quamitativc research .. l\l thoui:J;; ai~ ays has ,bcen
an important component of good research, it has become more frequently
use<l because of the avail.abiliry of easy-to-use books and c.:ompuc-crprograms. A particularly useful introduction to power nnal ysis, rep lete with
ii
..... --••-' - ·=-:-----:--::
.:-:::::
j examples. _;~nd ta hies, is How Man y Subj~ :t~~..(l~-:i~~~~E_{}<.Thiemann, t9m
ruaddirion, ,i- numher of computer programs arc availahl e-so m,e of whid1
only require rhe user to answer a few quest·ions before calculating sample
size. Because software is being updated quickly, we suggesr inquiring at your
Content of the Proposal
ith ~he appropri::irc heading) at L
m·~J
·
- . 1nr-io-r· h tllc
. or surnrnarJ7.Jng
• (. c. ~ ,tp s~
to Steps 1-7 in tl1~"G ·c11·
~
All e mes f
· or
on" (1 c1b/c4 ..'3).
lave heen c:ar.cfo!lyintroduced
he beginning and insert the
'
; in rbc prop()r loeat ion Tl . ·
f
• . l,lt
• i,k 4.2), '
·h of these questions are located.
r .1
•. to cac
5
•
•
•
•
11\!-i\\.'CJ·
• r·igraph outlining the organization ot the related hternture
,ii~ · ·
/5 rht:J'Ca p,1 '
1.S
he. c:omp Iere <.:ttallon
. · 1l1
· a spccia/ ~-,
• e for
." .
. , . ·s for Evaluating the Related Litcrnture Sect.ion
c;u1oe 11ne.
. 11,ll' .t..l
. , the first draft of the related li.terature section iSteps 1-11
1·
I ·c writrt.n
·
·
,
• , •(111 1: 1~
l
again as you prepare t l1e pcnultm1ate
(Ira f t (S
.. tep ·t4
. 1n
.
l) ·111dt icn
.
.
L l
,.
k
I
.
h
1\11,·1
e apJJrornatc questions ne ow. 1\..tar t 1c manuscnpc w ere
• • 1,lt-4.~ • ,
·11 1.1
• nswc r t 11
.
•
1
1
r. ,ecrion.'
o the paragraphs by (bcribing·'
them.
ed /itcrature sec1ion and use th
Related Lirer1liure Section,, th· e
, • .
· ,It
.VJSJonsseem necessary ·1 cl .
77
_ he order of the headings and suhheadings represent the rclarive
1
2- J)oes .
e of the copies and subtopics? ls the order of headings logical?
1
in1po1tart.
·e summary paragraphs for each of the two or rhree major sections
~ ,'\re rI1c1
·
d .. 1 overall summary at the encl?
;rn at
4
_ Is rbe relation of the proposed .swdyto past aud current research dearly
shown in the summary paragraphs?
ilt'<::
' 11 W;lJt Otlc
5.
What ucw answers (extension of the body of knowledge) will the proposed
,-r,scarchprovide?
Or coherence , C01Jtint1J'tv
..., an d
>t ro another. Check carefully for
)r mechanics.
6, What is distinctive or different about the proposed research compared
with previous research? ls this clearly stated? Is this introduced in just a few
paragraphs?
7. Have the results from your own pilot studies, when appropriate, heen
interwoven into the synthesis of the rdated literature~
iat cou ld not be improved b h
, I ..
Yt e
.r. 1er tt is to demonstrate the 1e
. 1·1. mstrumenta_rio~, the practicality
vana b1LJtyof ob serve d
c~pabihc '.es, or: rhe in vestigator's
s1ble basis for maki11gwise deci-
.ers,
~1:
.
·nt means b}' wh1'cl1to d
.
..
·
.
ctcrmrnc
•
E, cxpcn-
tc:int differences ·i.1110111.1
'
8. What arc the mosr relevant articles (no more than five} that bear on this
research? Underline these references. Are they listed under the first topical
heading?
9. Are these artides presented in a way that denotes their importance? Are
some cite<l so many times they lose their power through repetition?
10. ! las the evaluation of these key articles, as well as all other articles,
been presented succinctly in terms of both procedures and interpretation
of results?
•r_"p<~w~ _ar!_!1ly~ s" rcc~ntly has
:h.
. Although it always· has ·been
I
it lias become mo1·e•1l, equent y
use boo· ks· an d computer proreplete with
J_powe~--'-!:ll:1>'...:'~'!...
(~?~!.!.1~?
..~}'t1_i_~;~j°~~
_I
1
~87f
av~ilable-some of which
ons before caku!ating sample
,He
,, we suggest inquiring at your
institution's statistical consulting office or rnmputer center to find out what
is available. In addition, you also should inquire about sample size calculators that are available on the Internet. These may serve your purpose, but
you will want to make certain that the one you choose does the desired task
using algorithms that are appropriate.: to your research.
Pilot data and a few decisions (primarily related to the error ratcs you
want for the study) allow researchers tu estimate the sample size needed to
78
Writing the Propos~]
Im
find sign i_fica
nee, , f in
it exists in the data. It is possi b_l'.that the e.stj'¾t
sample s,,e ,·equffed will be so large as to be p,oh,bmve, on which e \\I
method and •ueasurcment tools shou ld be reexamined. fo the ideal
power analysis will infor01 the researcher of an appropriate sample :'. cl
based on pe,·missibl, error and the data-not just on acbitrary guessing_
better to find the appcopriate sample size in advance, rathec than after~fact. Both a samp le size thar precludes a significant finding and the us, '
more subjects than needed are a waste of time and effort.
Of
ca,·~
r"~
The use o/ even a few subjects in an informal trial can ceveal a fao,1Ra,
be/ore it can destroy months of work. The same trial may even provid,
fortunate opportunity to improve the precision of the inve«igation or t;
streamline cumbmome methods . For aU these reasons, students and ad~.
sots shou ld not insisr on holding s1tingent, formal standards for explorarocy
studies. A pilot study is a pilot study; its ta,·get is the practicality of Propo,o1
operations, not the creation of empirical tr.uch.
Examp les of purposes that pilot studies might seove include the follow;,,g
l. To determine the reliability of me,rn1remcnc in your own laboraror.y,
under field conditions like those proposed for rhe srudy.
2. To ensure thar differences rhat you c:xpcx;rto exist., do in focr exist-that
1f you are studying the different dfecrs of gender 011 mot·ivation, make
the gender difference ex isrs.
OI'
i~,
SIH·e
3. To "save" a sample thar is <lifficulrto ohtain 11nti!the real rcsean:h prnicct is
undertaken--that
is, it is prudent to tc:sr available subjects until proccxlura[
bugs arc worked out before tc:siing world-class athlete.~.
4. To determine the hcst type of skills to use as an indcrerident variable; for
examrle-w stlldy rhc cffecrs of d;ffercnr .:Jnkle braces on knee mobility-test
jumping vc:rrical!y,horizontally, and while running, then select one.
S. To <leterrnine the frasibility of co/leering at1dio recordings of participant t.1)k
in ,ln e11vfrnn111cnr
where there is a great deal of background noise and highspeed verhal interactions.
The presentation of pilot study results sometimes docs create a troublesom.e prohlem. Readers niay be led in a dvertently to expect mote of pilot
work than it can reasonably dc:!iver.Their concerns wirh the limitations may
distract from its Jimired use in the mai n line=of rhe argumcnr. being advanced
hy the ,utlw,. Acco,·d;ngly, the best cou,·se ;sto onake no mo,·e of thr p;lo,
srutly (·han it honestly is worrh-most a re no more than a report of limired
experience under less than perfectly co111rollcdconditions-and
do so only
when the report will best illuminate the choices made in rhe rroposaJ.
Content of thr. Proposal
the data. It is possible that .1
·
1.11e
esti
rge as ro be prohibitive, i.n whichfl.J;ite~
)uJd be reexamined. lo the ideal . <';i¼
;earcher of an appropriate sa Clase,~
I
,
lllpe ,
ata-not Just on arbitrary guess ·
SJ~
I •
•
ing
l
.e size tn ac.lvance, rather than f . fis1 . .. 'f·
.
a tet tL
- es a s1gm tcaot frn<ling and the u oe
te of time and effort.
se Of
79
1
rn ~nforma) trial can reveal a fatal
fl
-k. f~e- ~ame rrial ,~ay even provid:\v
e prec1s1011of the ltlvestigat ion o
··. ,'.ill tJ,,esc
· • reasons student s
<t
d t to,
'
. an adv·
~enr, formal standards for explo
I·
. . .
..
.
.rater,,
t.s t-i1get is d1e pract1calicy of pro
,.
:al truth.
Posed.
-Jiesmight serve include the £oil .
owing:
a.~uremcnt in your own hb
.
, oratory 0
·posed tor the study.
' r
~xpccr to exist, do i11facr exist-that is,
cts of gender on motivation, make SUJ'e
obtain lH1tiJthe real rcseard1 f'ro·
.
.
·
, Ject ts
tc.~ravailable suhjects Until proceduruJ
>rld-d,i.~s athletes.
o use as an independent variable; for
:nr ankle braces on knee mobility-rest
•hilc rnuning, then select
011
c.
g audio recordings of participant talk
deal of b~ckground noise and high-
Lt
some tim es does create a trouble:enen tly t~ expect more of pilot
.oncerns with the limitations rnay
·?Eth e argument being advanced
is co make no more of the pilot
io more than a report of limit ed
1
_1edG:ondir~ons-and do so only
ices made JO the proposa l.
. e to pilot work may be made in supponing the broad
h
.
f
h s
·r
. 1-eferenc
r~rid •
·es selected consequent to t e review o researe . ome p1 ot
1 5 eracegi
f
k
I
f
·
I
,ese:1rc
in fact, be treated as one o the wor s wort 1yo review. More
stLtdjesmaYhowever,the results of exp loratory studies are used in supportco111r11<>~1r~
ocedures proposed in the section dealing with methodology.
;11g specafi; pXilot study represents a formal and relatively complete research
1
Whe_n r e ~oper to cite the work in some detail, including actual data.
·t·Orr' itheisprelimimHy
p
kh b
· c
I
1· · I ·
L
·
cf
wor
as een mwrma or 1m1tet, 1t _ITJ~Y
·~ _11~tro\Xi'hent
footnote to the main text. ln the latter case, it may be desirable
J ced as.da a roore dera1·ted-account
--- o f the wor k in
· a sect10n
· o f t I1e appem 1·,x,
<u
ro p~·ovirbereader the choice of pursuing the matter further if desired.
k':'1VU1g e
Murphy,s Law: Anticipating the Unexpected
Law dictates that, in the conduct of research, if anyth.ing can go
·
g , it IJrobablv· will. This is accepted by experienced researchers and
wron
,·h advisors hut rarely is considered by the novice.
res ear
, '"
Within reasonable limits, the pr9p(.l~aL_i~_the
..pla.ce.. tu . pro.Yi.de.for_confrontation with the inexorable ~-pe~a~i.9-n_
of ~ lt\r_phy's l.aV\',Su_bject!l!!rj tion
can~ot be prevented, btit its effects ~an he circumscribed by ~areful planning.
The potentially biasing effects created hy nonreturns in questionnaire studies can be examined and, to some degree, mitigated by plans laid carefully in
rhc proposal. The handling of subjects in the event of equipment failure is
far better considered at leisure, in writing the proposal, than in the face of
an unanticipated emergency. Field research in the public schools can provide
a range of surprises, including indisposed teachers, fire drills, and inclement
weather, all better managed by anticipation than by snap decisions forged in
the heat of sudden necessity.
\I Iurphv's
,
Equipment failure may interrupt carefully rimed data collection sequences
or interview protocols, or temporary computer breakdowns may delay data
processing and analysis. At best, such accidents will do no more than altt:r the
time schedule for the study. At worst, they may require substitutions or substantive changes in the procedures. Each step of the research process should be
studied with regard co potential difficulty, and plans in the event of a problem
should be stated in the appropriate place within the proposal. for instance, if
unequal subjea attrition occurs across groups, the type of analysis to be used
with unequal Ns should be stated in the analysis section of the proposal.
It is impossible to anticipate everything that can happen. A good proposal, however, provides contingency plans for the most important problems
that may arise in the course of conducting the study.
80
Writing the Proposal
Anticipating the Analysis: Do It Now
The proposal is the proper place to reveal the exact nature o.f the analysis
well as anticipated plans in the event of emergency. For many students, e; ils
cially master's candidates, the ana lysis, if statistical, may represent ,~c1
knowlcdge recently acquired and not fully digested. In addition, the c1.1::
tomary time lirnitation of f 2 to 16 months, by which the master's candidat
is !:ound, adds to the_difficulty. The cand idate m~y even be in ~ 1e middle ;
0
a hr.st formal course in techniques of data reduction and analysis during the
same period of time used for wn structing the proposal. Consequently
students find themselves in the awk ward position of having to write lucid!'
about their analyr.ic roofs without yet knowing the entire armamentar iu;
available. As untenable as this position is, and as much symparby as may be
generated by the student's advisor or friends, the omission of a fol1explana.
tion of the analysis in the proposal may prove to be disastrous . Countless
unfortunates have found themselves with files full of unanalyzab Je data, all
hec.:ausethe analysis was supposed to take care of itse.lf.A step -by-step antic.
ipation of rhc analysis to be used is also a double check on the experimental
design. Fina Hy,as we suggest in C:hapr.er 5, in the case of qualitative research
forethought in planning the atrnlysis can identify prob lems in data coUection
thai· might have a direc..:tbearing on the persuasivcnes~ of larcr conclu.sions.
Oescriptive, survey, and normative studies re4uin: extensive data reduction to produce rneaningiul quantirative desc.:riptionsand summaries of tlw
phenomena of interest. Techniques for de1c::rn1.ining
sample charac..:terisci.cs
may be different from those anti,:ipated on the bas.is of pilot results, or the
study sa..mplemay be skewed, resulting in the need to discuss techniques for
normalizing rhc data.
Statistica.l techniques ;ire. founded o n assumptions relating to sample charbetwe en that sample and its respecrive population. The methods one intends to use to determine wheth er the sample
meets r.heassumptions implicit in the ant icipated ana lysis should be clearly
stated. For example, many sr<1tistirnltechniques must be used only when one
or more of the following assumptions are m::r: (,1) normnl distribution of the
sample, (b) 1·andom and independC:~
11t seJeccion of subieus, (,:) linear rela'tionshtps of variables, (d) homogern~ityo( variance among groups (in rcgre.~sion analysis, rhis is called hornosced:lstic.:ity),(e) independence of sample
''means and variances, and (f) units o.t measure of the deµcndent variable on
an interval or ratio s<.:ale.
:Kteristics and the rdationship
'
-- Accordingly, the process of selecting a star.istic brings with it a m1mher of
subsequent questions. What. merbods will be employ1.:dto der.ermi.neth"ic
aSS\1
_!:r!ption ~ have be:: 1 met? Wh,i'
t analysesw ill be USf.din i:he-cve1
1Ti:11e
Contmt of the Proposal
8I
Dolt Now
·cvcal d1e exact nat ure of the anal .
.. of emergency. F.or many s tudent Ys1s'<ls
Iys,s,
· 1'f .stanst1caJ,
· ·
s, es1,
may represent l"\:,
ot fully digested. In addition ti 11\l\v
> le ClJ
tonths, hy whid1 the master's ca d' s,
n td;it
l· 1ate may eve n be in the midd
l e
canu<
data reduction and analys is dur,· e of
·
ng tbc
:t.ructmg the proposa l. Consequ
ently
.
.
:ani posmon of ha~ing to write lucid!,
.t knowmg tl1e ent ire armamenta . }
•
•
•
<
< l'lllni
n 1s, and as much sympathy as Illa b
~ricnds, the omissio n of a fulJexpJ: e
.
na.
my prove to be disastrous . Count!
,ith files full of uuaoalyzable data e-s
lsl
k
'a
. 1·
,1 .c care o f ttse f. A Step-by-step ant ic.
o ,~~ouble check on the experi.rnental
:r ::i, 111the case of q u.aJitarive researcl
. I 'f
1
n t( em1 r problems i11data colJection
persuasiveness of later conclusions.
mrdics require extensive data redtKe descriptions and summaries of the
. determining sample characteristics
· on the hasis ot pilot results, or the
11
the need to discuss techniques for
.ssumptions relating ro sample charha( sample and its respective popu.t~ determine whether the sample
t1c1pacedanalysis should be clearly
'liqucs muse be used on!y when one
n~et·: (a) normal distrihution of the
ect ion of subjects, (c) Jiuear relavar iancc among groups (in regrcscity), (e) independence of sample
sure of the dependent variable on
tatistic brings witl1 it a number of
he employed to determine chat
:es-·wiH be used in the -e-veni:
·die
. ns are nor met? W ill the planned aJ1alyses be a p pro .pri ate io .th e
lost
. so t ha t there a.re unequ a l n umb ers o.f subi ects o r mals
5 11111.P~~~
· ar e
:,~
b· 1
·ectS
SLI
· ns ).
cvcnr . erent co n d'mo
.
;,, rhe d,ff, J is section of the proposa l sho uld be ou tl ined to co rrespon d
1'he arl~-::tives of the study so th at eac h ana lysis w ill yie ld evidence rel a tth O 1
,virh e . ·espondfog hypothesis . l n add ition , the reade r s hou ld be a ble
jog ro a ~ou how all da ta co llecte d are ro be ana lyzed . An efficient way of
1·0 dt:rcr1:1
~~e1g rJ1is
in Ta ble 4.4.
phs,111
' is by us ing a.mo del simi lar to tbe. one shown
.
,,,com are to be p resented 111 tabular or grap h1c form ll1 rhe comp lete d
Lf·rdara
an examp le of one such tab.le, includin g pre dicte d figur es, often will
repOl ,
T:ible 4.4
Example of a Table Showing How Each Hypothesis Will Be Tested
rlJ•pothesis
Variable
Analysis
I. The sclf-wnccpt of
J-vcar-olds is higher than
th~t of 5-ycar-olds.
Sum of self-concept
suh~cales
MANOVA with
follow-up if age factor
is significant
z. The self-concept of
Sum of self-concept
suhscales
MANOVA with
follow-up if age factor
i.~significant
Sum of self-concept
subscalt:s
MANOV/\ with
follow-up if location
factor is significam
4. Nint:-ycar-oldsuburban
children will have higher
self-conct:ptth~n 9-ycar-old
rural children.
Sum of self-wncept
sub~calcs
.'vlANOVA with
follow-up if location
factor is significant
5. The peer sociability of
7-year-olds is higher than
that of 5-year-okls.
Peer Sociability Index
l'v1ANOVAwith
follow-up if age factor
is significant
6. TI1e peer sociability of
.9-year-oldsis higher th.an
that of 7-year-olds.
Peer Sociability Index
MANOVAwith
follow-up if age factor
is significant
7. Suburban children will have
higher peer sociability than
urban children_
Pet:r Sociahiliry Index
M/\NOVAwir.h
follow-up if location
factor is significant
9-year-olds is higher chan
that of 7-year-olds.
3. There is no difference
between rural or suburhan
children's self-concept for
5- and 7-year-olds.
NOTE: This i~presented for illustrative purposes. An acrnal srndy may have additionalhypotlie.scs.
82
Writing the Proros~I
he helpful to rhe reader. The purpose of a tahlc or figure in a research l'e
is to summarize rnarerial and to supplement the text in making it dt 0 tt
undcrsran<lahle. Tables and graphic presentation may serve the same ;tly
pose m
· a proposa I.
llr.
Because of thei.r display quality, che inclusi on of tab les in the Prop
. or researc I\ des1.gn. For 111
. srance) so0 Saf
may expose errors ol.. ana lys1s
.
committee readers niay not <letec.:tthe use o f a n incorrect error ter n, & :e
reading of the text, but cmc glance at rhe degrees of freedom co.lu1 0 1 i<I
1111
an analysis of variance table will reveal the error.. In ana lysis of varianc:
comparisons, inclusio n of severa l tables may expose the presence of noniri.
dependent
variables. Although most comp l.e red theses and d issertatio n s n0
.
longer include analys is of variance tab les> at the proposal stage they a1ay be
valuable for stimulati ng feedback from yotu· comm ittee.
rf the analysis activity of the project is stu<lied carefully in advance, many.
headaches as well as heartaches may be avoided. Ir may seem to take a n inor.
dinate amouur of time to plan the analysis, bur it is time rhat will nor have
to be spent again. As the analyses are complewd, t he resuJrs ca n immediate ly
be inserred .in the results section, and the researche r can comple te rhe pro ject
with a feeling of fulfillment rnd1er than a frantic scramble to make sense our
of a puzzle for whic.:hsome of the pieces rnay pro ve ro be missi ng .
The StatisticaJ Well: Drinking the Greatest Draught
Students usually can expect help from rl1eir advisors witl1 the design of statistical an:dysis. At minimum, an experienced advisor will have some sugge.stions about the type of analysis that would be most appropriate for rhc
proposed investigation. i\.lany ~l~p~rtn,e:11.
~ _in~:l_~q_
e spe~·ialisi~. ~h~>__bayq _r~.of their profestisticaJ consulration with graduate students as a primary ·1;art
-o'ii"r.~iae-~r:{ristic;lsorial responsibi!i ~ ~ Ot her dc.•par-rnren tsi. vork cl®; lj,. ~vii!;:
be in <lepa~~t~
of ~d u~;ti·~.~-aI
and ..c(>mp uc~r co11; 1ltams who
choJogy, psychology, comp ut er science, or 6us.inessa a iiuh istrnt 1no;'n
n-. - You sh ou ld not, however , operate und er the faulty impression that when
the data are co Uected they ca11be turn ed over to a handy statistical experr
who, having au intimate relationship with a computer, will magically transforrn raw data into a finished form of findings and conclusions. Just as you
c.:annot expect the analysis ot data ro take care of itself, neither can you
expect a statistical consultant ro take care of it.
ma
y
psy~
1
The assistance of a statistician or computer cons ult anr, invaluable though it
may be, ordinarily is limited to rhe technology of design anJ data analysis and
help witl1 using a packaged statistical. program on a com puter. The conceptual
a
ll
h
;1
p
It
al
V{
m
-ii
Sc
Content of the Proposal
of a rablcor figure in ·1 resc I
'
ate l iplcment the rext in making . tflc1,..
.
.
, It c/ ,,
,resentarion mav serve th
e<lr/
- . .
c san,e J·
P11~.
c inclusion of cables in th
e Pro
csearch design. For instance Posa,
.
.ise o f an incorrect
error re >, So,,_
.,,~
: t11e degrees of freedom rm u:olll ii
1
.
a I t he error. In analysis ofco ur11n
. •11
. _
,
varia,
. may expose the presence of no/c~
mpleted theses and dissertations tri.
s, at the proposal stage they ma ~~
'our committee.
)' e
midied carefu lly i11advance
'd
, l1Jany
·.<'.ied. ~n a~ seem to tak e an inor.
,1s, but it is tune tl1ar will not h
1 I
P cte<,I the result s can j m ......
ed· a~
.
.. u .u
tare/y
esearcher can comp lete the p . .
c
.
ro1cct
rantic scrambl e to make sense Out
nay prove to he missing.
!t
1e Greatest Draught
· advisors with the design of stad advisor will have some suggesld be 11~ost appropriate for the
nclud e sp~a list.swho have stas ~ primary pa.rt of their p.roi;.
K CJOSelywifh
outs1ae Statistical
=~-8:'tments oieducat ional ~
SJriessad'ffimis~
he faufty llnpress ion rhar when
!r to a handy sta tistical expert
ompute.r, wiJ/ magically t1·.u15s and conclusions . Just as you
are of itself, neither can yo u
t.
msultant, invaluable though it
f design and data analysis and
n a computer. The conceptual
83
stud and the particular form and charact eristics o.f the da ta
Y
·
· d
I
1
, rids o.f rhe
th . estigator's province - co be explame to t 1e cons u tant,
JeJl"',\ced
are L1
~kwwvi·se
of resLtlts is a logical,. nor a. techni. •n•' •
e • , rhe
· interpre- tation
iI, c.: vice
versa. d thus is a responsibi lity for which only th e mvesagacor 1s
,or >f:1£10
. nan
--~--~
----11
ope
•
-- a red·
,. ' Iv Jrep
prllPcrY E
•
0
. a Friendly Computer
) fineItng
(. n
. . uf interrelated decisions must be addressed as you contemplate.
\ \"1necv
.
I
. . l
I
dd
' .,. . ' h e of thc studv . First, w 1at stat1snca too s are nee e to ana, -~-:..,......---,.-"7"-;--""7--;;----;---,,
l}·s1sp as .
11
rl ~ ;Ht;J
·.
• rrectlv? Then, after the statistica l rnerhods are selected, what
1e data
co
.,
.
d
fy-1,e
Cl .. t· efficient way to do the analysis? (Nore that ir is not a good i ea
·
) A
·
''· rhc·r11uS
. . I e order of these two questions.
· t one time,
stu( Ients had few
·
·
"
h
I
I d h
·
n·veJ.sc t1
to1 .·
.· h regard to methods of analysis. Either r ey ca cu ate t e stat1s·I o1n:s wit
.
I h
.
.
'·elves on hand or table calculators,
or they use< t e u111vers1ry
ri(s rhems .
.
d
.
f.
. . I
·. 1-.,
1,une. computer to complete the analysis. To· ay, a vanety o. stattstJca
1n·lll1
,·, available for computers. These packages
.
' 1rnges, ~re
are extremely versap .1c
.'
·
J
after a user
off
uJe
anci,
. learns the . basics, .are relatively easy to
. use. Par.t
L
·h
planning
mvolves
makmg
choices
among
a
growmg
numDet
o
rcsearc. '
·
,1ttractiveoptions for analysis.
.
The first step in selecting a software program to analyze your data 1s to
find out what packages are avai.lable. Your advisor., other graduate students,
and rhe university computer center can be helpful. ~"v1anyuniversici.es have
consultants to help students and faculty and may havc agreements with software companies that greatly reduce the cost of the software. fn addition, if
a needed program is nor available on a computer in your department, most
universities have thern available in a centralized site. It is important ro note
here tnat such careful search is just as important when selecting sofcware for
analysis of qualitative data as for the numerical data of quantitative designs.
The first step in dcciding how to analyze your data is to eliminate all
packages that will not allow you to do the desired analysis. This may involve
more than assuring yourself that the proposed statistical operation is available. For instance, you should be interested in the number of subjccts and
variables the program or computcr can handle, whether data can be convened to other forms if you need to conduct additional analyses at a later
date, and, shoul<l data assume an unex.pccted form, whether other tests that
might be needed are available with the same computer and package.
fortunately,
rhe most common and powerful statistical packages
(Statistical ~alvsis Svstem ISASJ..and -5tatistica L P11ekage for the Social.
-S~
ces
(SPSSJ)·;;·<~~;~~-~;~
.a~~·;;;
for personaLcomput~1:~~
'fhe._pa&ii'ges can
.- .. ·- ~
---,c_~;...
,__
acco1nmodate ve-;y- larg C( iatisc ts for analysis and storage (the limits, of
bi~
,.-
84
Writing the Propusal
l
course, will depend on the configL1ra
tion of the persona_! computer if
is be.ing used for the anaiy.sis), can perform vir tua lly any statistica l ~~
researchers require, and are available on most unive rsity campuses. Altho te~
sorne may find these packages slightly more difficult to :master than llllt&h
spreadsheets that have statistical opti,ons and those programs designed to st
form a single analytic task, several benefits accrue from learning to use oneer.
0~
more of the large statistical packages. Among these advantages are:
p°
l.
You will lie able co ohrain help with your data ar,alysis. because many facuJr,
mcmher.s, graduate students, anti consultants are experienced with rhe larger
'
well-known .~taristicaJ packages.
2. ff rhe data muse be reanalyzed, iris very likely that the package can complete
the analysis \V.ithour the time-consuming :prohlem of rcformartiug or rnypin~
data. In addmon , these programs can eas ily export data mto form:~ts that c~n
be used hy other more specific software programs.
3. Familiarity with a commonly used program will serve a studeut well when
moving on to an academic appoinrmenr-rhe
packages ,Ire avai!ahlc on !ll()st
campuses and little srart-11p time will be wasted on le::irning a new program
to analyze data.
l\o matter what type of computer or package you choose, it is absolu tely
essential to back up your fiks. This ca o be accomplished by simply saving
computer files on borh the hard disk an d on a flash drive O( CD . No matter
how dilig1mt you are about saving and backing up your file~, you also should
save a harJ copy (a printout of your data) in more than one place. This
applies equally co word processing text for the proposal document. i\s a
matter of practice, have both a printout and an electronic copy of the data
in at least two places. An extra word of camion-save
frequently. You can
do this rnanually or by using the autosave foatme of your word processor,
or, just ro be safe, hoth. Remember, data arc like eggs-they are most secure
when stored in more rhan one hasker.
Selecting the wrong statistical package rarely is fatal in the research
process. The selection of a software package tha t does not meet all your
needs may require additional tirne for entering data or may de lay comp letion
of the analysis. A little careful planning, however, may eliminate waste and
reduce aggravation at a later date.
The Care and Nurture of Consultants
To obtain tcdmical help from an advisor or consultant, you should be prepared to provide bask conccprs about. rhc content domain of the investigation,
Content of the Proposal
ration of the persona/
c
computer .
if"
, perwrm virtuaHy a
ny srat1stic "flt
on mosc university campuses. Alt~!t~
dy more difficult to master th 110118!1
>ns and those proorams d .
a11llio..
o
es1g11ed
t0 '"{
tefi rs accrue from learn·
p,,.
mg to USeo "fs
Among these advantages are:
'~c()t
h your data analysis because man f
>nsu!tanr.~ arc experienced w·th hy aclllty
i
t
c lar,,
.,er,
v:ry likely rhat rhc package can co
mr,g prohlern of reformatrin11 <)r. 11lr,fere
.
.., tetyp·
n cas11y export data into t·
rog
onnat.s rliat
,are programs.
cuQ
orogram will serve a stuJeru Well Wh
~nt-thc pad<ages are available on in en
be wasted ou learning a
Ost
new pro1;ratn
package you choose, it is absolute l
, be accomp lished by sin1pJy savi/
I on a flash drive or CD 1s.1
g
J.
• i ,o matter:
up your files, you a/so should
a) m more than one place. Th.
for th e proposa
·
l document. As rs
a
~nd _an electronic copy of the data
.aution-save
frequently ·y
t
. . ou can
e e~ture of your word processor
ire like
· eggs-th
·
ey are most secure'
:tn~
e rarely is· fata t in
· t hc research
that
does
not
meet· all your
:age
·
mg data or may Jelay completion
)Wever, may eliminate waste and
1ts
)r consulta~1t, you should be pre1tentdomain of the investi<>-ation
0
'
85
. e 1-eview of what js to be studied, a clear picture of the form
.. 111
din!5 a concisda preliminary estima te of alternative designs that might be
1:"w
\viii rake, a;e demands of the proposed research. In aJJition, whether
' pfvpriare to. g design statistics, or computer programming is sought from
11
P
'
. .
1cern1n
:1dvicec~' . dvisor, from a departmental spec1alist, or from an expert source
~our proiecc~ department, basic rules exist that must be considered if you are
11ro r nose info.rmacionand help for the smallest cost in valuable cone,xccrfll
,1cancheJ
10 g
. e
sulcMion run .
I
l · Understand the consu ltant 's frame of reference. /\s with any other sitknow enoug l1 aoout
L
[?//.e n ·,·nvolving exten dc d com municacwn,
· ·
·1r 1s
· use f u I to
uauo
e language, predi lecrions, and knowledge base of the consultant to avoi<l
1
r;ri ous misunder standings and ease the process of initiating the transaction .
:~eseMChconsultants are professio nals whose primary interesr is in che
rocess of research design, statistical analysis, anJ the application of com·
~uters in research. They use a system language unique to statistics and data
nianagement and appreciate those who understand at least the rudiments of
this vocabulary. Correspondingly, your consultant will not necessarily undersrnnd the system language to be used in the proposal, nor the peculiar characteristics of the data. For example, it cannot he assumed that the consultant
knows thar some of your data consist of repeated measures. Similarly, it
would be unlikely for a statistician to know whether these data arc normally
distributed across trials.
The consultant .:.-annocbe expected to make decisions that relate to the purpose of the study, such as those regarding the balance between internal and
external vaJi<lity. Some designs may maximize the valt<lity of the differences
that may be foun<l, but correspondingly trade off external validity, and thus
the generali:tahility of the findings. Decisions concerning the acceptability of
such research designs must he maJe hy the proposer of the study. The grounds
for.such a determination rest in the purpose of the stu<ly and thus in conceptual work completed long before the consultation interview.
Consultants can be expected to evaluate a propose<l <lesign, assist in selection from a group of alcernative <lesigns, suggest more efficient designs that
have not been considered, and propose methods for efficiently completing the
analysis. Often they can be most helpful, however, if preliminary models for
design and statistical analysis have been proposed. This provides a starring
place for discussion and may serve as a vehicle for considering characteristics
of th.e data that will impose special demands. Consultants can provide information about computer programs, the appropriateness of a particular program f01·the proposed design, and the data entry techniques required of these
programs, ,\gain, some rrcliminary preparation by rhe stu<lent can make
the consultant's advisory task easier and work to guarantee an optimal selection of procedures for processing raw data. This preparation might indu<le
talking with other students presently engaged in compurer use, reviewing
86
W riti.ng the Proposal
1
material on stat istical p!'ograms, and visiting rhe computer center ,
'\lr ~~
up<larc on availa ble services.
1\lormally, rhe statisti.cs and computer specialists in a university setting
he~iegedby frant ic grad uate studem, and busy faculty colleagues, all in :td~~t
rion ro r.hedemands of rheir own scudenrs. Funher, they may be responsible f "
the management of one 01 more functions in tl1eir own administrative Unit <>r
in the computer center. hnally, as act ive scholars they will be conducting th ~r
mvn research. Hoch the picwre .=indth e lesson should be equal ly clear to so, Clf
· ass1s1,U1cc:
·
Sca;1st1rnm5
· · · an d consuJtartts are bllSy peop le. 1'hey<'
tie,
one ~ccI{mg
.
.
.
a11
provu. Jc et'f.cct1ve
ass1sta11ce
on Jy w l 1en .mvesugator-s
come wit. l1 accurate expce,
tatiom for r.hekine! of help ,:iconsult ant can properly provide and come full
prepared to exercise their own re.~poH
sibiliries in the process.
y
Ru.le 2: Leam the lcmgucrge.The system languages of measurernenr, co Pllt111
er.~, experimental design, and borh inferential and descriptive statistics are
use<l in varying degrees in the proce ss of techni:al consu lrarion for lllany
research proposals. No one, least of all an experienced consultant, expects
flucm mastery in r.he novice. You 11111.;
r, however, have a working knowledge
of fun<lamenca] c:onccprn.These ordi narily include measures of central ten.
dency and variabiliry, distribution models, and the concept of statistical sig.
n ificance. l:\a.~ic research de.signs, such as those described in inrroducrory
i:escarch method books, -~hm1ldbc familiar co any novice.
Jr is, of course, prciernb!e r.o complere at least one scatisrics course before
actcmpr.ing any srndy that will demand the .inalysis of quantitat ive data . If, as
somccimcs is rhe case, the student is learning basic statistics concurrently with
the preparation of rJ1eproposal, special effort will have ro be concentrated on
preparin~ for consulratious concerning design and analysis. The sit11ationwill
be awkward ar. best, air.hough many consultanrs will rem.=iinsymp:uhcric anu
parien.r if sr.u<lents:1rchonest ,tbout thcir lin1it:1tionsand wil.!ing to exerr heroic
dforr once it bccomcs dear which tool.s and L·onceprs111usr.
be JUastcred.
Beyond rhe problem of mastering enough of thc languaf:\eto parricip,ttc in
usciul di5cussion is tJ1emore subtle prohlcm of undc::rsrandiug the parricuhr
analysis and technique!; selecred for rhc stucly. You musr not drif.t inro the positio11of using :t scatisric:¼I
tool or a mcasurcmcm tcclmique thar you really do
nor under.~tand-evr..n one en.dors,'.<land urged by rhe most compet.ent of advisors. lJltimatdy, you will have ro 111,¼
ke sense our of che n:sulc.sobr.aincd
throu~h any analysis. At rlnir poitll, shallow 01: incorrect interprernrion~ will
quickly betray a failure to unrlerst:rnd che natmc of rbe ;1nalirsis. You al.~owilJ
have to answer questio11s::ibolli chc iiudi11gslong ,tfrr.r th,'. advisor is no longer
arnnnd. Expert technical aclvicccan be :111invaluable asscr in devising a srrong
propo.~al, bm in the final ,:inalysis,such advice c.=innorsuh.stirute ior rhe r_:ompeten.ce of the investigaror.
Ru.le 3: Under,;tand the {.noposed stud)'· If thc novice rei;card1er does noL
\mdr.r.srand r.he study $Uffic:.iemlyto idencify anci ask important and explicit
Content of the Proposal
uns, and visiting ti
le computer cente
r fot
computer specia/jsts in a
.
.
•
udenrs and busy facul
university sertiri
n students. Further ti ty colleagues, al/ in~,.
f
.
, 1eymay be res
~rl,
e .unct1ons in rhei.row d . . Ponsibt'
as
·
h
n a mrn,strativ ~,.
e 1111
• 1t
accwe Sc olars they . ·11b
d tl I
e co11do · 11',.
n ,e esson sh0uld be c
Ct111g
llil
ns and consulr:ant.sare b (Jually clear to so·~
Nhen investigators come us!teople. They~·
isulrant ca,1pro ·J.
W'.t l accurate Cl( ~
.
. ..
pei Y provide and
~
,pons1b1lJCies
in the p
conie fti1
,
rocess.
~,
,ystem larrguc1gesof rn
rJ . f
easuremenr c
i tn crential and desc··· .
, OllJp~I
.ipt,ve stati t'
·
rocess of technical
. . sics ~i..
f
cousultar
~
"
100
o a II an experienced consu(can. or Fllany:
nusr, h<>weverhave
t. ·
t, C)(p~
.
•
a wor,ong k
rdmarily include n1ea
nowledne
I <>
sures of ce
models, and die conce
ntra ten,
such as those des ··bpdr~f s~atisricafsig.
· ·i·
ci, e m mtrod
-am,,ar to any novice.
ucro'JI
Ip Ic.te a r least one sra .
nsr,cs course b r
1d t I1e annlysis of q
. .
ciore
.
uam,rative dara If
earnmg basic staristics cone .
. , as
aJ efforc will have t b
urientJy Wirf,
g design and anal ~ e con~entrated on
:onsulcanrs wl11ys1s:The s,ruarion WI/I
. 1· . .
rema111syrnpathct
,r 11111rac
1ons and ·11·
,c and
w, mg to c xerr. Iiero1r
.
sand conce
PCSmusr be tnastercd
...
tough of the lan u·
oblea1 of d g age to parncipare in
1111 ersrru ··
JC!ingthe particular
itndy y
.
. ou must not drift in J
mremenr tedmique ti .
to tie POSJ·
tu·gcd 1,, I
!.lt you really do
,•
} t le cnosrcompeteII(' of advi,e .sense our of ti
!low or · · . . '.c resulr.~obtained
.
m_co1nee lllter.pretMion.swill
nature OJ the a 1 .
'
na ys1s. You also will
gs long .ifrcr the adv·
·
i .I
isor rs no longer
ova uabfe asser in de ..
:I ·
vismg a strong
vice cannot snbstiture for the com-
w,
ff the n<>v,·,..
J
~e
f . ·
. resr;irc 1r, . d oes not
y and a:;k Hnportam an(/ ex 1· .
p ICIC
87
. ns. t hac lack is a major obsrade to a succe.ssfol consultation. On ly
sultant underStands the quesrions of ceucral inrerest in the study
rhCcon
1vhen
'ble co translace chem inro the steps of sta tistical a nalysis
tfuesrio
.
pOSS I
is ic
. n of the appropriate computer program . Even if you employ a
·
· 'f1 you
:ind ·selecrio
only co help you w ·it h data preparation
an d a na Iys1s,
sultanr
.
con- communicate exactly what you want, you may get back a pnnto ut
cannot analysis rhar is icrelevanr co your needs. Fu.tther, a host of specific
111
fr<> a~ rs assoc.iaced with the nature of the study will condition the
o11sr
ra1n
.
.
d
c suIcanc's decision about wh,ch an_alysts to recommen .
011
c You sliould be ready ro provide answers to each of the follow ing
qllesrion.~:
1.
Wh,1tare rhc independent variables of the study?
2. What are rhc dependenc variables of the study?
3. What arc rhc pot.cnrial confounding variables of the study?
4.
Whar is the measurement scale of each variable (nominal, ordinal,
interval, or ratio)?
5. Which., if any, of the variahlcs are repeated measures?
6. What, for eac.:h variable, arc the reliability and validiry of the scores
prodm;e<lby the instrumcnrs?
7. What are the population distribution characteristics
variables?
for ea.:h of rhe
8. Whar. difference between dependent variables would be of firactical
significance?
9. Whar are the monetary, safety, ethical, or educational
if a Type I error is made?
risks involved
10. What is the natul°e of the loss if a Type H error is inade?
In summ ary, before cons ulting wit h a techni ca l spec ia list, you must be
able to express exactly wh at the stu dy w ill be designe d to acco mp lish, identify the help needed in pro duci ng suc h a des ign, aa d prov ide a ll the e..xplicit
derails the consultant wil l need io formulating advice.
The Scientific State of Mind:
Proof, Truth, and Rationalized Choices
Scientific inquiry is not so much a matter of elaborntc technology or even
rigorous method as it is a particular state of mind. The processes of science
88
Writing the Proposal
rest, in the end, on how scientists regard the world and their work. Altho
some aspects of scientific thinking a1·e subtle and elusive, others are llgh
These latter, the basi.c attitudinal prerequisites for the conduct of scien;.~.t.
11
inquiry, are reflected in the way a novice speaks and writes about propo ~
research. More directly, the proposa l will reflect the degree to which :~d
author has internalized critical attitudes coward such matters as proof, ttuthe
1
and pub licly rationalized choices.
What matters is not the observa nce of particular conventions concerni
phrasing, bur fundamencal ways of thinking that are reflected in the seleng
tion of words. When, for examp le, students write, "The purpose of th~·
study is to prove (or co demonstrate) char .. . , " there always is the clange~
ous possibility that the inrcnr is to do just chat-co prove whnt they have
decided must be true.
Such phrasing cannot be dismissed simply as awkwa rd or naive. Studenrs
capable of writing such a sentence withour hearing at once its dangerot1s
implications are students wirb a fundamental defect in preparation. They
should be allowed to go no further until they apprehend both the nature of
proof and the purpose of research in the scientific enterprise, for clearly nei.
ther is understood.
Proof, if it exists at aJJ in any useful sense, is a probabilistic judgmem
based on an accumulation of observations. Ordinarily, only a series of care.
ful replications ca n lead to the level of confide11ce implied by the word
"proved." Research is not an attempt to prove or demonstrate, it is an
attempt to ask a carefu l question and to allow the nature of things to dictate
the answer. The difference between "attempting to prove" and "seeking
proof" is subtle but critical, and a scientist must never confuse the two.
If scho lars have no illusions about proof, it is wrong, neverth eless, to
believe that they never care about the direction of results obtained from their
research. As human s, they often a1·c painfully aware of the distinction
ben-veen results that will be fornmare or unfortunate for their developing line
of thought. As scientists, however, they recognize the irrelevance (and even
the danger) of allowing personal convenience or advantage co intrude in the
business of seeking knowledge. In the end, researchexs must sit down before
their facts as srudenrs and allow themselves co be instructed. The task lies in
arranging rhe context for instruction so char rhe answers to questions will be
clear, but the content of the lesson must remain in the facts as revealed by
the data.
A second critical sign of the student's ability to adopt the scientific viewpoint is tl1e general way the matter of truth is tr eated in the proposal. When
students write, "The purpose of rhis study is to discover the actual cause
of . .. , " there is danger that they think it is possible to do just that-to
Content of the Proposal
the world and their woi;k A.!
. th
subtle and elL1sive others
O\ioL
. .
'
ate QO,
Jws1res for the cond uce of s . 1101
Cle . .
e speaks and writes aboltt p tlt 1ti1:
vill reflect the degree to wlr?Po~d
ltch ·
1
:oward such matters as proof
tic
' tr11th
)
f parricular conventions cone .
k'
h
ern 1n
·&
mg t at are reflected in the
.
"Th e purpose ofsefec
dems wnte,
.,
t
. the danger
thJs·
· . h.," r·here a 1ways 1s
ist t at-to
prove what they I
,
lave
ply as a,:kward or naive. Students
>ut hearmg at once its dange
.
rous
ema l defect in preparation. They
~ey ~~_prehend both che natu.re of
::1cot1f1
c enterprise, for clearly nei:ense, is a probabilistic judgment
. Ordinarily , only a sei:ies of care::onfidence implied by the word
p.rove or demonstrate , .it t's an
ow the nature o! things to dictate
mpting to prove" and "seeking
mu st never confuse the rwo.
of, it is wrong, nevertheless, to
on of resu lts obtai ned from their
1 .fufly aware of the distinction
munatc for their developing line
>g111zethe irrelevance (and even
e or aJvamage to intrude in the
=Searchers must sit down before
:o be in st ructed. The task lies in
the answers to questions will be
:lain in the facts as revealed by
ity to adopt the scientific view. trearcd in the proposal. When
s to discover tl1e actual cause
s possible to do just that-to
.
Inmate
89
face of realiry at a single glance. The most fondamenral
d
d
., ·ern rhe LI, •11be required if such st udenr s eve r are to un erstan ·, muc
cJ1s'
·on wi
.
b
•1J'l
edi:1r1 scientific inquiry.
,o ncfttct,
archers seek and revere veridica l knowledge; they may even
feSS
. ced rese
~"perien . k of research as the search for truth, but they also understand
th
'.~ ' 'e and probabilistic nacure of scientific trud,. Know
. , ose co
ledge is
c111
.
e
hagu. '
.
h
51
,Jwelu " '
tarive decision about the world, always .held conrmgenr on t e
l
·ded
as
a
cen
•
rcg:11 f ·t _future.
ri:
O
~ ie
. of rhe researcher is st riving to und erstand . Correspondingly,
.
knowledge . T,rnth LSheld gently, how. •h va1ne
. accor d'mgIy. rt 1.s
;1l11g '
b ISexperienced investigator spea ks an cl .wntes
t e co lard a proposal wit· h reservanons,
·
·
·
l,vei,· and
·
provisos,
an d di sc Iaimers
necessary
·
·
f
h
l
· of
nor
... ems." Jt is necessary co wnce with .respect or t e comp .ex1ty
f as ir se
.
I '
su~,
d with modesty foe what can be accomplished . The researc 1er s
rl~mgs
/;x ecrarion for any study is a small but perceptib le shiic in the scale
hi_ghe~d
Most scientific inquiry deals not tn the heady stuff of truth,
of evi en ·
·
f
I b·t·
" . bl' hing actua l causes," but in ha1·d-won increments o pro
t tty.
eSC<~ IS
.
h_
d
'
.
'£'
. 1s
. the
·
d
s
r
'
gn
by
which
co
estunate
t
e
scu.
enc
s
sc1ent1
1c
maturity
I
A t1tr
·
.
.
.
.
.
.1blc
(and willingness} to exa mmc alcernac1ve mterpretat10ns of evidence,
11
'
~ble
r·,val
facts that hid co disconfirm
the theoretical framcp1ausi
' 1,y,,otheses
t
,
.
.
.
work , and considerat ions that reveal the l1m1t,~ttons of the methodology.
[r ·IS ,·n1portant not onlv, co lav• out the alternar.rves for the reader but also
co explain the grounds for choice among chem. The student who neither
acknowledges alternatives nor rationalizes choices simply does not understand research well enough to bother with a proposal.
The mature researcher feels no compulsion co provide perfect interpretations or to make unassailably correct choices. One does thl'. hcst one can
within the limits of existing knowledge and the present situation. The author
of a proposal is compelled, however, to make dearly rationalized choices
from among carefully defined alternatives; this is one reason readers omside
the scientific community find research reports teJious in their actenrion to
detail and explanation. It is the public quality of the researcher's reasoning
that makes a communiry of scientific enterprise possible, not the construction of a facade of uniform certainty and perfection.
Scudent-couducced research often contains choices that must be rationalized less by the sh,1pc of existing knowledge and the dictates of logic and
rnore by the homely facts of logistics: time, costs, skills achieved, and available facilities. The habit of publ ic clarity in describing and rationalizing
choices must begin there, with rhe way things are . An honest accounting of
hard and often imperfec t cho ices is a firm step fo r the stud ent toward achieving the habits of a good researcher-t he scientific state of mind.
,;0 nrenr
11I (;. bust ness
· placed o n hard-won
!
,a
90
Writfog tlic Proposal
Note
1. Sracistical significance, of course, is not synonymous w id1 scientific sig .
cance in terms of the evolution of knowledge, or practica l significance in tei:
111
solving professiona l prob lems. Statistical significance largely depends on sample/I
and selection of an ttlpha level (the level of confidence necessary t-o reject the
hypothesis). 1t can be demonsrrared berween alrnosr any two groups using alni11
any variable selected, if the sample size is large enoug h and the power of the test s::t
fi.ciently high. Such dHferen.ces berween groups may be statistica lly significant b •
sc.ientifically trivia l and professionally worthless . The pilot study is an cxcelfe~t
device by which the probab ility of a Type I error may be estimated and an aPPro~
priare san1ple size selected. ln rhis way, the investigator can increase the probability
rhat a sracisricaJJy significant result also will reflect a difference of sci.enrific and
practical significance.
:'fr.
°-'
~~I
s not synonymous with scientific .
.
dge, or p ractical significance in s,glli~
gm"£·
1cance largely depends 00 s terllls or
am~ ~~
t ence necessary ro re·e .
' f con f"d
J er the
en aJrnosr any two groups using a) 11~
1!
·ge enough and the power of th
llloit
c tesr
>ups may be starist ica/ly significa Su(.
·thlcss. The pilot study is a11 nr b~l
exceJJ
em
error may he:cstimared and a
. .
.
n appr
nves(lgaror can increase rl1c b
o,
·
.
pro abii'
II reflccr a difference of scienr·f· '!):
I IC an~
5
Preparation
of Proposals
for QualitativeResearch
Different Assumptions
The Only Constant Is Change
When this book first reached print (197tl}, the probability that any of our
readers would elect a qualitative study for their di~sertation or granr proposal was small. Only students in sociology or anthropology would have
heen likely to know that such an option even existed. In that year, with the
exception of the small number working in history or philosophy mosr graduate students and young scholars would have begun their apprenticeship
in research with studies cast in the familiar quantitative mode of natural
science.
Those studies would have presumed views of the world and the process
of inquiry that were then so pervasive in the disciplines of natural and social
sciem:e (and applied professional
fields such as education,
nursing,
and
social work) as to be callee.Isimply "the scientific method." It was an orderly,
understandable, and innocent rime. There was only one way to do good
research; one learned i.t, and then did it. Recause science is not a static sec of
prescriptions, however, the natural evolution of the enterprise was to produce some dramatic alterations in rhat familiar landscape.
91
92
Writing th~ Proposal
What changed was not the viability of the th en-dominant natural Scie
tradition. Experimental and qua~i-expffimenral d esigns (with all their asst, ti~~
tions abour the nature of truth and m11.ity)remain tl1echoice for many sc_~ip,
tific purposes. What changed was our growing und erstanding that quan.titat lmeasurement·, manipulative experirncnts, and the searc h for objective trt ith '"e
not the only way to do research-and certainly not the only means of S,,steil
te
atic investigation that deserve to be called scholarship.
Ill-
A reconsideration of assumptions about such fundamental things as the
nature ot real1cy,what constitutes knowledge\ and the role of human Values
in the process of resea rch led scho la rs co challenge the adequacy of sotne of
the established norms for inquiry . Such challenges led, in turn, to the devei.
opment of new stracegies fOl' formal inquiry in the social sciences. Th.ose
alternatives created bo th the necessity of an expanded curric ulum in research
training programs and new options for research proposa ls.
As a convenient :;ir:nplification, the alcernati ve way of thinking abour
research questions (and the new forms of inquiry that it produced) is referred
to in tbis text as "qualitative rcseard1." As an alternative paradigm, sonie
forms of qualirnrive research have had long his t0ries of use in parricu!ar
areas of social science (for example, cultural anthr opo logy) but until recenrJy
were not a significrnt part of mainstream scholnrship or resea rch trainfog in
orher disciplines or applied professional fields. 1 Ju the last two decades, how.
ever, contributions from qualitative research h:we burgeoned in chc literature of virtually every area of social science.
The same has been true of ptJblicario11sabour qualitative research. At the
end of the 1960s, only a handful of rdativcly obscure books and journal arti·
cles <lealing with qualitati ve research existed. Suddenly, a mountain of print
appeared conraining dis cuss ions of theor y, alternative designs for inquiry,
and debates about t.ed,ni cal appl ications and qualitative standards. !nc,,itably,
then, 4ualitative research . bas been a "wo.rk in prog ress." Full ot tc.,ty academic disputation and ex ploratory studies pushing the envelope of acceptable science, qualitative research is slowly bei ng ddined hy the uses of its
practicioners.
That evolution has been reflected in successive editions of Propo:;afsThat
Work. Certainly, our. own vision of what consr-itures a sound proposal for
qualitative study has changed. r-o,-example, readers with cll:cessto c;idier
versions of this drnpter would detect that we now have introduced attcnr.ion
to the particular problems of \.vri.ting proposals for focus-group research, a
format for qualitative inquiry th:i( we had previously elected co ignore. Also,
over time we have progressively altered our advice concerning the use of
mixed (qualitative and qw.1ntitati"\le) methods, the ne(;essity for including a
comprehensive review of the literature in the proposal document, and the
Preparation of Proposals for Qualitative Research
93
·.iftJ1e then-domi nant naturals,·
Cte"
·
.
nema I <Iestgns (w 1tb a ll their a
··~
.
.
•
~ ~h
Y) remam the ch oice for man,, .''!>
,
.
I
J SC(r>,,.
wmg Ul'.I<ersranJi ng tl1ar quanr itat~'"
and the search for objective truth 1"'t
tamfy cwt ~lieonl y means of syst/"c
!iCholarsh1p.
Ill,
•ur such fundani enta l thing s
- .
as tne
;dge, and the r.ole o f hum an v
a1Ue
chal!engc the ade qu acy o f sorne 8
1allenges led. in turn co che d Of
. .
,
,
evef.
UJrym the social scienc es . Tho
.1
d J
. I
Sc
expan e curn cu um in researc/i
,earcl1 proposals.
ternativ e wa y of chinking . b
.
, .
' 1 · 01Jt
tqu1ry tl1a c Jt produ ced) is referred
,s a n _alte~na tive paradigm, .~o,ne
ng hJStone s o f use in particular
I anthropology) but until recently
:hobrshrp or research l-r;:iiningin
1
ls. In the last two decades, how:h lrnve burgeoned in the Jirera-
:>omqualit,Hive researd,. At the
obscure hooks and journ::d arti. Suddenly, a mountain of prim
al(·cmative designs for inquiry,
1ualitative standards. fnevitably,
in progress." f-uJIof zesty acaishing the envelope of acccpt~ing Jefi.ned by d1e uses of its
;iveeditions of Proposals 1hat
1stitutcs a sound proposal for
readers witl1 access to earlier
ow have introduced attention
!s for focus-group research, a
iou sly elected to ignore. Also,
tdvic e concerning the use of
th e necessity for includino a
b
'.)n>posal document, anJ the
to address thr eats to va lidity . We make no apo logy for
wa'.)c 1.l anges in opinion
an d shifts in emp hasis . To th e contrary,
apP,optiare
d Othe1
·
• o~·e (art hat tbe experiences of o ur own scho la rship , as well
as what we
cJ,
.
e bcljevei:
from ou r students and co lleagues, have given us a better
l
.
I d .
. .
'"
co learn .
•0 11rinue .
f cbe particu la1· demands and pro b ems mvo ve m wntmg
"'
tand111
go
l
0 nders
proposals
for qua litative researc,.
Disagr cements an<l Diversity
· tlnt
eco<rn1ze
., some of you will co.me ro chis chapter un fami
f liar
h with {or
\Y/e r "b t) q ualitative research. For yo ur use (as we ll as or t e purpose
a. ou
I
d
·11
"d
unc·lear 1
1 · g a common point of reference for
. a I rea
. ers),
I w.e w1 ,prov1 e
0 f estab is un
. f intro
. . duction
to che qualitative paradigm. Ftrst, 1owever, we must
a b.rie
.
.
f h 1 h'
Because this type o inq uir y 1s re at1~e y ne': m some areas o sc oa rs 1p
and becau-se it is eve tywhere undergomg a smg ula r spurt of development
. .
' d d" ·sif ication, the field is anything but tidy . In consequence, we a nt1c1a.n 1ve1
.
. ..
.
.
·11b
.
h. t some of the ddm1t10ns employed m tb1s chapter w1 e unsat1s.
II
Th.
pare t a
ts
factory (if not outright heresy) to .some of _our academic co cagues.
book, however , was not written for established researchers who already
appreciate the subtl e distinctions in the field-and who already know how
ro prepare proposa ls.
.
.
.
Further, there is no agreement on a universal label for this kmd of
research. [n th e literamre of socia l science and applied professional fields,
such terms as interpretive, naturalistic, constructivist, ethnographic, and
fieldwork are vario usly emp .loyed to designate the broad col!ection of
appr oaches that we call simply qualitative r esearch . Some of those terms
reflect importan t cuscinctions in the minds of the people who empl oy them.
In contrast, we have selected the generic labe l "qua litative" as an arbitrary
2
coavenience. It is intended to be a working term for writing abo ut research
proposals and shou ld not be assigned any particular theoretical or ideological con notat ion .
csrablish severa l tmpofrrant ~av~ats.l . I
Also, despite the deceptive simplicity of the single term qualitative, you
will be confronte d at the outset by the need to choose from au1ong a numb er
of possible formats for inq uir y. Every research prop0sa l mu st reflect the
a uth or's selection from among a lternat ive approac hes to doin g researc h . If,
fo.r examp le, trad itiona l .qua11titative Lnquiry see ms most appropr iat e to a
research quesrion, th ere still rem ains the p ro blem of determinin g which design
offers the best fit. Will yo u emp loy an experiment, qu asi-experiment, descriptive sur vey, case study, or mixed methodology?
94
Writfog th e Proposal
Qualitative researchers must face a similar question. rf the broad assu
tions of the qualitative worldview seem appropriate (for reasons that typi~f·
mix the demands of the question with the investigator's practical intenrio ~
the availability of resources, prior training, and personal dispositions), th;s,
still remains the problem of deciding which of the various qualitative trad~c
1
tions will best serve the needs of the smdy. 3 Do you think it appropriate •
utilize the assumptions and methods that characterize ethnography, ground;:
theory, phenomenology, critical theory, or mixed methodology-or would
more generic qua litative format best serve your purposes?
a
Alrhougb true beginners may have the freedom co make sud1 decisions
more in theory than in practice (in graduate schoo .1s, for example, the availability of advisors often constrains the actual range of possibilities), they
nevertheless should be aware of the alternatives. The choice is important, the
different research traditions present very real advantages and limitations
and the decisions made in selecting from among them have profound conse'.
quences for the proposed study.
To survey some of che traditions under the generic umbrella of qnalitarivc
research, we suggest one or several of the following resources. They provide
not only the economy of an overview (often with emphasis on conrrasrs
across severa l perspectives and their characteristic strategies) but also a sense
of the tensions that attend the process by which scholars begin to stake our
territorial claims in a new enterprise.
In his textbook Research Design: Qualitative, Quantitative, and Mixed
Method Approaches (2nd ed.), Creswell (2003) focuses on the relative advantages (and demands) of qualitative and quantitative designs for study. In a
companion textbook (1998), however, he describes the conceprnal and operationa l consequences of se.lecting a particular qualitative approach . To illustrate, he examines the respective traditions of biography, phenomenology,
grounded theory, ethnography, and case study.
[n addition, edited collect ions such as The SAGE Handbook of
Qualitative Resea1'ch (3rd ed. ) (Denzin & Lincoln, 2005), and The
Handbook of Qualitative Research in Education (LeCompte, Millroy, &
Preissle, 1992), contain chapters that illuminate distinctions among various
qualirntive approaches . Several authors also have explored the major qualitative traditions for the purpose of developing useful taxonomies that clarify both differences and similarities. Among those are Jacob ( L987, 1988,
1989), and Thornton (1987). 4
Some o.f the textbooks that introduce qualitative research also give careful attention co helping novice researchers distinguish among the traditions
witltin the paradigm. Among chose are three that our students have found
particularly helpful. Frecbody (2003) devotes several chapters co sorting out
.Preparation of Prorosals for Qualitative Research
nilar question. If the bro d
.
.
. a a
,propnate (for reasons tha Ssullin
.
l.'typ· l<
e tnvcstigator's pracricaJ inte t~<llfy
g, and persona! dispos itio ,iitio~
ich of rhe various qua!itat~s > theix.
ivet
1 .
y.·' Do you rhink it approp . tadi,
ttare
.
.
,
.
11<1uH.:tenze
toethnogi:aphy .
.
> gcound
· mixed metl1odology-o
Cd'
r. W0l 1fd a
your purpose:;?
· freedom to ma ke such d ..
ecis
te scl100Js, for examp le t11 10ns
, e av ·1
:.:~ualrange of possibilities),
t1vcs.The choice is importa 1~
nt, the
rea l advantages and limitat·
10ns
.
1·
nong them have profo und c
rt.
onse.
w gct!cric umbrella of qualitative
lllow,~g resources. They Provide
ten. with
emphasis on conrrasrs
'
::-r}st1c
strategics) but also a sense
luch scholars begin ro stake our
ative, Quantitati11e, and Mixed
3) focuses on rhe relative advanltitative designs for study. [n a
·crihes the conceptual and operqualitative approach. To il!us)f biograpl1y, phenomcnolovv
f.
bl>
; The SAGE Handbook of
\c Lincoln, 2005), and The
ition (LeCompte, Millroy, &
tc distinctions arnong various
ave e~plored the major quaJi: useful taxonomies that clarhose are Jacob (]987, 1988,
·ative research also give careinguish among the traditions
har our students have found
:veral chapters ro sort ing out
95
. ,, O f qualitative inquiry, Morse and Richards (2002) use
· ·
· text, an <l
. •·c,1regor,es
·
d" as a cen.cral theme ·m orgarn7.mg
t l1e1r
1a1"r
1· .
"
,, ff. .
rnerho
.
11 _ inS a
. (2003) creac altern ative trat mons as genres, o ermg
· h
,.,e(r.?.,r I Ral115
k l ·
~oss111.111
. tine
vionetces ro suggest how graduate sn.idents m1g t ma e se ect1ons
,-,rive o
1'fh16f'
diem.
.
·
·
d.
r11nroong ·
idebook designed to assist people 111 the cask ot rea mg
fro \ s p,irt of ad.gu research reports, we deveIopcl I our own 1·
1st o f " types "
1
l
ersran
ing
·
O
d
d
··
d.
I
d ull' ' '
& Spirduso ) 20 4 .). fnten e tor a rea mg au d.1encef~
tlat
,Lil
s·1vennan,
-.
1
(I·0 ckc, b th reseal·cli consun1ers and entry-level researchers,
. . the
. text
. o rers
O
jnchrdcs
of five distinctive ap proaches to qual1rattve mqrnry. Each
•
l cl
I
brief· Jescripoons
d. ·ons is i_llust.racedwith the example o f a pub 1·
JS 1e stm y.
10 se n·a ,a
I
J ·
· I
1·
I
·
f
of r • u can see, qualitative
researc 1 1s not
'
. a smg
. · e, mono
. 1t 11cway o.
t\S yo . - I ·esearch. We have mac.le. this .a particular
pom.t
of emphasish
. , emp1r1ca .1 •
. .
. .
doing
b - ing an overview of the d1vcrs1tyw1rh111qualitative rcscarc
. se o ram
. .
· .
I,c,au
111
dmg you to hold your conmutmcnts
·11have th e saluta1-y effect of re111
wi
u·t you have a complete map of the. territory.
f
1· hrly-un .
,g
I ave d,e advantage here of havmg read the reports o some
ff you 1
d
d'
·
.
.. ·ve research it is virtually cerram that you have encountcre stu 1cs
qual1ran
.
. I
1· ·
· h. h
·
not be assigned to any par.ticu ar tral 1t1on wrt m t at pararhat conld
. .
I
.·
.
~.i·anvpublished reports, pcrhap~ even the maJOn ty, emp oy a geneJJc
d1gll1. IV. '
•
•
•
•
•
h d b I
·ich rhat clearly .is qualm1t1vcm its assumptions and met o s, · ut t 1at
appro,
·
.
. .
.
f ,,
.· .,
·. no.t ol)viouslv
,s
. the product of a smgk rrad1t1on.1 he concept o a genenc
~nalitativc approach was first forwarded by Merriam (2001, .2002). We
have pn;viously included ir among the types of research des~~1hcd 111 our
companion textbook (Locke et al., 2004), and will make use of 1t here as the
"plain vanilla" version of qualitative inqu iry.
Given such rich possibilities for inquiry, you might reasonably ask why
we have opted for a generic model in the introductton that follows. Aside
from the need to ;:ivoid overwhelming readers who are ju!;t starting to learn
about research and the obvious need to keep this text to a reasonable size,
the answer lies in our purpose.
All the forms of qualitative research with which we have experience have
at least some characteristics in common. Foe example, a concern for mainraining flexibility in the execution of the research design is virtually universal arnong the different traditions. We also have found it true that all
qualitative approaches make some demands that arc different from rhose
encountered in quantitative designs. A clear illustration is seen in rhe foct
that the relatively straightforward recipes used co confront investigator bias
i11 some quantitative studies (such as the double-blind experiment and analyses that identify investigator ef.fect) arc not possible iu qualitative research.
fn consequence, no matter which of the qualirnrive research traditions is
%
Writir:igthe Proposal
involved, the problem must be approached in a manner 'very different f
for example, wbat might he used in an experime nt .
tolli,
Observations like those have led us to conclu de tbat for the PU.tpo
helping novices make decisions about proposals , it is the similarities s~ 01
bind distinct traditions into a broader family that really matter. Our si,nt a\
dichotomous division of paradigms into quanti i-ative and qualitative Pie,
our use of a rudiment ary gencri1.:model to exemplify the latter, are pra;ttl'lcl
if inelegant, teaching stra tcgics for encoura ging you to focus first 011 the /al,
~t
t?
others the task of discer _aing and illustrating all
~ame~c~ls . ~e leave
tme d1srinct10ns that l,e both between and w1thm the broad paradigrns i
inquiry in the social and behavioral sciences.
Or
A Brief Desuiption
of Qualitative Research
What is q1.1alitative research, and how is it done? On first hearing, the
answer. seems disarmingly simple. It is a systemati .c, empirica l strategy for
answer ing questions about people in a particular social conte:x:t.s Given any
person, group, or lou1s for intera ction, .it is a means for describing and
attempting ro understand the observed regularities in what peop le do, or in
what tl1ey reporr as their expe rience .
For example, one of the most common purposes of qua lita t ive research is
served w,hen investigators pose the basic q uestion, "What's going o.n here?"
Venues for their q uestion might include a religious community, a hospital's
administrative sraff, a halfway house for paroled felons, a classroom, or a
school district in which textbook selection .has created controversy.
Alternatively, the experience of being a first-year social worker, an older
aduJt returning to a community col lege, a Litt le League coach, or a nu .rse in
a hospital's imensive care unir might be the focus for study. In each instance,
it is the total contexr r.hat creates what it means co be present, ro be a participant, to be a member, and co ha ve a role to p lay. It is the participant's
experience in that context that the researcher seeks to capnu·e and under stand in this kind of qualitative investi gation.
The mattet of definition is complicated, however, by the situation we
described in the preceding section . The q ualitative paradigm actually is a
collection of research traditions, each with its ow n prio1·ities, political agenda,
preferred means of data collection and ana lysis, and-unhappily
for the
beginner-technical
jargon. "What's going on here?" is just one of the many
kinds of questions that can be raised abouc people. Further, such quest ions
can be answered in ver y different ways when investigators start with difiei·ent assumptions abo ut what matters, and where and how to search . That is
PrP-parationof Proposals for Quahtative Research
ied in a manner very differe
:xperiment.
nr frolli
:o condudc that for the PUt
,
>roposa1s, it is th e simil . _Pos
e or
·i
arities
rruy that really
matter O
th.a,
··
· ur si
> quantitative .:ind quaJ1·tat· tllpJe
.
.
·
lVe
'
J exemplify the 1arrer ar
' arid
., .
• e Practj
,tg'.ng you to focu s lirsc on th cal,
f discerning and ilJustratin e fut\,
. .
g all h
·d w1th1n the broad parad·
t e
:es.
ig tns for
ive Research
s it done?
··
. On · first
· lie·anng
th
empirical strat · ' c e
. I
.
egy cor
tJCU ar SOCJa) context 5 c· .
.
·
1ven an
t is a means for describin . . y
.. .
· g and
u Ianues in what people CO,
:I Or 10
.
5Yst cmatic,
urposcs of qualitative research ..
,
"
·
IS
esri'.m, What's going on here?"
eligrous community, a hospital's
,Holed
felons ' a classroo
.
·
m, or a
non l1as created controversy,
st-year social worker, an ol<l~:r
ttlc League c<ndi
.·
' , oi· a nurse 111
:>eus for study. In each instance
~ans to be present, to be a pa/
~ to play. 1t is the participant's
~r seeks to capture and undcrhowcvcr, by the siruation we
litative paradigm actually is a
,wn priorities, political agenda,
lys,s, and~tmhappily
for the
f
here}"
·
·
1
5
•
· Jusr one o rhe many
:ople. Further, .~uch questions
investigators start with differre and how to search. Tliat is
for
rJ1 e differ ent appr oaches
,rl>
' cI1e casethe rubric
97
to inquiry that arc commonly
f
1· .
I
0 qua 1tat1ve researc 1.
, -c:i
der
oureJ un . . es uneasy and frequ ently fractious scholarly bedfeliows colr
i;
. n1et1rn
.
'fhCso . be qualitative urnbrella are bound together (albeit loosely) by
iec.:ccd
uJ1der tsuroptio ns about the sod a! world, the nature of social realities,
-c:vc1-nlkeY asquent natur e of inq uiry. lt is not always easv, however, for the
•
I~..
,,rid r 1c
discern exac tly wbar th ose shared assumptions may be. Part of the
i,cg.inller t~i es from the dist inction betw een theoretical rnodds and real life.
p1·oblel11at :he development of a quali tative paraJigm was stimulated by a
Bcc
·
'cl
·t· i · I
· · rcscarc h (t· h e fonn
.~iusef rhe assumptions
1 en.ti. 1ec wit 1 quantitative
f
. I posmv,sm
. . . ), 1t
. h as
.,1
·
c..:r
i
on
o
·
·
I
I
·1
I
r....
. shap ed by its roots 111 t 1e p 11 osop 1y o Iog1ca
111 1
of q_u,:~rnon co describ e qualitative research by noting how it differs from
bccnco
· s t 1mt purport to d1stmgu1s
· · · I1 t I1c I)C 1re
·f
Ider paradigm. L'1sts o f assertion
rhor O of investigators are use d to de fme
· quantmrnve
· · anoJ qua t·1tat1ve
· ways
5
wsrern
.
.d
.
1 .ing research.
. dersta11d111gthe wo.d -and c.0
of un ,
We have some grave doubt s, however, about the veracity of such dichotomous portrayals when app lied ro rea l people. Those reservations may pr.ove
ro be useful as you negotiate your proposal through networks that may
indLideadvisors, reviewers, or co-investigators, aII of whom will bring their
own assumptions to your document. To be direct about our advice, you
should not be disconcerted to discover that lives as lived rarely imitate
sdence as performed.
Our careers have aliowed us to become acquainted with a considerable
number of active research workers from a variety of disciplines and applied
fields. We can testify with confidence that it is difficult to find any investigator who will profess to all the beliefs that purportedly arc required of
an adherent to either qualitative or quantitative research. Put another way,
when it comes to worldvicws and personal philmophy, researchers are
like most other people. Not only do they display the usual wide range of
individual differences but they also seem perfectly comfortable with some
remarkable inconsistencies in their thinking.
Accordingly, our advice is to take those lists of various assumptions about
the world as pedagogical roois that can help you understand the qualitative
paradigm, but noc as portraits that accurately describe the beliefs of everyone you actually encounter on that (or any other) side of the paradigmatic
street! With that caveat in mind, we offer you the foilowing brief description
of assumptions commonly attributed to (if not perfectly shared by) all who
employ the qualitative paradigm.
Qualit:uivc researchers assume that thcrc arc aspects of reality that c.:1nnot be
quantified. More particularly, they believe it. is both possible and impor.tanc to
98
Writing the Proposal
discover am\ understand how people make sense of w hat happen s in their live
That includes as~ing research que.srions ah~ur the meani ng s people assign
particular expenenccs, as well as d1scovermg rhe processes th rough Whi.cJ
they achieve their intentions in particular co n.texts. lt also is assumed th:/
all persons cnnstrucr. thc.ir individual accounts of each evenr in which they
participanis. Those subjective consr.ructions are accepted as the reali.ries of
the social ~vorld. Thus, what is real is regarded as invariably multiple and
immutably relative to person and conrext .
t:
ar:
Given those assumptions, it is pre.-iimcd appropriate and effective to inquire
about sp~cificsocial processes or. parric ular persons' perspectives thro ugh dirccr
contact with those involved-ob.servin g, interact ing, and as king questions-in
natural contexts where people function. In doing so, it is accepted th at the
investigator mu.st he che primary instru ment fo r data collection, and thus par t
of rarher than separntc from wh~tever is invest igated. 1n tum, rhar requires
the assumption rhat the n:se:~rcher's own per.spectivcs and values ir1evirably
will become part of the rese:ncb process aud, ultimarcly, the fiu<lirigs a J
co.ndusioa.~.
11
Over tirne, the assum ptions laid out above ha ve led qualitative researchers
o f different designs for study, methods of data collection and analysis, and conventions for discou rse aboul .vork within each
tradition. Alrhough the result somet imes resembles an academic Towt.:r of
Babel, tho1>e designs, methods, and conventions do reflect (with varying
degrees of explicirness) the broad philosophic perspective assurned by the
overarching paradigm.
to generate a plethora
To il.lusr.rate the conceptual di.fficultics thar confroru· newcorucrs ro rite
research, we can point to the incorrect (though commonplace) assumption thar. there are such entities as "qualitative research
methods." We say "incorrect" because, abscn1· the underlying ;1ssumption.~,
th ere is nothing in the long lisi: of research procedures commonly employed
by qualitative resea rchers that could not be employed in a quantitative study.
Use of a method of data collection commonly employed in qualitative
research, however, does not make ,1 quantitative study one whit less quantitative in ir.s scientific nature-unless
rhac method i.s used (and the accumulated data subsequently imerpreted} in acrnrd with the assumptions of che
qualitative paradigm. Of course, the reverse ;1lso is true. The Curr.her possibility of actually mixing research modl'ls (panidigms) within the same study
is a topic we will address later in this d1;1pter.
wodd of qualitative
Having argued that no research methods are cxclu~ively "qualit,itive" in
nature, however, we have to admit chat sorne research tools ,ind conventions
are closely identified with qualitative inquiry. Tn that sense, it is fair to say
Preparation of Proposals for Qualitative Research
of what /i:~ppens in tJ . f'
1e1r tve
uut rhe mc,inino s people a .
.·
,.,
, ss,gn ts.
.rng the process es through wh· 0
;ellSt'
tt'
corucxts. fr als o is assumed
Hs ot iach event in which th
at
ey are
s :;ire accepted :~s rhe rea/ '1t·
. . .
.
.
' tei; Of
ai<led as rnv:ma bly multiple and
99
. regies for inquiry are characteristic of much, if not all, quali5t.L
i.~rsofl1e 1a Those characteristics include many of the following {though
IJO·
sear<::
l,
f11rive
re
dy it would be rare for alJ to be rep.resented).
one sru
irt :1nY
• cive researchers usually work inductively, trying ro genera te theories
l· QuaIitaIp them understan d ne,r
I · data. Tl11s
· 1s
· ·111 cont rast, for examp le, rot he
1
char ,e
· quant itanve
· · researc I1 ,n
· w I11c
· h IlYl?O theses usua 11
·n,encal tra d..
mon m
y
e"pen
nopriare and effective ro in .
,
quire
:rsons perspectives through di .
•
lCCt
·ac?ng, and aski ng questions-in
is accepted chat t.he
·or ~lata coll.ecrion, and thus part
resr1ga~ed. fo rum, that requires
n.pecr'.ves a(Jd values inevitably
i<l, ulurnarely, the findings and
?oing so, it
have
led qualitative researchers
, study, methods o( data collec-
a re set
2
a priori and then deductively tested with the co llected d ata.
In rnosc qua litari~e studies, r.hc ccnrra I problems are ro identif~ how people
· . e ·act with the ir world (whar. chcy do), and th.en co determme how they
111t l
•
x x:rience and und erstand that world : how they fee l, what d1ey believe, and
I
e
• structure am I re Iatt·ons h'1ps WLtlln
. I . some segment o f cl1e1r
.
how
they exp Iam
existence.
J. Interviews and various forms o( observation are the mosr. common means of
data collrction, though they arc somcrimes supplemented by the collection
oi dou.rrncnts.
4 _ Data most commonly
take rhe form of words (field notes, interview transcripts, diaries, etc.), although quantities, frequcncic:s, and graphic representations also can be used.
Jursc about wor.k within each
~ml,les an academic Tower of
~l-<Hls do r~flcct (with varying
ic perspective assumed by the
5. Ir. is common for reports of qualit,uivc research to contain detailed descriptions of parr.icipancs, as well as both the physical aud social structures of rhe
lt coufront
6. In many forms of qualirarive resean:h (though certainly not all}, rhe investigator collects data in the field-the place where rhe behaviors of interest natura IIy occur.
newcomers to the
:o the incorrect (though com1ritics as "qualitative research
t
the underlying
assumprions,
,cec.lures (:ommonly emplovcd
Jloyed in a quantitative sru,dy,
•nJy employed in qualitative
;e scudy one whit less yuantiod is used (and 1·he accumuwith the assu1t1pt.ions of the
30 is true. The funher possiligms) within rhe same study
exclusively "qualitative" in
:earch tools and conventions
1
rhat sense, it is fair to say
context within which rhe study cakes place.
7. Qualitative research designs frequently involve collection of dara from different sources (somer.imes by means of differenr merhods) within a setting
for the explicit purpose of cross-chei::king information, a pr.ocedure called
tri{(ngulation. Inspection of such dac.a secs and subsequent follow-up where
discrepancies appear make this a primary 111cansfor establishing the truthfulness of sources.
8. Jt is rare for a qualitative: researcher to intro<ltKC a deliberate intcrvcntio11 in
rhe field of .study. For the mosr part, inve.~tigators rry to he non-intrusive,
reduci11~ r.hc causes of parr.icipanr reactivity to chc smallest possible number.
The cxi::eptions here are partici.pacory action reseatch srudies in which the
investigator plays an active (though circumscribed) role.
9. le is common for qualitative researchers co have a primary interest m
identifying an<l undci:sr.anding rhe social processes by which particular end
results arc created, rather than simply describing rhe results themselves.
!00
lO.
1 l.
12.
13.
""
Writing the Proposal
Although a researcher may mah use of intervie w gu ides, systematic fo
for recording observ:c1tiou data, and even mater ial from responses to tilJn~
tionnaires, in the final arndy.;is r.he researcher is the primary instrumei{lles.
inquiry in qualitative research. Wich rare excepti ons, he or sh e must int t for
di reedy wirh study parr.icipar1ts, dctcnnining fro m moment to mo uenr ~<lct
1
to beha vc, what to no r.ice and record., and how a pai:ricu l ar line of in ~Iv
does or does nor offer promise for atl.'iwe.rir1gthe researc h question at hqlltry
,
and.
Qual.itarivc rcseMchers try to he conscious of th e perspective they brin
,1 study. For ch.at reason, they ofren exrlain
their own backgrou nd! to
. l .
' h
I
.
f h
parucu a r mteresc 111 t e rcse::irc l quc.stton as part o .. t e research repond
r,Zescarc I1e.r b.
h
·
I
·
d
1as, owever, 1n t 1e ser1sc ot a vcste pe rsona I ·Lnteresr in p rt.
ducing a pa rticu lar find ing, i~ regarded as a di fferent matt er . Bias muser:
conrrollcd if rhe resulrs of a s1udy arc to seem rr mhfu l. Accordingly, tacrj
for counrering rhe inclinat ion 10 see ~11dhear wh at is desired often are Ce:
rral ekmc:ms in qua.lit:u ive research designs.
lrrcsrccrive of the paraJigrn, parricipanr rcacrivir y to the investigator or to
the condirions of rhc study is a lht·eat r.o the: integri ty of rc.seacch. for a varj.
ety of reasons, however., this is a particularly sen sitive prob lem in qualita.
rive re:;cardi. A..::rnrdingly, many studies include tactics intended to limit
that source of d:~ra distortion.
Only rarely arc: .~amplcs of parricipanrs created by random prncedur.cs.
Selection is more likely to be purpo.~efoJ, with rhe i.nr.en1ionof maximizing
the utiliry of data for rhc research goals intended.
14.
Designs for qualirativc .qudies usua lly are carefu lly though t o ut during
a per iod of planning and preparnrion. In some instances, the p lan may be
specified iQ considerab le derai l in the form of an exce11sive written proposa l.
Neverc hc!css, absolute fidclit)' in execution of a particular design does nor
offer the s.ame benefit it yields in quantitative stud ies. Ins tead, it is common
in qualiratjve research for plans to be regarded as tentative and concingenr
on the realities p resented by data co llection and analysis. At least in the case
of experienced inves tigators, in-course adjusrmenrs are regarded as part of
do ing good researc;h ra rher rhan fata l breaches of protocol.
lS.
Qualirarive research .reporrs ofren :ire wrinen i.n rhe fi.rsr person and may
employ exp ressive lang u~ge intended to make finding.~ both accessible ::ind
powerfully per s uasive .
Those are eharacte ristics typ ical of studies that would belong under rhe
qualitative umbrella. A particular scholar might add or delete one or several,
o.r might modify som.e of otu· explanations, but we believe that when taken
toget11er,the 15 items collectively come close to a consensus mode! of what
ir means to do qualitative research.
Preparation of Proposals for Qualitative Research
se of interview guides
.
d even materia l from, sysre.n,atic
fo
respon.
rill
researcher is the primary . ses to n ~I\
inscr
·"~t1.
1 rare. exce:prions
he·
h
unic11 ~
.
.
, 0rs en,
. tr.
ermuung from moment t
list t11rec11
d
d
o molll r~c
r , an how a ·parricu la· 1·
Cnth 1
.
1 ine of.
o\\nswenng the research q
.
•11q1r
ucsr1onac L 'I.)
11a
•
11
isc101.1s
of the perspective I .
<l
t 1ey h .
.
I
n exp am their own b:i k
r111g
t
·
c grou d 1,
.iesr1on as pare of the
.
n a"d
research
·•
,e of a vesred persona / int
report
ed
d'ft
eresc in . .
as a ' rerenr marrec Bias . Pro.
. to seem rrurhful. Accordin I lllusrht
od heat whar is desired oft g Y, tacti~
~signs.
en are ce0•
nr reactivity lo th e i11v .
.
.
esr,garor o
1 th
r: iriregrit y of researcl1 p
rto
· or av .
cu Iar/y sensitive proble
·
~ri.
m 1n qua/
res ;tidu<le tactics intended
.. tta.
to lun;1,
ts created by random
,
I .I
.
Proce<lui·es
.' wit l the intention of maxim. . .
Jtltended,
tling
arc carefully thought· out d .
.
.
lll'Hlg
some rnstances the pl·'n
b
f
'
~ tllay C
-:Jo an extensive written proposal.
m_of a pan1cular design does not
t1ie stud,c.~. Instead, it is comrnori
arded as t
·
cnrat1ve and concingenr
t and analysis. A,: lease in rhe ~ase
I
usrments are regarded as parr of
~·he.,ot protocol.
·ten in tl1e first person and may
1kc fmdi11g.~
hoth accessih!e ~nd
hat would belong u11Jer the
add or delete one or several,
we believe that when taken
a consensus model of what
.
101
tha t assertion may be, however, you sho uld be aware
. ,ornforr•~g::ement about the exact location of boundaries for the
j\)rbcre is disa~f inquiry ca lled "critical theory research," for examp le, is
t
n:1'
Ji~111.
A cypernescholars co differ so sharply in its basic ass umptions as
· ·
di
I
r.irti I bYso
siderec
eparace and d1sr10ct para gm. In ot 1er words, whether a
; ro st udy is inside or outside the theoretica l boundary of
,o
proaco
.
.
.
rch depends on who 1s domg rhe looking.
(ic:1l :ip
1
c
. e resea
.
.
.
.
q'orJliWc•V earc h is well established in bot h the soc ial sciences and some
· ic-il
bl
·
·1
k
d d
·
c rir
' .res
0fessional study (nota y nu rsing, soc1a wor , an e ucat 1on) .
:irc11sof P'.shelcers a some times bewildering number 0£ its own pernmta1\JrJwugh is a core of characteristics that does give defin ition to a critica l
. • rhe1c
d th
k f .
.
,.,
n_on:., f both society an
e tas s o mqUlfy. oecause you are sm·e to
O
v,cw
. reports of critical studies, we will detour briefly here to suggest
,ounre1 .
,
k .
.
.
I
cl'
en 1 , urccs that will ma e It eas ter to app.reci.ate w 1ar you are rea mg.
·ev~ra_1::sr readers, Tho.mas (1992) will serve to introduce this complex and
fo\ ,es controversial approac h. For readings that offer mo.re detai l there is
smnenn
· on cnt1ca
· · I et1nograp1y
I
I or any o f severa I examma·
r n's (2005) creaase
i\.11.ic
isQ
1
·
d
·
cl
fi
I
I
f"
d
·
{C
&
K
emm1·s,
o~s • of critical theo ry as app
. .1e m 1e cc o e ucat1on arr
!986; Carspecken, 1996; G1.tlm, ~994)_-None of tl~ose make heavy dem ands
for background in the areas of social sc1e11c.:c
and plulosophy. There arc source
re;idingsthat go far deeper into critical theory, but most of our own students
felt more than sufficiently introduced when they had finished the chapters on
rhar subject in the handbook edited by Denzin and Lincoln (2005).
To really learn more about what constitutes the critical approach to
research requires more than reading. ft is our judgment that ac{;css to a mentor who has actually performed critical studies is a support for which books
can never fu lly substit ute. In accord wit h that ()pinion, except for the brief
comment that follows, we will nor attemp t here to address the rnyriad imp lications of critica l. theory for the product ion of a research proposal.
As with its paradigmat ic neighbors, critical theory includes a number
of traditions that a1:e only .loosely (and not a lways comfortably) related .
Participa tory, empowering, action, materialist, and feminist research perspectives are among those . Ac the most fundamental level, however, what
they share is an interest in and a concern for the ways that power is disrriburecland maintained in socia l settings-and how those arra ngeme nts can be
challenged.
cOJl0 risricurea
!'
On the surface, that kind of interest appears not to req uire a new sec of
assumptions about inquiry. As the concerns of critica l theory begin ro shape
the relationship between the .researcher and the researd1ed, however, they
begin to have important imp lications for both method and the investigato r 's
purposes in the study . In some forms o.f critica l theory, for examp le, research
102
Writing th~ Prnposal
becomes a vehicle for urging or facilitating the redistribution
of power
Whl't~
improving the life circumstances of the particip ants . At that point ,
the imagined line between politics and scholars hip begins to blur, You
be sure that we have moved into new and contr oversial terri tory. Our all
t~
2Q;l't
guide for rea<ling and understanding research re ports (Locke et al., O
describes several studies that serve to illustrate so me of the dilemmas c,eat 4)
L
. •
I approac hcs to mqmry.
.
.
Cd
uy
cntica
(t is the very nature of those dilemmas and co ntrovers ies, however, tha
will attract some i.n<lividualsat the outset of their research careers. Persoo ~
values and deep commitments concerning iustic e a nd equity may prese:
compelling reasons to explore critical theory as a mode of inquiry . .Never~
thdess, whatever you may discover about your own motives, it will be vita(
to remember thar a sound critical study must begin with a sound proposal
for becoming critical. Given that injunction, this brief introduction to critical theory will close wirh notice of one last variati on on the theme.
Some investigators who do what is called "fe minist" i-esearch also have
made claims to a separate paradigmatic status for their perspective-a
tradition rhat they regard as distinct from both cri tical theory and the larger
collectivity of qw~litativc research. Whether or not that proves to be a useful
perspective, we suggest that if you arc interested in the way gender enters
into research, you take a short excursion into the often lively literature on
that topic. Good places to begin woul<l be Lather (1991), or the col!tctions
edited by Eisner and Peshkin (1990), or Gitlin (1994). A somewhat more
cautionary text edited by Ribbens and Edwards (1997) emphasi,:cs the
considerable difficulties that have attended efforts to pursue a consistently
feminist viewpoint in qualitative research.
Preparing a ProposaJ for Qualitative Research
With regard to advice concerning proposals for qualitative research, we
wish to 1m1keonc point clear from the outset: Virtually all of what has been
said about function, deve lopme n t, writing style, organization, and format
for quantitative proposals will app .ly bere. The qualitative proposal is not
substantively a different kind of document . Our experience, however, has
convinced us that it does present a particular set of problems that will
demand your attcnrion-cither
because they constitute common sources of
difficulty for revicwers, or because they are matters better confronted in the
proposal than at the later point of preparing a report.
In drawing up the following lisr of 12 key points, we have made three
assumptions about your situation. first, we have presumed that your proposal
Pr~paration of Proposals for Qualitative Research
:I.ting the redistribution of p
. .
ower
<IQ~
te participants. At that poi
I scholarsl1ip begins to blu/ t, Whet~
.
, You
in d controversial territory. Out ca~
esearch reporrs (Locke er I o,~
a., 20o
ustra te some of the dilemmas . 4)
Cteate~
,as and controver sies howe
.
'
vet' th
:!t of their researc h careers. p .' ilt
·
· .
etsolllil
mg Jusuce and equity may
.
.
Prese
ieo1y as a mode of inquiry. Ne llt
ir your own moti ves, it ·will be "er,
must begin with a sound . . Vital
PtOposa/
·
h.
ion, t is brief intr od uction to : .
I· • ' '
Ct1t1,
.st vanauon on the theme.
died "feminist" research als{>h·
·. .
.
·
ave
tc status for their perspectiv
both critical theory and the la;:
er or not _thatproves to be a usefu~
tcrested
in the wa" gender· e
•
J
11ters
mro the often lively Jiternture on
: ~a~her ( J 991 }, or the colJections
Gitlin (1994}. A somcwh"r
E
.
·
" more
~dwards (1997} emphasizes the
:I efforts to pursue a consistently
tive Research
¾Is for qualitative research we
all of what has been
style, organization, and format
The q ualitative proposa l is not
Our experience, however, has
ular set of problems that wi ll
t: Virtually
constitute common sources of
1attcrs better confronted in the
a report.
y points, we have made three
,e presumed that your proposal
l 03
·ecr to review by someone who does not accept qualitative
•
•
~iiJI 11or enuine scbo larsb1p. When char 1s the case, th e proposal must
rescM'h as g defend the legitimacy of a paradigm - a task we thi_nk is better
undercak;coseasoned scholars . On the other hand, reviewers who simp ly are
rcserv~c~
orwich qualita tive research usually can be dealt with by patient
1111
tt11fiJn ~r and by providing a judiciously short selection of imroductory
lc:1nar
10ll
r!1'P
readingsd ve have assumed that at least one of the reviewers (for graduate
Sccon ',,vos
t commonly the committee chair ) will be familiar with the
denrs, t
sru. . d methods you propose, as we ll as the literature chat explicates
des.ign1an1·ces Without the support of such expertise, the burden of explain·! ose c ,o
.
.
11
ing, and persuadrng may be greater than can be born by a docuing,cerc,·fy
r of modest length.
me;bird~and finally, we have assumed thar .reviewers for yo ur proposal will
be looking for answers to a familiar sec of questions . For example, do you
ear co know the concepruaJ and methodological rurf? Does your plan
;:fiect careful th0ught? Do the parts of the proposa l fit together? ls there evi.dencethat you are fully aware of the prob lems to be overcome? Is the na tur e
. cl.,ecope
an
· of the study reasonably well matcbed to your skills and resources?
ff the five assumptions noted above arc met, the following items can be used
as a checkli~t and, when coupled with the generic advice about proposals
be
SLIb)
offered in the other chapters of this book, should help you deal successfully
wirh the problems that are particular to qualitative research.
In our experience, these are the proposal topics that most commonly
attract the attention of reviewers.
I. Why qualitative? Make absolutely ckar rhat a qualitative design is appropr.i·
ate to hoth the .stndy's general purpose (why you are doing the sr.udy) and ch.e
more specific research goals (such as formally stated research yucstions).Your
training and personal values are nor. irrelevant to this argument. In the end,
however, it is the match between the paradigm and the prohlcm that must
carry the day.
2. Plan flexibility. Present a plan describing what. you will do from the outset to the finish of your .~tudy. ff you are a novice, adhering closely to the
general specifications of that plan is prudent policy. Qualitative research
frequently involves some circumsrances, however, for which a degree of
anticipated flexibility also is wise. Some procedures muse be responsive
to what acmally happens during data collection, as well as to the nature
of the data chat begin to accumulate. Jf there arc such points within
your proposed plan, showing that you have antil:ipated the necessity of
selecting (or devising) alternative courses of acrion ahvays is reassuring
to reviewers.
l0,1
Writing tl1c Proposal
A delic.:ar.cbalance has ro be maintain ed in this aspect of a proposal
qualitative research. Present a careful plan and stick co it unless there for
compelling reaso11snot to do .;;o (dependence on "emergent design" is
expcrien..:e<linvesrigacors}. Give d ear indicarion, however, that You h 0 t
given careful thought co alternariv es-should they be required .
av~
;t~
3, Build a framework. Presenr a conceptua l framework that helps to exp( .
and clarify your proposed design, Define rhe main. consrrucrs and show th:1•11
relarionship to one another, to r.he research quesrions, to the methodo lo ic
am( r.o the related lirerncure. Absent a concrary regularion, rbjs should
take the form of an extended general review of the literature . Previous scho~'.
ar.ship is best limited ro assistance in defining the precise conceptua l territory of the proposed srudy. lhe lir.erature can provide conscrucr definitions
thcoretic.:al frameworks, examples of successful fcsearch strategies used i~
parallel c.:ir:cumstances,;ind a display of where your study would fit into
rhe ongoing conversarion among scholars . The primary emphasis, however
should be on the c.:oncepc.•,
and relationships assembled for your own stud/
A graphic forrna r often is useful for acl1ievingclariry here (see the excellent
advice an<l examples provi<lcd by MaxweU, 2005).
;Y,
4. Ar ticulate. the parts. Take special care at each step to write brief. !mt explicit
exp.la nations of how the parts fit together-purpose wirh qllesr.iou, question
wirh framework, framework with methods, and collected data with means
of analysis. lo the absence of the stru cture provided hy standardized designs,
it is easy for authors of qualirarive proposals to lose the sc11s1::
of cohesive
ur1iry among rhe several pares.
5. Plan ( 0 1' validity. Deal directly wirh the issue of validiry.6 H you complete
the proposed srudy, everyone who reads tl1e report will have: a perfect right
to ask "Why should 1 believe you?" Tf you wanr to be prcp,1red with a persuasive answer, the propo sal is the place ro search om the threats to validity inherent in your plans. Morse and Richards (2002) have an exeellenr
,;:hapr.cron convincing reader.s about the rigor and tn1srworchiness of your
srudy, an<l ;\faxwcll (2.005} goes even further co s11ggcsrth,1t qua]icatiw proposals should have a .scparare section devoted to answering r.he question,
"llow could 1 he wrong?" That is a tough question, but a healrhy one.
At rhe least, you muse dc,ll with the three threats ro validity char most
commonly ;ine.nd rhe procedures used in qualitative research. (a) How will
you ensure that descripriom of participants and context are :Kn1ratc ,rnd
complete? (b) Arc your per.,onal biases a threat? lf not, wh)' noc, or if so,
what do you plan ro do about them? (cj In what ways amt ro what degree
will parr.icipa1u reactions to you (and co the procedures used in rhe smdy)
impc-deacquisition of valid dara, and whar arc your pl/\ns for dealing with
rhat prnhlem? J\gain, experiences (and dat:l) cited from pilot work are powerful ways of showing that you arc prepared ro deal realistically with threats
to validity.
Preparation of Proposals for Qualitative Res~an:h
3imained in chis aspect of
ireful plan and stick to jr alp~<>Posa1
'
(<:Iependence on "em . , un ess tL••ete'()t
·I , . . .
.e1gem destg ,, . at
c Cal md1carion, however cl
n ts t t
1 ld
.
' lat Yo o,
ies-s iou they be required.
lt ha~~
iceprual framework that h I
h
epsroe
e me t .e main constru .crs and J Xpf~i~
·
s 1<hv h
.· res earc h (Juesaons,
to die method
t Cit
ot a contrary regulation rl . h ology
J
•
, llS S 0Uld
,
a review of th e liternrure P1·e. .
•101
. d .
. . v1ous I
efining tl1e precise concepr
101.
irature can provide construct
tetti.
of successful researcJ1 srrateg· rn1tio11s,
•es LISed.
ay f I
o w iere your study would fir . . in
holars . The primary em 1 . h
lllto
.
h
P 1as1s, owe
ions ips assembled for .
Ver,
. .
your own stud
y,
ac blevmg cJaricy here (see the
axwell, 2005).
excellent
l)f
m.
/c
dct.
. ..
ear each step to wr ire brief b
'ther
.
lit explicrt
..
-purpose with question que .
c tl10 d .~, an d collect ed dar
·' I Stl011
' a w,c l means
:ture provided hv, st andardizec
.
'
, f d es,gns
•roposals to lose the sense of coh .. '
esrve
:he issue o{ vali<litv 6 ff
,·
you complete
_ds the rcporr will have a perfect right
you want to be pn:p·,r~d w· J
.
' ~
ttl a pertee to search out the threats t ,. 1·d
:I R'·h·
.
, o \a 1 .
t~ ards (2002) have an excellent
he l'l "11or and t ru st wort h'mess of your
. .
'.lr.ther to suogcsr ti
" "' ' lat qua 1ttanve pro<levored ro answering tlie qucsnon,
.
:1ghquestion, bur a healthy o ne.
! t h ree threats to
.
· va 1I·c1·
ity t 1lat mmt
n yuahrative research. (a) How will
Janrs. an<l co ntext are accurate an<l
a threat? If not ' why not , 01 . I'f SO,
) In what W·\ys
'
'.1n d to w Jmt <legree
> the proce<lur.es used in the study)
hat are
your •Jlans
for deal·
.. h
.
'
'
• lllg Wit
!ta) cited from pilot wo1·(,
·1
, , re powred to deal realistically with threats
JOS
ExpJain exactly how you will maintain a paper trail.
,. records·
,or
if you propose ro use some form of category system for ana ly6. pore xiin1PIe,r'ipts how
· and w I1en wt·11you recor d thc exacr source of each
. O f cransc •
..
f/(111
sis
~ Where wi ll you document rev1s1ons as they become necessary?
,arego~Y· ow wilJ you record your speculations about the data, the particf
,. h
h
'kew1se, 1l
Lt
the study, or yoursel when suci1 t oug rs cannot be hand(ed as
ipanrs, d insertions into field notes or interview transcripts? W,e can conr,racke~:r such records will be essentia l when writing the report-many
firnikt r months lacer. The length and compl.e.xity of most qualitative studi ormat1on
. :tu th ere 1s
. no
w·ee. so aUy guarantee t h at yo u w1·11tose .1mporranr •111
·es
v1rru
1
,.
,ed regimen for recording it promptly and in adequate derail. We also
pr'Jlll
. . d .111d up 1·acate-ar separate locat 1.ons.
hat all recor ds be mamrame
urge C
De/llonstrate /~rocedm·es. Avoid_the ~in of n~~inalism. Because q~a~itacive
7
· research cradic1ons usually are nch with specialized nomenclature, ir 1seasy
t<> slip into the habit of using the names of comp lex operacions as though
chey were magic incantations. To say that you will produce "grounded
cheery," or employ "ana lysis through constant comparison," "tr iangu lation
of Jara soi1rccs," or a "peer debriefer," tells the reader little more than that
you know how to spell the wor<ls. Explaining why you will employ the operation, showing exactly how you will use it in the context of your design, and
giving citations for the literature sources you have consulted constit11tc a far
more persuasive presentation.
8. J)on't anticipate findings. Be careful ahout using language that might appear
to build your personal expectations about findings into the study procedures. l;or example, a research question such as, "How do student interns
deal with feelings of hostility toward supervisory staff?" presumes that such
affective .~tates will be experienced by the participants. Whether char
assumption is correct or incorrect, the question too easily translates into
interview questions that can cue participants as tn how they "ought" to feel.
Of course, you will have anticipated at least some aspects of what is going
on for your participants. In most studies, the conceptual framework itself
reflects, directly or indirectly, what the investigator suspects is going onor at least what he or .~he helieves is worthy of attention. Such expcctarions,
however, become hias (threats to validity) when they go unrecognized,
unmonitored, and unchecked. The proposal should he written in a manner
that is sensitive to such dangers.
9. Re explicit about relationships. Your proposal should demonstrate that the
nature of your relationship with participants has been thoughtfully planned,
and will be carefully monitored during the course of the study. What people
say to you and how they hehave in your company is conditioned in large part
by the nature of your relationship. Thus, what happens hetween you and
your participants will reflect how you present yourself and how the mutual
perception of roles is progressively defined by subsequent interactions.
J06
Writing the Proposal
Presenting youndf (an<l then acti ng) as an inre resred and respectfu l visi
- . I co IIe:iguc, a genuine
' fr'1en d. an d. companion,
· a nee dy suppJicator•
a protessrona
(common among doctoral students), a potenrial po litica l ally, an om1 iscj I.It
1
scholar (occasional with profess oria l ryp es), a tota Uy dispassio nate and obtlt
rive observer, or a warm and sympathetic listener, shoulcf be a decision that
made consciously find for <lcliberace pw:pose . Make no mistake; how you str1.1:
ture relationships with partici pants wi ll have an effect on what will be collecred
a.~ data. Ir is inevitable that those wh0 read your proposal will ask whether
those social imenic.:tions will serv e the purpose of comp leting a so und study,
10. Plan entry and exit. Think thr ough procedures for entry to and exit fr
0111
your research rnntext (both the sire and the bumaa relations h ips) and 111akc
those phns explicit in the prop osa l. Negotia ting conditions for your pres.
encc and dcpanurc crn be delic ace matters tha t have both ethica l and prac.
tical consequences. This aspect of your study is likely ro contai n problems
that would not be encountered ia a ryp ical qua n tirative stu d y- nor leasr of
whid1 are close person:11 relatio nships with some or all of the part icipants.
11. Treat tr,ms(er cautiously. /:le ca reful ro w r ite about the potential genera liz.
ability (applicatio11 to popula tions outs ide your study) of your conclusio ns
in ways that march the pro posed procedures f.or se lecting participan ts.
Absent rnndom s:impling, cla ims to acquisitio n of valid k nowledge aboat
other groups (either within the s tudy context or externa .lto it) almost always
arc inappropriate. Careful and th orough desc ript ion of th e context and participants can make it possible to cJiallengereaders of your report with the
later question, ''Why would the se conclusfons nor app ly in anotbe r context?" Thar, however, is not the same as atrernpr.ing: to generalize your findings to ariocher .~erring.
l2. Name your oum perspecti1ms.Eicher in the main body or in an appcridixro rhe proposal, include a brief sta tement highlighting those aspects of your
personal biogrnphy (work ex per iences, educatio n, mentors, sa lienr events)
that havr. shaped your persp ect ive o n d1e proposed srudy- ics questions,
parricipants, venue, and gene ral pui-pose. Reviewers at this stage, as well as
readers who wif! later co11sulr your report, have important reason to know
what baggage you bring: Lo d1e proposed study in the for m c:,frelevant
beliefs, values, concerns, corn mitmentS, and int ent ions. You wi ll be the prin1ary research instrument, and publ icly naming the ways you relate to your
study is a vital p:1rc of preparing for qualitat ive research.
New Territory: Proposals for Focus Group Researd1
\Xie have three reasons for inserting a special note here about proposing
the use of focus groups for data collection in qualitative or mixed method
studies. First, this is a research strategy that has been applied to social and
Preparation of Proposals for Qualitative Research
~~) as an interested and respectful ..
tend and companion, a needy s V1s1
tot
a porent ial politica l atly
upl>lie;i11'
t
)
.
, an ornnis . t
ypes, a totally d1spassiona•e
C1eo
. .
, •. and b· 1
1ettc listener should be a d . . 0 1~A
•
ec1s1on "''
.trpose. Ma ke no mistake· h0\1V that i1
'
Yous
1ave an effecr on what w'/1b
true.·
1
eco u
ec~ij
io read your prop osa l will ask
, Wheth
purpose of completing a sound
et
111
Study,
proce d ures for enrry ro a d .
n ex1r f
nd rhe Jiuman relationshins) I :roilJ
N
. .
,- ant ll1a'egot,artng con ditions for your ~c
atters that have bod1 etllica l and Pres.
ur. stud y is likely ro contain pi·o:, rac.
' PtcaJ quanritative stud y- nor
ellls
I.easr of
·h
W it some or a ll of rhe _
parttc
· ·1panrs.
o _write abo ut the potential gcner 1·
·s d
a ti. t e yo ur stud y) o.f yowr condu s1
·
d
~
0~
roce u res ,or selecting par cicipa .
·•.
, , nrs
cqLU
s1tto n of valid knowledge abou;
•nrexr o.r external ro it) almost alw
I d
· ·
ays
. 1 escn pn on of the contexr ar1d
I
d
~~
.nge .rea e.rs of your reporr wich the
tclusrons not app ly in another cons attemp ting to generalize your find-
1
rh~ ru.ain body or in an appendix
ir h1ghltghting rhos e as peccs of yo
d
.
ur
e ucatton, ment ors, salient evenrs)
the p~oposed study-it s questions,
·. Reviewers ar this srage, as we.I!as
·rt, have imp orranr reason to know
;ed stnd Y in the form of re levant
ind inrenrions. You will be the pt i:~m'.ng the ways you relate to your
itatJve resca!'ch.
;roup Research
::ial n~te l_1ereabout proposing
1 qualttat1ve or mixed method
has been applied to SO(;ialand
107
rch with increasing frequency in recent years. Th ere is a
. al resea
·11cons1'der using
. f:·ocus groups e1t
. I1er as an
• '[int chat )'O U w 1
1
71
51
bt 11
· n cechmques,
•
·
· Ie for
g PoS· her., data coll ecno
or as t he p nmary
veI11c
5r (0
,ncr co or
.
!IdJl
1,,quirY· rbis is a particu larly attrac tive format for ga thering informa tion
0nd• . sight in to participa nts' feelings, attit udes, an d pe1·ceptio ns aboLLt
s~c
thiir ,illo;s tr\c. Unlike indiv.id ual interviews, for example, a focus gro up pre " sclecte copnacural environment because part icipants appear to be .infl.uenc·
· li£e. Ln tur n, nI at sense o f
,c_·11rs a'dmore
. f[uenced by onI ers- Just
as uI ey are UJ
111
·11
· to che clata .
,g 'an . ·rv can lend a useful degree o f aut11onty
1rh
ennci.,
c
l
·
·
·
l
· k' d f b · us1
11 •• ·cl and finally, i:or t1e novice invest1gator t 1ere 1s a m o o V1o
l hib, ' tr the method chat allows brief acquaintance to obscure the tru ly
11CSS a OL
··spects of its app lication. lf_ you are rea dy to use focus group tec.hf'
dif t<.:LI1t "
.
with skill it can be a po wer ful tool. If you a re not ready it also can be
n1que
· 1 mto
.
.
. , that lures tI1e_unwary an d unpract1cec
wast .mg time,
or, worse,
1·
:i n ai
I
·
d.
f
d
·
·
. erroneous cone us1ons masquera rng as pro o un ms.1g11ts
.
uiroOur advice on th'1s topic
· can be d'd
·
L act:1·c an d exp 11·c1t:
r,;1v1011:
• Heginby reading some references that go beyond the hrief descriptions found
in srandard reseal'ch rexrs. We have found that Krueger and Casey {2000),
Morgan (1997) and Puchca and Potter (2004) are ideal for thar. purpose. In
addition, to begin sensitizing yourself co the mysteries oi focus group moJeration, you can do no better than to vi.sitthe world of marketi11grese:Hch in two
books by Grccnhaum (l99S, 2000). You also should inclu<le inspection of
some critiques and cautions about focus group nierhodology such as Kidd aml
Parshall (2000) and Webh and Kevern (2001). And, finally, be sure to read severa.lreports from your own discipline in which focus groups are employed in
a variety of different research <lesigns.
• Find a menror, whether a colleague or an academic advisor, who has had experience in leading focus groups as part of a research scudy. Talk with him or
her to c.xplore the aJvantages and limitations of the method, giving particular
attention ro the demands of skillful group leadership and the complexities of
analyzing transcriptions of group inr.craction.
• If rhe use of focus groups continues to be an attractive option, at the earliest
opportunity try your hand at both leading gronp sessions and working with
actual transcriptions of data. Some pilot srndy experience will reveal rather
quickly whcrlier preparation and practice will allow you ro become comfortable with rhc tricky nuances of focus group dynamics. l.ikcw.ise, when
confronted by the enormously messy realities of what people actually say in
corwersations, you will learn whether you really have the patience, penchant
for cigorous use of an analyr.ic system, and rite necessary eye for subtle regu1:uities wirhin convoluted text th;it arc required ro exrract useful meaning out
of focus group recordings.
108
Writing the Proposal
,. By the time you have taken those rhree prior steps, the preparation of an effet,
rive proposal will presenr no mysteries. Aside from the requisites for an
sound plan for research, the keys to making your proposal persuasive will b:.
(a) explaiuing ex,accly why focus group methodology is appropria te ro You;
purpose, and (b) dcmonstracing that you. actually have "been there and done
that" with a credible record of familiaril'y and facility with the technique,
New Territory: Proposals for Mixed Method Research
Jf. there is a growth area in qualitative research right now it ha~ to be
rhc swirl of .interest in mixed method designs, AnJ, although quantitative
researchers have loug made use of such tacr.ics as interviewing subjects to
obtain insight'., that supplement their primary analysis of numeric d;1ta, nod
qualitative researchers have for generations been noring frequency counts of
all sons of objects cliHi events a~ pa1·t o( rheir "thick description$," the rules
for such casual and mostly opportunistic blending of methods have changed,
In fact, the majm ch::mge is that there now actually are such rule~!
Confrrcnces, workshops, college courses, monographs, textbooks, and special journal issues, all devoted to the theory and practice of using mixed
merhod designs, 1nake it impossible to be anyt·hing other than conscious,
cautious, and deliberate in making such a choice. Being nicely sensitive to the
changing winds of investigatory fashion, our own graduate studems have
become increasingly eager to suggest that they might do a mi.xed merho<l dissertation. le is alrnosc certain that you will at least consider such a possibility for your own proposal.
It is not difficu It to trace some of the roots for the new status of mixed
methods, The tension:; rhat arrendcd the growth of inreresr in qualitarive
research produced a lively deba1e ahout its legitimacy and, ultimately, about
the compatibility of qucintitative and qualitative paradigms for inquiry,
Inevitably, rhat led to some uneasiness about studies that employed any
mcrlrnd ot d,1ta collection whid1 ordinarily was associated with rhc procedure~ of ,rnothcr paradigm, From that concern scholars moved on to d.ispurcs over the question of whether ir was possible for one investigator to
sinrnlr.aneously hold co.llflicting world-views about the rrnture of research,
For a tirne, the whole \opil: appeared ro be spirnLing into ch,10:-;,
In rhe natural cotll'Se of events in the r<:scarch community, however, some
level-headed and industriou!; people decided to define terms, parse the possibilities into ordnly categories, decide which debates were not worthy of
continued effort, and, in general, set the stage for researchers to get on with
their work. Thar effort achieved much more than just lowering rhe volume
of disputation. Once the possibilities for mixing methods were examined
Preparation of Proposab for Qualitative Research
or ~teps, the preparation of an eff,
Aside from the requisites £ e1:.
or "-n
ig your proposal persuasive Will >'
iethodo .logy is appropriate to , bc:
0
ictually have "been there and > lit
· with rhe tec.hniqu done
and ·fac1·11ty
e.
Method Research
esearch righc now it has to be
ns.
And,. although ouantit
.
.
1
ative
r1csas tnterviewing subjects
to
. ,
' ana Iys,s ot numeric data , all d
een noti11g freq uency counts 0 f
"It i .,cJ< descriptions,,. the ru les
di11g of methods l1ave changed .
now actually are such rut es.1
nographs, textbooks, and speand practice of using mixed
iything other than conscious
:e. Being nicely sensitive to
own graduate students have
might do a mixed method dis1.eastconsi<ler sud1 a possibil-
th;
; for the new status of mixed
,vth of interest in qualitative
timacy an<l, ultimately, ahout
ttivc paradigms for inquirv.
: studies that employed
ts associate<l with the proc:1 scholars moved on to dissihle for one investigator to
)out the nar.ure of researd1.
:1linginto c.:haos.
community, however, some
de.fine terms., parse the pos:lcbates were not worthy of
•r researchers to get on \~ith
111 just lowering ~he volume
tg methods were examined
adv
109
·e closely, it was. irn_°:ediately clear that casua l use had not allowed
•• to discover s1gmficanr advantages that could be achieved by delibsch01t'i; e)ecring particular formats for combining methods .
e~re
cou nt we could ident ify no fewer than 23 distinctive designs for
A_cng
use of mixed methods . Not one of those plans, however, is defined by
r,1ak
i rticular methods being mixed. Instead, the designs are defo1ed by how
rhe P~,mu:es of methods are distributed, ordered, executed, and utilized in th e
cher1.11 .
.
•
·equent data analysis. Thus, although some members of the 23 designs
5
b
· w1'cl1111
· a c Iuster o f rel ared stra teg ies, you can
su · more than c Iose cousms
-,re n0
.
j e su.re that in preparmg your proposal there are de arly defined mixed:1echod choices to be made, an d that making the right cho ice can have posirive consequences for a study .
ft is gene rall y accepted that alJ methods have their limitati ons as well as
their strengths. Thus, the idea is co mix them so that the strengths are complementary: Such hlend: c_an produce a conver~~nc: of evidence that reinforces findings, can chmrnate or. at least m1111m1zcotherwise plausible
altcrriativc.:sto your conclusions, or can enrich your conclusions hy revealin"o
a
divergent aspects that would otherwise be invisible.
We.:will not undertake even a survey of the possibilities for mixed method
designs. What we can offer is to identify what we think are the best sources
(most understandable and most economical of time) from which to extract a
sound introduction. Ignoring journal articles and limiting the list to books and
monographs, these are commonly available and generally reliable sources.
,11°
1
~:st
• Creswell, J. W., & 1'lano-Clark, V. L (2006}. Vesignin1; and wnductinK mixed
methods resemch. Thousand Oaks, CA: Sage. (Creswell has a well-descr.ved
rcpnt;:ition. for constructing step-by-step guides for the decisions that must be
made in designing studies. The hook is oriented to the need.~of graduate .~rudenrs,
and foeusc:son prnducing a proposal that envisions a manageable mixed method
project for the beginning researcher. If you are entirely new to the idea of rnixing
methodologies, this hook offers three chapters that will be particularly helpful:
Chapter 1, "Understanding Mi1<cdivlethods Re.~earch: Purpose and Organizatiou"; Chapter 3, "Locating and Reviewing Mixed Methods Studies"; and
Chapter 9, "Questions Often Raised Abour Mixed Methods Research.")
• Tashakkori, A., & Teddfie, c·.(Eds.). (2003)~ Handbook of 1nixedme~hods in
socialand beha11iomlresearch.Thousand Oaks, CA: Sage. (There are 26 chapters
here with varying levels of mrnsp;:ircncy and usefulness. We suggest starting
with those by 'l'eddlie and Tashakkori [Chapter 1], Green and Carncclli
/Chapter J), and Morse !Chapter 7]. As your interests an<l needs may dictate,
there are more specialized chapters that offer discussions of ropics ranging
from compttteri1.ed analysis of mixed methods research, through m.cthods for
teaching ahout mixed methods, to writing ,·eports of mixed method studies.)
110
Writin g the Proposal
• Thomas, R. M. (2003). Rlending qualitative and quantitative researcJ.
1
111 I
ods iri theses mui dissertations. Thousand Oaks, CA: Corwin Press. (\'(,'tiet 1•
in trans parent prose:, and wirhour any scholarly pretensions, tllis book ten
directly with what the title proposes. Rich wirh illustrative examples fr:Cals
actual proposals that employ mixed methods, and built around a simple ll\oOrti
of five kinds of research purposes, rhis book will get you started. lf norh~el
else, the first chapter will cake you further into the subject in 13 pages r~~g
11
many other sources can manage in LOO.
)
• Reichardt, C. S., &. Rallis, S. F. (Eds.). ('1994). The qualitative-q11nntita
t·
Ille
debate. San Francisco:Jossey-Bass. (To make sense of where the mixed rncthOd
movement bcgm1, yoLJvvill need at least a little hisrory. Here, in a small PaPcrba,k from the publisher's series on New Directions for Program Evaluation, the
uHHribLJlOr
i; have captured the t:.-sence of the controversy. Remember, thoSc
who do nor read history are dt:.,rincd to repeat it!)
• Green, J.C., & C:iracelli,V. J. (Eds.). ( L997). Advances in mix ed method evaf.
11atio11:The chaUenges a11d benefits of integrating diverse paradigms. San
Francisco: Jossey-13ass.(From the same series as the item above, this collection
offers one of che first clear descr,iprionsof the benefits to be derived from careful construction of mixed method designs. The fact rhar the authors are primarily inrcresrt::din .research as ::i tool for evaluation is nor an impediment to
the utiliry oi rhi., $01m::c .)
In dosing this brief section on mixed methods we want to make clear chat
we ar e not unalloyed fans of combining approaches to research. For example, folding in a frequenc y cotu1t of something is hardly likely co produce
severe stress in a qualitative srudy. But assuming that rhe investigator can
easily swing back and forrh between rhc world views that are characteristic
of positivistic and naturalistic sde nce (as in so-called "sequential mixedmodel" designs) is a far more troublesome requirem ent. As Datta (1994) has
succinctly pointed our, " mixed-up" methods aod models (those that are used
incor rectly, inconsistently, or without reference co any kind of consistent
philosophical orie.nracion) are not to be confused with mixed methodology!
While rhe mi.xing of paradigmatic models (as distinct from methods)
clearly is feasible, we Still think it a less than prudent oprion for the average
graduate student. He or she must have sufficient flexibility of mind co move
back and forrb between two vantage points when contemp lating the data.
That is asking a great deal of anyone who is not yet perfectly confident of
his o.r her own grasp of research technique.
Resources for Qualitative Research
Our recommendat ions for resources that will assist in preparing a CJLU
lita·
civc proposal have heen divided into 12 r.opica [ areas. We 1..:,rntion
you nor co
Preparation of Proposals for Qualitative Research
litative and quantitat·.
,
· tile researct
rand Oaks, CA· Corw;., P
, '11e,,
, h
.
u, ress. ('>;,. ,,,
s~ olar~y P~etcnsions, this book tittcq
Rich wnh illuscrari,ve exam l dcij{~
thods, and built around
. p es ftoru
a sunpJe
,,,
; book will gee you started If n1odcf
.'
h
. llOtL'HQ
·
:ti
. ie1 into t c subject io 13
tL g
Pages
)
.
~
l. ( 1994 ), The qua/itative-q"
.
,
..aht1t.
nak c sense of where the · . d Qtf11e
'
ffiL'(C me L
a .Imle. history · Herc
;" a smafl t,1oq
· , u,
)11,ect10ns
(or Program £ al . Paper.
f tJ
~ l(tlt1011 L
o ie conrroversy. Remember ' t,,e
·epear it!)
' rhos~
97). Advm1ces in mixed meth d
·
· diverse pay"d: 0 · e11a/
mteg,·atmg
·
.
, .. ign,is s
r1esas the item above, this coll;c/"
the benefits ro be derived fro
ton
. Th f.
m care.
,. e act char the authors are .
. .
pr,.
Cvaluat1on
is not an imped'
iment to
lods we want to make clear that
•roaches
to resc·irc!1
For exam.
•
•
< ·
,
1t1g
. ts haedly likely
, to proJ u~
!llltng that the investigator can
·Id views that are characteristic
, ~o-calleJ "sequential mixed1u1rement. As Dntra ( 1994 ) has
nd models (those that arc used
}CC
t0 .
I. d
.
any <m ot consistent
seJ with mixed methodology!
' (as distinct from method~)
niden~ <~~tion for the average
u flex1bd1ty of mind to move
rhen contemplating the dara.
tot yet perfectly confident of
,sist in preparing a qualitilrcas. We caution you not to
3
l 11
ex.haust ive set designed to include all the resou rces that are
11r rbcse as ~·cacive research. Our purpose here is more limited . We have
rte
to qua t .
I I
d d.
1
. l
•
•
1cv:i11r k and joucna s t 1at atten
11:ecc
y to top ,.cs rnat novice mves,~
ed b OO. · sipal ly our own stu dems) typ 1·ca 11y I1ave clOUn d. re levant to the
rtfcct"
:ig-1
r<>rs(prt:eacing
a study design and writing the proposal. The order of
r,vosreps ~ with d1e needs of a novice ac the earliest stage, and then adds
. begins
.
d.
I
Id
1op1CS
:al oreIer chose invesrments 111 rea ing t 1at wou accompany preparain se1if,
ualirative proposal.
tiOll O
aq
User-friendly Introductions-The firsr step in considering use of a
1
·. ·ve appr.oach is co gain some sense of what it can and cannot accomqu:iltraria research strategy, and what sort of ski lls arc required to design
lish as •
.
f
.
P d'1 .. collect and analyze data, and write reports. Our person .al avontes
sru :; purpose are Bogdan ~nd Bik len (2003), Glesne (2006), Lofland,
Anderson, & Lofl.and (2005), and Merriam (2001}, all four of which
Sno 'r •1e virtue cJf care f uI rev1s1ons
·.
. use .
af rer Iong peno. ds o.f extensive
I,ave L
Delamont (2001) and Rossman and Rallis (2003) are equa lly competent as
introductions, however, and as they have distinctive styles they may better
fir your tasre in texrbooks. Finaily, Creswell (2003) is widely used in introJuctory research courses because its juxtaposition of quantitative, qualitarivc, and mixed models for research helps :;tudents begin to grasp the range
of formats that arc available for inquiry.
for proposals invo.lving smaller scale qualitative studies, Denscombc's
The Good Research Guide for Small Scale Research Projects (2003) is aptly
named and thoroughly practical. For case study research, Stake ( 1995) offers
an eminently readable survey of its many permutations, and the several texts
from Yin (2002a, 2002h) are widely used by students who are contemplating use of the case format.
for:
Finally, even at the introductory
level you will encounter
a good deal
of unfamiliar vocabulary, muc h of wbich wiJJ not be found in a standa rd
dictionary. Fortunately, a dictionary devoted exclusively to the .language
convenr ions (a kindl y label for jargon) o.f qual itative inquiry .is avai lab l.e
(Schwandt, 2001). With suggested readings and cross-references given i.n
most entries, th.is is an instance for which cover-to-cover reading of a d ictionru'Ymight represent an eminently practical strategy.
2. Specfrnens-Almost
any modern research journal in the social
sciences Ot' app lied profess ional fields w.ill conta in qualitative studies. (t
wiU be wise to locate journals that publish such reports in your own area
of interest, and we urge you to do so early in th e process of explori 1~
the quaUtative opt ion. On ly concrete examples can make theoretica l discussions come alive . If you need a convenient starring p lace, both Qualitative
l l2
Writing the Proposal
Sociology and Qualitative Health Research offer many reports that
relatively brief, inrrinsicalJy interesting for rnost readers, and quite Un~.r~
rnanding of technical background.
( e,
Many areas of study also have collections of qua litative research i·epo
in book formar. Excellent examples include Mer dam's Qualitative Resear:
in I'1'actice (2002), Milinki's Cases in Qualitat ive Research ( 1999), t'i n•
Cc,se Study Antho logy (2004), and Riessman's Qua litative Studies in So,; S·
Worf?Rese,1rch (1994) . If you wish ro read repo rts that are further our 41
0 11
the cutting edge of evolving methodology for qualitative study, the co.llectio
edited by Denzin and Lincoln (2002) will take you there.
11
3. Theo,-y-Books <lealing with the theoretical foundations of qua litative
research arc notoriously difficult-and
for the unin itiated, ponderously dulJ.
Sooner or later, however, you will have IO make a start. Many beginners find
it be.st temporarily ro bypass texrs that deal with the epistemological roo ts of
th.c paradigm, and stare instead by reading material that dea ls with how to
do research tliat is fait hful to those philosophica l origins.
The standard in the field for rhat ptu·pose is l inc.:olnand Guba (] 985),
a book that is consulte d, cited, quoted, and, in mo st cases, owned by vinu.
aJJy all who are ac.:ti.vein qualitative research. A treatment of foundationa l
theory that will be even more accessible for ma ny readers is offered by
Patton (2001 ). This is a textbook on theory and methods for researc h and
evaluation that is just ifiably famous for the author's light rouch. Another
gentle way to case youn ,elf into the literature deal ing with theoretical foundations is to consult tl1e book by Creswe ll (1998), in which the autl1or illustrates how each of five different qualitative researc.:htrn<litions shapes the
nature of study design. If you become serious about' doing a qualitative
study, there will be many more challenging theoretical mountains to climb.
for the present, however, this is enough.
4 . About Qualita tive Proposals-ff you have gone thi.s far, you prohably
a re going to write a proposal for qualitative research. Presently two textbooks are subs tantia lly devoted co that topic, and in this case our suggestion
is that yo u p urchase and llSe both of them. Marshall and Rossman (2006)
and i'v(axwell (2005) arc both ideal for the novice in qualitative research.
With contrasting styles and cmpliases, they form a pcrfec.:tcomplementary
pair.
· If you are preparing for a dissertation, the next step would be to read
Piantanida an d Garman (1999), who offer a four-chapter treatment of the
proposa l process . Richly illustrated with real-life examples of how graduate
students st ruggle wirh tha t task, this is a tour through all of die notorious
tough spots in gaining approva I for a qualitative study. [inally, available 011
Preparation uf Proposals for Qualit,itiv~ Research
'.esearch offer many
c
•
reports th
1g .1.0.r most readers and
. ar a
'
qUJte
.
't
ll11~~
-
~cttons of ~ualitative resea1·ch
.lude Mertiam's Qualitative R.rePorti
. Qu~iitative Research (1999 eseq,-c
/,
11
.~s'..:in'~ Qualitati ve Studies ;~'/i~·s
read reports tliar are furthe
Octa/
:y for qualitative study the t' °llt . 011
>
coe,·
11cttoll
,·11ta I(e you there.
hcorcticaJ found:ui ons of
.
h
qua 1ttar
>rt e l.!ninitiated, ponderous]y l\te
> make a start. Many beg·
duJj
,
.
innel's f
·al w1tl1 the cpi.srcmolog1·cI
IJ)d
.
a roots f'
•1gmntenaJ that deals w1
·r11 I
o
.
.
. .
•
10w
soph1cal ongms.
.to
pose is Lincoln and Guba (1985
most cases owned by . ),
'
virtu.
ire Il. A treatment of foundati
.f
onaf
. or many readers is offered b
•ry and methods for research a
he auth?r's ligl1t toucl1. Anoth:r
ire dealing with theoretical f
'1998). ·
oun. · ' , in which rhe aurhor illus,~ research traditions shapes the
nou.~ about doing ·1 "u· 1· .
'
-i
c1 lfattvr
d1eoretical rnountaim; to dimb.
1d, rn
l
lave ~one this for, you probably
e rcseard1. Presently two textand in tht's··,--1 t'.
'-< S our suggestion
Mar_shal! and Rossman (2006)
~iovicc ID gualital'ive research.
·orm a perfect cornplementary
e next step would be to read
~our-chapter treatment of the
ife examples of l10w graduate
tllrough all of the notorious
,e st udy. Finally, available on
113
·//WWW
.ssrc.org), the Social ScienceResea rcb Counci.lprovi.des
~h_crp~ide
for authors, The Art of Writing Proposals (Przeworski
ii>coP>
' of ,rs ~ 995). Prepared by veteran reviewers, the advice offered is
~ 5:ilc>n100 ~accicaland applicable to proposals for either qua litative or
,wriclYP
h
,rri! . ·ve ,:esearc. ·
'" 0 bsicC
111rrtatr
. in Qualitative Research__:Wc placed ethics here in the topical
th,cs
•
£
· dea I'rng wit. h met110 do [ogy liccause we t I1m
. k
.>,
f e the categories
order
!,el1ere
or 1·t belongs. Thinking through the design for a study should be
. -15 w
rh16 . terms of the ethical consequences that might attend each decision.
fC(ll'l,ed
tfl requires acquainting yourself with such problems early in the
1° do _1~\rocess . As we noted in Chapter 2, the topic too often receives
proposl,·Etin research train ing at all levels and, sad to say, that continues to
horr sin
.
.
.
s . . of qualiracive research. Aside from a passmg encounter with ah urn,m
be,.·rrue
,
I
b .
.
I
h
I
.
crs review prococo, most eginners give scant t 1oug t lot ,e quest10n
suvJ
e they wish _to treat t heu·
· parttci
· ·p,~nts-u_nt1.
·1t I1ey wa ll< hca dlong mto
·
of how
Hieof the nasty dilemmas that abound in qualmn1ve research.
( That ir may be necessary to consider the topics of participant anonymity
.ind confidentiality is fairly obvious. But how many novices wouJJ anticipate
the need to deal with situations in which it is the participant who makes
.:inonymityimpossible? Likewi~e, it is one thing to plan for development
of attentive and sympathetic listening skills vvhen interviewing. It is quite
another thing, however, to anticipate die need to handle interview sicuarions
in which the participant discloses sensitive and pot:enti;:illydangerous information. Even the seemingly simple question of determining when a participant is free to withdraw from a study can be more complicated th,m it might
seem.Decisions made in the relative calm of preparing a proposal are almost
always better than those made in the iiel<lwhen the right and wrong of
things is so easily obsrnred by panic
You certainly can begin your preparation for designing an ethically
responsible study by reviewing our introduction to that topic. From. there,
however, you will need to expand to resources that deal more particularly
with the wide range of ethical dilemmas encountered in applied social
research (Kirnmel, 1988) and the processes by whid1 your institution will
enfon:e ethical standards when reviewing your proposal (Sieber, I 992).
Ultimately, of course, you wil l have ro consulr readings that are more
directly focused on qualitative inquiry. One way to begin that process would
be by simply surveying the foll range of ethical problems rhat can arise in the
conducr o.f qualitative studies . For tbat purpose, a recent text by Mauthner,
Birch,Jessop, and Miller (2002) will serve admirably. As a more traditional
alternative, however., you might consult rhe c.:bapters devoted to ethical
qt1,
114
Writing t ht Propo:;:d
qucsrions in.foundational sources such a:,;Eisner and Peshl<in ( l 990), l)c, ..
·lll 111
and Uncoln (2005) , or LeComp te ct al. (E/92).
If they provide a closer fit with your interests, more focused treatni
of ethics arc av:-1ifa ble for qua litarive studies int.he field ofc ducati on (SinicntN
& Usher, 2000) and for designs that fall under the broad rubric of crir~ns
ethnography {Madison, 200.5). Finally, in a sma ll paperback prepatcai
explicitly for undergrndu ates and bcgi11ning n.:se~rchcrs, Caro l Bailey ~:d
elegantly underscor ed the ub1qu1rous nature of ~rh1cal concerns when inv $
rigntors intrud e into peop le's lives. Ea.::hsucc;~ssivechapt er in her Guide~T-ii'IJResearch (1.995} comdins a cliscu!ision of the sticky ethical problerrio
8
th~r can entrap tbe nnw,1ry. [fan hon::st ,~pprnisal of your background i
rese:1rch suggesrs that a book for novice:; wo uld be appropria te, tl1is ori:
mighr be a ,ound i1wes1rneni.
6. Nlethods-- ·rak en in thP.genP.ric sense, method s arc the coo ls for doing
resear1.:h.They include chl:.prcm::dmes and instruments used by the investiga.
tor w generate data, as well as the ced u1iques use<lto ana lyze data. Although
we ordi11:-1rilyrhink of interviewing (listening and conversing), observing
(watching people ), an<l dornment analysis (reading) as the primary means for
collecting data in qualir.ativc swd ies, a quick s11rvey of published reports will
reveal that there are rrurny or.hers. Q uestionnaires, ~urv1.:ys,systematic observation instrumenrs, L1t.t0btrusive 111e
asun.:c.,videotapes, and photo graphs also
serve as data sources. Likewise, <lo:teusof metho di, are avaifoble when organfaing dat a for the purpose ol <1nalysis
. ,t\;o single sourc e can cover all niethodc., so it is necessary to narrow any sean .:h co what cau be found in more
specialized rcns . We haV!.
: pro vided sugg<.
:stions below for such sources in tl,e
uoH.:.s
. com pmer managcmenr. of data,
broad ropic areas of inrervicwing, Cielc.l
nnd analy ;i:;;.
A useful first step in retrieving sources that exp lain parti cular research
method s is to use a dictionary or glossa ry of qualitative terminology co look
up the synonyms, definitions , and standard references associated with the
method about which you want information. Again, we suggest Schwandt
(2001 ) as particularly helpful for that purpose. A second seep would be to
consu lt the index to any of the introdu ctory-level textbooks recommended
above in order co c.rack down citati ons for method- related articles and
book s.
lf you think it would be helpful to browse articles that treat different aspecrs
of qualitacive methodology , the collection edited by iVlichael Huberman and
the late Matthew Miles (2002) is a sound and generally accessible place to
begin. Fina Uy,we suggest that you take a fewminu tes to scan recent cata logue
listings from publishers that off-erbooks ond monographs dealing with social
G
p
t(
re
cl
ht
St
nr
(2,
me
tht
or
an,
olc
Preparation of Proposals for Qualitative Research
'S
Eisner anJ Pes hkin ( 199
OJ,De,
>..
1
(l992.).
"~
..
mterests, more focused
treat1
.
..
· ws 1r1 the tidd of educ·'t
.·
'l1eh,_
« ion (S· .,,,
11under
the bro ad "'lb .
i111on
.
... ric of . ''\
Ct!tiC;if
, m a .StnalJ paperback
·
·
Pre
nmg researchers Caro l B . P<1te,1
aile,, b"
.
'
ur.e o f e:hical co ncerns When .' .i\\
StKcessive chapt er in her G,/n-vC/,.
;ion of the stick y et/'1·cal
tde l11
.
.
•
l
< P.rob
1
r appraisal ot Your backg .
Crtis
,.
.
tOur1d .
.s would be appr opriate l . •n
j.
.
' t 11s Ont
:~, methods arc the tools for do·
mstrument s used by the in
_lrlg
vest1ga,
.
d
,es use to ana lyze da ta . Altl
·
lOugh
:nrng and conversing)
b
.
.
) o serv1n
read rng) as the primary m
g
k
eans for
s~i"Veyof pub lished reports Will
~a1res, surveys, systematic obser.
rideotapes, and photographs also
1
~thods are avai lab le when <>
rgas1ng le so urce can cover all meth1 to what can be found in more
>ns below for such sources i11the
, computer management
· · of da ta,
·liat explain partic.:ufar research
qu~litative terminology to look
rckrc::nc.:esassociated with die
· Again, we suggest Schwandt
,se. A second step would be to
-level textbooks recommended
r method-rdate<l articles and
eides that treat different aspectli
ed by Mic.:hael f Iubenuan and
~ generally acc.:cssibleplace to
tmutes to scan recent catalogue
onogrnphs dealing wirh social
115
. us social science discipli nes and applied fields (for exa mpl e,
· varJ0
rch
1.
n
faJroer
Press, Jossey-Bass, Longman, Pine Forge Press, Rou tledge,
1
r\.vir1
r,_ess,
. 15 and Teachers Co.liege Press). Me thodology for qua litat ive
Go
l '
5-igerubhcat!O
. articular
ly active area o f pub Iication and many new resotu·ces
,ch ,s a p
,cseo ach year.
~
,eo' e
,1pl
viewitig-Because th is is a panicu lar ly common form of da ta
7·. fnterinterviewing
·
· accor ded a cI1apre.r 10
· near 1y every qua litative
·
is
collccrion, xrbook.1n chat regard, we think that Patton (2001) and M etr iam
,c:5c,trchrensrirutegood p laces to begin. Seidman (2006) and R ubin a nd
1
1200_
are more specia lized, yet quite accessible . Fina lly, va le (1994)
Rubin logued the mosr common objections co the use of interview dataI1as cata
e appropr iate responses.
\;i05)
K
and som .
B. Field Notes- Again, virtua lly every basic text book covering qu a lita-
. , research offers a chapter on the art of recording observatio ns in t he
~~cd-rhe ubiquito us "field notes ." Lofland et a l. (2005) prov ides a thoug hcf I treatment that has guided several generations of st udents in the socia l
s~iences. More e.,xtended instruction in the techniq ues o.f wr iting eth nographic field notes is ava ilab le in Emerson, Fretz, an d Shaw (1995).
9, Computer Management of Data-Computers
make it possible.: to
perform the complex tasks of data management, c.:oding, retrieval, and
manipulation with a speed, economy, and accuracy never before available
ro qualitative researchers. They also make it possible to waste time and
resources, make egregious errors, and create the illusion of substance where
there is none-more swiftly than ever before. Read, c.:onsult, reflect, and plan
before you decide whether (or how) a computer might serve your proposed
study.
Although many of the older textbooks can provide an adequate introduction to the potential uses and abuses of computer software for manipulating qualitative data, the frequent appearnnce of new software quickly
serves to date rnost of their commentary on panicular systems. At the
time of this writing, Morse and Richards (2002) and Bazeley and Richards
(2000) have published recent guideiworkbooks for the use of NVivo, the
most widely used software package for qualitative data analysis. By the time
the present text 1s in your hands, however, there surely will be other such
resoun:es available, and, 4uite possibly, new software systems as well.
10. Data Analysis-Whether you use a sophisticated computer program
or something as simple as a large accounting lc<lgcr, ana lysis req uires a pla n,
and the place to sketch out the initial shape of that element of your meth odology is in the proposal. The strategy you employ for making sense out of
l 16
Writing the Proposal
~our dar'.l_willbe determined in the first instan ce by the p~~ticular Gllali
rive tradmon you have adoprcd for the study. Some tradmons offer . ta,
flexihility, whereas others make no prescription abo ut the particu lar for 1lttle
an::dysisto be employed. The purpose of your study, the nature of the; Of
set, and even how much time you have avJilahl e will influence the decista
and thus the resource texts, that will be relevant.
,
11
By far the most cornprehens1vc collection o f analytic strategies can b
fo1111d
in the second edition of Miles and t Iuberrn an's sow·cebook Qualitat;ue
Data 1111,:dysis
(l 994). On a scale tliat more closely resembl~ standard text~
books, howt:ver, several other s01H"cesoffer excellent overviews of whati
s of analysis and data repr/
a vailablc. CrcswelI (1998) compares techniqt.Lc
sentati'.)n across five qualitative traditions,_a!1d Coffey ~nd Atkinson (199 )
6
r techmques of ana lysis can
use a single data set to demonstrate how c..hffcrcn
he employed in a compl(.'rnentary fashion.
You will find that one of the most carefully defined and foll.y explicated
forms of analysis is employed in the social science rradition of "grounded
theory." Although rhe term somt:rimes 1s11:;cdin nonspecific ways to refer to
any approach for developing theoretical ideas that somehow begins With
darn, when employed to indicate a specific q ualitative research tradition
something very different is denoted. Grounded theory involves a specific sec
of highly developed, rigorous, and intellectually demanding analytic techniques for generating substantive theories of Socia.I phenomena.
Among various steps and techniques, the grounded theory model oi
analysis employs the operation ca1Jed "constant comparison ." That term,
unfonu11;1tely,has been considerably misused by gradt1ate studenrs to give
an ;1irof scientific respectability to proposals for analysis by means of unsystematic data-snooping. if you wish to avoid the stigma of such amateur
usage, we urge you to read l:Jasic:sof Qua/it,itiue f<ese,irch by Strauss and
Corbin (1998). Although demanding, wirh reasonable diligence it is an
accessible soun.:e, particularly when coupled wit·h the extensive collection
of reports and readings found in Grou11ded Theory in Practice (Strauss &
Corbin, 1997).
:I l. Writing a11d J>uhliccllion- T hcre now is no scarcity of resources
intcnde<l to help you write tl1e report of your swdy. Although that task
might be considered beyond the purview of the proposal, that is not at all
the case. There are aspecrs of the report rhat are not encountered in the task
of writing the proposal, bur the work of crafting clear and precise prose
about a study, whether prop osed or completed, is the same. Just a few hours
expended on a st1rveyof what will he demanded in a sound report can .have
a powerful influence not only on how you write, but also on what you pro·
pose to do during the study.
t
s
r,
ti
p
ti
()
ll
ti
g
rJ
l'f
Pn:paration of Proposalsfor Qualitative Re:-carch
'irst instance by vhe panic l
rl
d
l1 ar
1e stu y. Some tradfrion
%c111
escription about the
. s Offet1· t~,
.
part1cula
'ttt
of your study the
r for"' ~
•
nature of h .,,nr
e available will influence ti t e cl.it
1e relevant.
le decisioa
llection of ana Jytic strategi
o,
f Hubennan 's sourcebook Q~ c~JJ be
nore closely resembles standa;f1tati11e
offer excellent overviews f cl te~t-11 ·
f
O, Wh
• rnques o ana lysis and dara ar I§
ns, and Coffe)' and A·1..
rel?re
.
dif
~1nson (199
' forent techniques of
6/
.
l.
aaa 1Ys1scan
ireful/y defined and foll
.
..· I .
Yexplicar,1
• 1a science tradi tion of "
"~
.
ground ct·
u~ed in nonspecific ways to refere
I ideas tl1ar som ehow b . . .to
,·'f ..
. .
egios with
~r IC qualitative research tr d . .
cJ d
a ItI0
tn e theory involves a specifj n
cctu:illy demandin g· a - • . c set
. .
,
n<1.1,yt1c
tech.
of social plienomcna.
, the grot1ndec..ltheory model of
msram comparison · " Th·at term
rse(I bv· grad uaw studcnr,· t > • '
I F
,, < give
s_ or analysis by means of unsysord tile srig
.f
·
.
.
, , m:i o. such :immeur
lztatwe Nesearch by S
.
·J _ .
. _ . • trauss and
. l reasonable d1'1gen,-,,1·t •
d .
,.,. is an
~ wah the extensive collection
i Theory• in I'racti·ce
- (<'
,-,trauss &
>w is no sc-:ir.·
' ctty ot- resources
our study. /\!though that task
the· pro posa
. J• t Ilat Is
- not at all
are_not encountered in the task
·aft1nodc1r
.,-~nd precise
. pro.~c
•
l">
'
3.,is_the same. Just a few hours
led
. tn a S<>
un(:{ 1•eporr can have
ite, but also· on w h·at you pro-
~
117
bac you begin with Becker's Writing for Social Scientists
e suggesr rnt and highly personal exposition by a master of rhe craft,
9g6),tttl eleg~thGolden-Biddle ru1dLocke (1997), a lively im:roduc tion to
11d , onrinue
" I. texr. WI
·eporc as a form of " storyte u·mg."l n a more tra d'mona
:11I
c ve c
,- 3 I
~ c qua ,c, Holliday (2002) recounts the practica l problems which writers
111
bOok for ~\:~ ·, actempt ro transform rich data from real-life research into a
wee\v1icn r 01~ 1t. The book j s rich.lydiagrammatic, transparent in style, and
(otffl•11doCL'
for studenrs in any di.scipline or app.lied field.
~re
.·
.
.,ppro he.re you can cum to texts that have been honored by vut1:1a
lly unifr()l11c AJcl1ough specific to a single tradition, Van Maanen (1988) has
al use.
.
.
.
vets
a standard reference for the p.reparauon of ethnograpluc reports.
ton~ b~:; generic in its purview, Wolcott's (2001) smaU monograph on
nd 1
,\ _,. g up'' qualitative research (volume 20 in Sage Publications' Qualitative
" \vririn
·
. d
· ) . w1.deIy regarded as a cIass1c
. an d, m
. ttse
. If, consti/Je$l!arCl:
," Metho s series . 1s
.
w~<l. model of good wrmng .
.
Finally, if you can allow yourself to dream beyond the labors of writing,
rhi:r<:;ire, of course, far more attractive activities. Those may include pre~,:ncingpapers, designing poster sessions, locating an appropriate outlet for
publication,perhaps the somewhat less sal.ubrious task of dealing with your
iirsr rejection notice, anc.l, ultirnatdy, the heady thrill of discovering that
your efforts have influenced the thinking or practical decisions of other
people.
To sample some of those delights, we suggest the unusual collection
edited by Morse ( 1997), prepared by 2.4 active researchers in the field of
health care. The focus of Com/Jleting a Qualitative Project is on what
remainsto be done after your data analysis is finished, as you move through
the finalstages of a long journey for which your proposal constitutes the first
step .
12. Standards for Qualitative Research-If there were dear, explicit,
an<lreasonably parsimonious standards for quality at the other end of the
c~searchpipeline (criteria for evaluatin g the adequacy of completed qualitativestudies), those could be used for critiqui n,g and strengthening the designs
pr~s~ntedin p.ro_posals. We are sure you will not be surprised to learn tbat
tlus_1snot possible, at least not in any simple and straightforward way. One
obvious impediment rests in the diversity of qualitative reseru·ch traditions,
~any of them still evolving, and aUof them infused with distinctive intentions and commitments that would have to shape judgments about the adequacy of eacb study. Less apparent to tbe beginner, however, will be the fact
that there has been a distincc reluctance on the part of members of the
research community co engage with the question of qualitative criter ia for
qualitative studies .
I J8
Writing tlif! Prorosal
The most common explanation for this aver sion has been a desire
impose rigid guidelines for prncess within a pa radigm rhat puts the c ll<l't!Q
and individual journey roward undersi·andi.ng at the center of i;ea~i~t
Wliethcr that explanation is entirely true or not, there recently hasq~1ti,:,
sorne tentative movement roward elucidation o f standards to be held\
reviewing a reporr of qualitative research.
~~
~: r'
h~
lf you wisb to make use of the literarure o a standards as part of
,
preparation for writing a proposal, we suggest beginning With an
011
tion of what constitutes evidence in qualitative research (Morse, Swa
1t).a.
Kugel, 200 J) and then opening the topic of stand ards with the brief 11800
butijl~
minating introduction offered hy Rossman and Rallis (2003) . lf that sec u.
J1eJpful,rnore e~tensive discussions of the an and science of critiquing qu~~
itarive research may be found in Lincoln ( 1995) or in the triad of chapre~
authored, respectively, by Sally Thorne, Phyllis Stern, and Judith Hupcey ill.
the collection edited by Morse ("I997).
exa:
1
11
Some of the best writing about rcasornl!Jle expectations for quality has
appeared tn discuss.ions about the nature ot validity in qualitative research
A useful introduction to th.at topic can be found in either Kvale (1995) ;
0
Johnson (1997). If thai brief overview secrns fruitful for chinking about YOUr
proposed study, howeve.r, you will wish to consult much rnore thorough dis.
cussions such as rhosc offered by Eiscnharr and Howe (1992), Mays aa4
Pope (2000), or Maxwell (1992).
:"'!l r~ I
1.,,,1~
!(
,m111:
~ ...
,n.
k,S:
f /.ulr,
:er~-:
0 (1
"L1 l}1 n
A1
,···,, /ti(
(bv•lfc
Q!lcdi.
~.:!II[ ...;
~
1heorc
;,s,;1
.~r
$()Li:il
[l"(llJii(
m;ttic i
i,,n11ar.
In closing tl1i~section dealing with resources, we draw your attention
many of the 12 topical categories
ahove, constitute by themselves an import ant reposirory for information
about qualit ative research. Whatever you are unable to find in the places we
have suggested, you surel y can locate in one of rhe following.
to several more that, while overlapping
In 198(-i, Sage Pllblications initiated a series in Qualitatii 1e Research
l'vlethods. Through 2002 rlie series had accumulated 48 volumes, making the
collenion one of the most comprehensive efforts ever undertaken to provide
support for quafitarive research. Packaged in the forrn of p::iperbound monograph., of 50-100 page~ in kngtl1, the series provides beginners and veterall
rescard1ers with a resource that is both inexpensive and carefully rarger.ed ar
the most troublesome aspens in the planning and execmion of qualitative
studies. Thougl, nor lirnired to qualitative designs, the parallel A.p/JliedSocial
Reseanh 1Hethoclsseries from Sage PubJi.:ar.ionscontains a ntunber of mono·
graphs that may be equ<1Hy11scfriin preparing a proposal. Those dealing with
case study research (Yin, 2002,1, 2002h), ethnography (Fetterman, 1998),
participant observation (Jorgensen, 1989), and qualitative research design
(1vlaxweJJ,2005) represe111excellent (and inexpensive) introductory sources.
rive nu.
i11galt
Fi11,
-tlly.
Q;
irll l'rclis
appear
llJ)d
il:al intt
The f
l'hc dee
th
thO!iC
arc 11Her
(:ind d1f
been 1.:0
l fere, he
ciccums1:
Preparation of Proposals for Qualitative R.eseord,
or this aversion has b
. h.
een a des·
¥1r JD a paradigm thac puts
lte llo
.ders randing ac rl1e
the Ct... t I~
· center f . -~r
y true or
·J
o LtJ t~,
.
.
no t, c1ere recent!
%i
Jc1dar1011 of st andar ds to b Y has b I),
arcl1.
e held
t~~.,
' \\rh
retaru.re on stand . d
t'I)
ar s as Pan
,e ~ug?es r begjnnfog witl1 an of Y0iq
1alttacive research (M
e:ical'lli
· f
orse, Swan Ila.
J1c o . sta nd ards with the brief son,~
sman and Rallis (2003). If tha~ll.tillq.
the art and science of . . . seelh;
Jn
cntiqu1n
·~
J
(l 995) or in the triad of
q4al.
t
~, Ph yllis Stern, and J udith B c ap~er,,
Upce11
·
, IQ
sonab!e expect·ui ons t 1.
.
· . .
' · .ro quaJ,h, I
re ot vaJ1d1ty in q 1· .
., las
h
ua ttanve resea
e found in eitl1er K l (
rch.
va e 1995)
:ems fruitful for th ink·
b
Or
· mg a our
co {;Onstr!tmuch mo
h
YOUr
·
re t o.rough d'
Ihart and Howe (1992) M
is.
,
ays aud
·esources we I .
· '
(taw
your art ·
many of the 12 to . , I , cnrron
' .
.
pica categories
ort,lnt repos1rorv f. . · ,
.
.
. , o1 llltonnanon
are unable to find in the I
.
>neof th f' JI .
p aces we
e o ow,ng.
a
series
. .
,
· in Qi ,at·t
t attve Research
,umulared 48 volumes
I. h
effo.
. .
. , ma ong r e
. tts e,e, undertaken to provide
in rhe form of ,
l
,. . .
pdper )<Hlnd rnono·
.s pt0v1des
bcginn~.
. I
.
· dS ,1m veteran
xpcns,ve and c·11. full
·
' e
Y targeted at
ung
and execu r·ion ot· qualitative
~ .
~signs, the parallel Applied Social
JOnscontains · 'a mun ber of· mono·
ig a proposal. Those Je·1I,·
.h
"t]
.
'
• ng Wit
·:u:nogrnp_hy_(Fetterman, J 998),
research des,gn
,·
,, d (jlla!1tat1ve
.
.
.xpens1ve)llltt·. Ol Iuctory sources.
J l9
format has become a standard means for periodically col1< .
h .in ·vano. us disc1p
. 1mes,
·
•fbCl,:1ndlJOO
• ving researc
as we JJ as presenting
d revie~
.
,
r1,1g on f technica l matters and broad .issues attendrng the research
0
115
'~seu$si0
. If. The Sage Handboo l~ of Qualitative Research (Denzin &
J'
. •e ,rse
.
.
f tl
di
.
.,rcrPr•~ OS)provides a genenc overview o 1e para gm, including its
20
e[JJ1, ol11,
ent cradicions,strategies for inqu.i.ry,methods of data callee·
ornpon
.
1· .
future deve lopment. Althougb
,ut
Y
c
lys.is
111 d ,1na , practica l app 1cat10ns,. and
.
.
.
,ion:in . ·ous in scale and mo.re specific co the field of educauon, TfJe
JtS~ amb'1'
of g,uafitative Research in Education (Le Com pre et al., 1992)
t1a111lbcV:Vei:agc
of many broad topics that relate to qualitative research in
of(ers lied field of professiona l service.
nnYa PPrnber of journals in rhe socia l sciences and applied fields have long
Adnud qualirarive
·
· t I1eu· coverage, but hve
·· have been
researc b. reports m
i11clured
e exclusive
· 1y to t I1at form o f mqu1ry
'
, . Tl 1e l nterrzattona
· l j ournal of
cfevo·t(ltive Stu d',es m
· Educa t·wn began pu bl'1catt0n
· ·m 19 88 and now repre·
.
. ..
Quilt
invaluable
resource
for
locatmg
ongma
l reports, research reviews
5enrs' an
.
.
.
. .
.
'
hc:orericaland technical articles, and book reviews. Now m its third decade
issue, Qualitative Sociology deals with rhe qualitative interpretation of
social life. [ncluding both research reports and articles on theoretical and
rechnicaltopics, the journal is made particularly useful by its freq uent the·
maric ii;sues and the indusion of book reviews in an extended rcview·essay
format. Jn the broad field of health care, the journal Qualitative l Tealth
Researchoffers an interdisciplinary forum for studies that employ qualita·
rive methods. lt has been particularly effective as a re:;ourcc for encouraging alternative approaches to inquiry in the applied fields of medicine.
Finafly, relative newcomers to this spccializ.e<lgroup, Qualitative Inquiry
an<l Qualitative Research focus primarily on rhrnrctical issues related to
:f
interdisciplinary and cross·paradigm research. Accordingly, research reports
appear only when they prcsrnt poinrs of special methodological or theoretical interest.
The Decision to Go Qualitative
The decision to undertake a qualitative study brings two types of problems:
tho~e that: are external and mostly Mltecedent to rhe proposal, and those that
ar.(•internal and as~ociated with devising an appropriate design for the study
(and then writing the prnposal document). To chis point, this chapter has
design and its written presentaLion.
been co1Kerned with the latter-the
l~crc, however, we want to mrn briefly ro the precursors and surrounding
eireumstances that will exert their influence on that process.
J 20
W,·_i(j11g
the Propos,,I
The problems that precede the preparation of a qualitative proposal b
·ep·1
with the author as a person. Everyone who is tempted ro employ a qtiat
1
rive design should confront and honestly answer one ques tion: "Wh tta,
l 'Nant to do a qwilicative study?" Some novice researchers, trautna;. do
·
I1-gral Ic encounter wit· 11 f racnons,
·
· · research. tzed
by a fourt
see qua 11tat1ve
f
·d·
I
·
1
d
·
·
·
· Iar. So Ionasa
way o. avo1 mg num >ers 111 genera an s-catisncs 111 parc1cu
question and paradigm truly are well matched, however, a choice tnad; as
0tll
· lf, JS
· ne1r1er
· I improper
·
·
such persona j groum Is, ta I<en l)}' )[Se
nor inevirab
.. . I.
1/
dy~.tunctrnna
It also is true_, howcve1:, th,H findi ng certain kinds of questions appea(i,
. not the same as .t1av111g
. t l,e perso na I capacmes
. . ancI mte
. 11ecrua I intereslg
1s
demanded in r.he conduct of a quali tative study. If avoiding staristi .cs is
:m much ,l bad reason for ekcting co go qua litative as it is an -irrelevantrea~
son, tbcn personal v;:ilues that are compatible wid1 a qualitative woddvie\1/
::ire not so nHH.:h,in irrelevant ration ale as they are insufficient. Determining
a mode of inquiry that matches both your research goa ls and your research
c:Jpabilirics require~ a more e.laborat e calculus.
ll;s
\\le are not suggesting here th at qual itative research must pJ·esume
unusual capabilities or exceptional inceJlect . To the contrary, ow· experience
has been that a .large prnporrion of propedy trained individuals can do perfectly compctcn1 qualitative research-if they are strongly motivated to do
so. The demands, however, are very real, and your capacity to meet them
should be considercrl with honest rnr e. At th e least, factors such as interacti.vc social skills (people skills), a sensir.ivc ear for nuanced language, and
,maly1i( capabili.ty (pattern recognition) are far more relevant to success than
either an interest in soci;:d dynamics or a bad case o[ marh anxiety!
In the same vein, a graduate srndent who elects qualitative research
because it appears eithe r to be relativdy "quick" in terms of time commitment or "easy" in terms of inte llectual demands has, in the first instance,
simpl y never ralkecl wit h anyone who ha s completed such a study and, in the
secon d instance, not read published r eports of field research wirh much care.
Q11alitative studies al'e never quic k and rarely arc completed within the pro·
ic,:red time lines. The aualysis ot qualitative darn demands a sustained level
of creative thouglu rarely required of the investigaLOr once data are colll'cted
in quantitative study. Qualitative research may be enormously v:lluabk for
many purposes, as well as irnrnenscly satisfying to the invcstiga1or, bur quick
:-rnd easy it is not.
,l
The vital codic.iI to this discussion of person,,[ factors in sekc(ion of a
research paradigm is that whatever one\ abilities or imercsts, it. still is necessary to match the procedures of inquiry with the purposf.s of inquiry. l'ur
anorhcr way, a close and comfortable rnarch. between your abilities and
Preparation of Proposals for Qualitative l{esearch
arnr:ion of a c1uaUr .
at,ve Pro
: who is tempted ro en J Posa1h
,.st IY answer one
lp ou
<:r;•
. , a % :"'l
question- " afi
ome novice researchers
. ~,,., t.l.
•
> (f
I Q
cr,ons, see qua litative
attnlat· 4
. .
rese
'~·d
d
'.l srac1sr1cs in part icu l
arch as
narcbed, however a cl1a~.So l0110 ~
·
.
.
,
Otce 111 <>14
, is ne ither improper
. nde
nor ine,.,- ()~
"
.
.
ttabi~
~erra1n kinds of quesci
,
..
ons app
capac1t1es and inrellecru · / . eafi11&
·e study. If avoid ing st _a _ltlrer~
• .. .
.
at1sr1cs· ·
qua 1rtar1ve as it is an ,· . ,
is 110,
"bl
ne,eua
m e wid 1 a qua /itarive
nt te.i_,
WOdd .
j
~ t iey a re ins1,1,
t:'l·cientD
view
11
eteritl' .
' ·
•
r research goa ls and
tn1ng
:u/us.
your research
ialitative
research in
ust presu
o the contra ry, our ex
. Ille
·/y trained d1" 1.d I
per 1ence
" ua scan d 0
hey are strong ly
.
per.
moc1vaced to d
and your capacity to meec th o
tbe least , factors suc h as .
eni
. T
t.
w·
. .. .
•nterac-
e,11 for nuanced language
d
far more relevant to
. , . .' an
:l
.
. success than
( case of rnath anxiety!
who elects
1· ·
.. · ,, .. , qua iuitrve research
utck IJ1 ter·ms o 1·
.
· tnnc
com ·•
lands has in the 1·· . . .
m,t
.
'
irst mstance,
nplcted such a study ·111d 1· I
ff I l
' '.n tle
. le t research with much care.
; are ~ompleted within the pro:~ta c,emands a sustaiued level
)tlgaror once dat~ :ire collected
tTy be
<
enormously valuable for
~ to the investivato1·
I)
.k
·o · , ut qu1c
~ma! factors in selection of a
cs or interests, it still is nccesthe purposes of inquiry. Put
berwecn your abilities and
1 21
11e 00 e band, and th e oatw ·e of qualitative research on the
05 on t uinely desirab le as that may be, still.w ill no t make it pos ·'llc:"riv
reti I •1S oen
P 1.erb111J',' 0d met hods into sq uare questio ns. If )'O U start out committed
,,,,,
d
If
·t·lc?co f,r rou.n. rive researc h, then yo ·u Irnve Lim.
1te youi-se . to on ly those
1
ii '
· tJUa ,ra
· ·f·1c pa ra d"1gm . Se Iect1on
· o f"t he means for
JoiJJg
besr yield tot I1at sc1ent1
,aucs110
. 11s chat
.
f
I
. aJ
b
h
. .
. ro identification o t 1e quest1on · ways ears t at restnction .
q qtH'
· ypflO(
I
bl
h
..
,
b'li
d ·eproposa pro . em centers on t e mvesuga ro r s a 1 ty ro
111
pi quant 1·tat 1ve
· mo de o f t houg I1t. For many gra d uate stu dems ,
"A ~·ectirl f che
O
,r,o"c our of adopting assumptions that are consistenr wit h the qualitat ive
.
. we II-worn ha b.its of tho ught. Fo r most of us,
,he:Process
. means
breakmg
ecave
. .
pc:rSP pcions of quant1ta t 1ve research have been presenred an d learned
ssum
f
I ,1
rie
• . ce" t11ro ugh many years o school and university ed ucation . Th e
, sc1en
11
ken premise of the perspective use d in the ph ysica l and natura l sciences
0
u119P ·f somet hing tru ly exists, it must exist in some quantity , and exist "ou t
j5 tl,lt
1 I
,, in some finite form . Although t hat certainly .represents th e most
cbere
.
f.
. . . thoug h r, .m .funcnona
. I terms .1t mar(<S
i
remc fo.rmulat 100 o posmv1st
the
ext
I
.
"
J"
d
"
l"
.
h
f
disrinccion berwe~n w 1at 1s cea _
not rea . m. muc o. our everyday
a?.
Ji{e.To adopr a different and unfamiliar perspective 1s somet imes more difficult cban one might expect.
To think anc.l write in a consistent fashion with the assl.lmption that
people construct reality, allowing truth to reside as much in our heads as
~our there," requires a sharp alteration in the habits of intellect. Even imperfectly accomplished, this is difficult for most, and, as experience warns us,
impossible for some. It is important to confront this problem during
rhe early apprenticeship stage of research training, when patient and sympathetic mentors can assist with the difficult transition between familiar
ways of pcrceiving and conceptualizing, and a different vantage poinr-the
qualitative worldview.
Before making the final decision to go qualitative, there also are logistic
questions to consider. Prepar,uion for qualitative research is most effective
when it takes the form of apprenticeship, wirh intensive field experiences
and closely supcrvised opporrunitics to practice the analysis of actual data.
Where those opportunities arc not available, some hard questions must be
addressed. Will it be cost-effective r.o acquire training and expcricnce
through some alternative source? Ts there sufficient time to invest in both
~ reasonable level of preparation and a lengthy study? Will the best solutton he to transfer to another department or institution? These are har<l
questions, but better raised earlv than too I.ate.
Hav ing fulfilled Our ob ligacio ~ to sound reasonable.: cautions, however, the
la5t th ing we wo uld w ish is to leave the impression that the opportunity to do
qua lita tive research is anything less than an exciting, thoroughly fascinating,
122
Writing the Proposal
and deeply fulfilling option. Herc is a form of research that invites qu .
. est1
that deal with how real peop le think and feel. Here is a way to produce
/.l~s
ings that are tho roughly grounded in the stuff of a recognizable reality •net.
world as it is experienced. Finally, here is an oppo rt unity to join With -.th~
scholars in an enterprise characterized by fresh ideas, energetic expan~~hct
and as-ycc-uncxplored possibilities. If all of that sou nds appea ling to '00 ,
then yes, you should consider goiug qualitative.
You
,
Notes
I. The word /)CJrad
igm, as used here, dr:11or.esa conceptua l framework that p
vidcs a particular way of thinking about mcanir1g in the conrext of fo.rrnal inqu(~·
Tims, raken collcctivdy, the beliefs, values, perspect ives, comin itmenrs, and co :~
11
q1mir methods of inquii: y shared by a gro up of investigators const itute a scientific
paradigm. JJJst as each pcr.~onhas a cognitive schema for making sense out of his Ot
her daily experience in chc world, scholars who share the assumpt ions of a paradiglll
have a particnlar w;iy of making sens,~ om of their s.::iencific world. Social scientists
who perform experiments, for example, slrnre a gen era l perspective on their Work
lime is distinctly different fro111thar. shared by inves tigators who do srudies in the
feminist tradition of qualitative re.~eareh. Accordingl y, it wou ld be asserted, for elearnple, th:?t c.xpcrimencal (quantitative) and feminist (qualitative) research have their
roots in differenr scientific p;~radigms. Any modern textbook on qualitative methodology will provide a srnning place for defi11ing the qualirarive parndig111,but nearly all
of them will lead you b;ick w Lincoln and Guba ( J98S), and from there to the modern
origins o[ the rern1 paradigm in Kuhn (l996}.
2. Our use of rhe term qttantitati/le ,1lso is ,1rhitr;~ry. The simple presence of
numbers in a srudy, per se, does nor serve to disringuish one paradigm from anorhcr.
BorJ1quantirative and qualirarivc: research can employ quantificmion. fr is the tmderlying assnmptions ahom those numbers that provide rhe distinctive differences.
3. 'fhe rerm trnditio11. in this rnnrexr represents a co11vcnrion for coUectivdy
designating rhe various disrincrivc forms of qualil-arivc research. fr docs not carry rhe
genernl, common-use sense of something pa.,sc:ddown, gencrnr ion ro generariorJ,
over a long period of time. Instead, it is employed in rl1e 11a,:rowc.r .~ense of designat ing a coherent body of precedenrs intended to govern some se t of actions -i n this
c<1se,a mode of though t and a related set of research procedures. Hence, phenome nologic~l research is a tradition within the quautarive pa radigm, as are ethnogra phy, fife history, symbol ic interac tionism, grounded th eory, and case studies.
4. Taxonomies for qualitarive approaches ro research may be exrended beyond
what ha~ been ronchcd on in the present chapter by including artistic as wcU as scientific modes ol inquiry. For an introduction ro that top ic, sec Ei.sncr (J 981) and
Tborntou (1987).
5. Our use of rlie term empirical here: i.~not employed as it sornerimes is in
rhe discipline of philo.~ophy, ro dcsignare the family of rheoric.~ called empiricism.
Preparation of Proposals for Qualitative Research
·m of research that invites
c
quesf10
ris
xeeJ. Here is a way to prod
uce fi11
stuff of a recognizable rea Ji
d.
, an opportun ity to join w-·thty_t
he
.
I
0~
Y fresh ideas, energetic exp . et
ans1
of t hat sounds appea ling ro 0n,
You,
tative .
res a conceptual framework tl
. .
. iar Pro
ung 1n. the conrexr of formal ·1nqu. ·
1
rspect1ves,commitments and co ry,
·
'
nse.
1f ·
mvest1gators consrirure a scie,ir·fi
h
'
.
c ema ior making sense our oft · l~
.
.
11SOr
sJia1e the assumpt ions of a parad·
. sc1enti
. ·c·
1e1r
1c world . Social sc,·en •gni
.
tlSts
a• genera
l
perspective
on
their
w k
.
Or
rnvemgarors who do studies in h
t e
I
.
1'.1g Y, it would be asserted, for exarn.
1st (qua litative) research have ti .
<
lCJr
! m rexrbook on qualitarivc method~ qualitative paradigm, bur nearly all
l985 ), and from there to the modern
arbitr ary. The simple presence of
nguish one paradigm from another.
ploy quantification. It is the undcriJc the distinctive differences.
enrs a convention for collectivelv
1tive rese:m.:h.Ir does not carry th~
down, gencrarion to generation,
!<lin the narrower sense of desig;overn some set.of actions-in this
rch procedu.res, Hence, phenome·
·ativc paradigm, as are erhnogra·
:d rheor.y, and case .~tudies,
escarch may be exrended beyond
Y inclu<ling artistic as well as scihat topic, see Fisncr (1981) and
employed as it sometimes is in
ly of theories cal!e<lempiricism.
1n
in che co.mmon-use sense of designating inquiry based on che data of
0 111
,fc u~eir :·ngs
char the investigator saw or heard chat can rhen be employed as
1
· research 1s
· empmca
· · I.
, p.:n·ence,t rfor a claim. '-u1 t I1at sense, a11qua Iiranve
should be aware that the word validity is not common ly used in
111e wMr.a~~ers
6- Re,. of qualita tive research. In fact, there is considerable discomforr with
tiecoric
..
rhe r
some rejection of the construct. Our dec1s1on, nevertheless, co empl,oy it
,111dl!ve
hn c this chaprer is based on two simp.lc facts. First, most of our readers will
ug ou
,1tro 'li·ir wirh the general use of the term ro deno.re a datum that accuncely repbc:flit11'
11'e phenomenon to which it refers (it is true), or a research finding for which
resents'. ~ evidence has beea presented (it is certain). Second, the formal language
Iace rJ1e construct of validity
pcarS lHISIV
been invenred by qua I'1tat 1.ve reseai·chers ro rel_)
113 5
t1wr sive complex, and far from sufficiently universal to constitute a reliable sys,~
' lf you dec1
'de to wnce
· a proposa I fo.r qua 11·tac1ve
· researc h, 1·c w1·11be
· exren
hinguage.
re~. sary to assimilate this new language (we suggest that you begin with Lincoln
ne(;es
•
· .Morse et a I., 2001,
. and Kvale, 1995),
d Guba, !98.>;
Gu ba an d L'mcoIn, 1989;
~::r for our present purpose rhe single term va,fidiiy will suffice.
6
StyleandFormin
Writingthe Proposal
T
he writing style of rhe thesis or grnnt proposa l may be the most imporwnt facto1· in conveying your ideas ro graduate advisors or fundfag
agencies. Even exper.icn,::ed researcher s must critica lly eva luat e their writing
ro ensure that the best bid plans are presented in a clear , straightforward
fashion. Tl-re sections that follow represent primary concerns for proposal
writers.
Praising, Exhorting, and Polemicizing: Don't
For a vaticty of motives arising principally from the reward system governing other writing tasks, many students use their proposal as an opponunity
to praise the importance of their dist:ipline or professional field. So.me use
cxhottative langm1ge ro urge such particular poinrs of view as the supposed
importance of empirical research in designing professional practice. Others
use the proposed research as the basis for espousing rhe virtues of particular
social or political positions.
There is no need or proper place in a research proposal for such subjective si<le excursions. The purpose of a proposal .is to set fonh for a reader the
exact narure of rhe matter co be investigated and a detailed ac'-=ounrof the
methods to be empJoyr.:d. Anything else distracts !lml serves as an impediment to dear communication.
12':i
12G
W:·iting ti,e Propo~;a]
th a general rule, it is best to stick
resisc the telllPta .
ro soL1r1d"properly positive~and enthusiastic." Do not at tempt to rnan~10ll
fate the opinions of the r~arler in areas other than rbose essent ial to the irl~lltigarion. The simple test is to ask yourself this question: "Does the reacfes.
1·eallyneed w consi<ler r.hispoint in order to judge the adequacy of Fl'\ythi ~r
iug?" If (he answL:r is "no,'' d1en the de,:ision to delete is dear , if nor ahv-n•
easy, tor
. the author.
ah
to the ropic and
Quotations: How to Pick Fruit
Frum the Knowledge Tree
f( [
Too often, inexperienced writ·ers are inclined co equate the number of citations in a paper witb tbc weight of rhe argurn ent being presented. This is an
eao1·. The proper purposes served by ihe system of scholarly citation are lini.
ired to a fevv specific tasks. When a <locurnenr has all the citations needed to
meet 1.hedemands of those few tasks, it has enough. When it contains more
citations; it has too many and is defocrive in rha c regard. Reviewers deem the
-~'SL' of nonsekctivc rc(crenccs as an indic:nion of poor scholarship, an inability to discriminate the centrn] from die perip heral and the important from
the trivial iil research.
-rhc proper uses of direce quotation are even nwre str ingendy limited than
the use of general citations for paraphrased n1.ncriaL The practice of liberally sprinkli.ng r.hc proposal with quoted rnatcria!-particubrly
lengthy
q11oration~-is more than pointless; it is self-defeating. The first trmh is that
uo one will read rhem. The second tn1d1 is thar most readers find rhe presence of. llJJC:'Ssential
quotations irritating and a distracrion from tlie line of
thought being presen1ed for examinar.ion. When quotations are imroduccd
at poinrs for which even general cit;:itions arc lumecessary, rhe writer has displayed clear dis1·egard for rhe reader.
There are t\-vo legitimate motives for direcc use of another scholar's words:
(a) tlic weight o( autho1·it;1tivc judgmcm, in which "who said it" is of critical
imporrance, and (b) r.he nature of expression, in whid1 "how it was said" is
rhe important clement. In the Fonner insrance, when unc:'{pected, unusual, 01·
genuinely pivotal po ints are to be presented, it is reasonable to show the reader
thar another competent crafrsperson has reached exactly the desired conclusion, or observed exactly the event at is.~uc. ln the larrcr instanl:e, when another
writer has J1i.ton die precise, perfect phrasing ro express a difficult point, it is
proper r.oemploy that talent on behall o.fyour own argumem. The rule to follow is simple. If the substance of d quotarion crn be conveyed by a careful
parap.bra.se, followed, of course, by the appropriate credit of a citation, with
Style:and Form in Writing the Proposal
rhe topi(; and resist the tem
·1s··t1'
, . c· " O
- o not artemin to Ptati011
ma,1'
icr than those essential to th . 'Pu,
. h'
.
e Ith,,
•'. t t 1s question: "Does the l' ~'
t_ojud~e the adequ acy of rny t~~clcr
,wn to delete is clear , ii nor I
'l'lk,
a Wah
1ed to eqliate the number of .
.
· c1ra.
umenr being presented. This is a
·stern of scholarly citarion are 1· n
111).
cm has all the citations needed to
s enough. When it contains more
1 that regard. Reviewers deem the
>n of poor scholarship, an inabi[.
·iphernl and rhe important from
1
en more stringently !i:mitedthan
' materia l. The practice of Jibermaterial-particulady
lengthy
-defeating. The first truth is that
that most readers find che presI a distraction from the line of
7
hen quotations are introduced
unnecessary, the writer has disuse of another scholar'swords:
hich "who said it" is o.f critica l
in which "how it was said" is
when unexpected, m1usnaJ, or
s reasonable to show the reader
red exactly d1e desired condu1elatter
in.stance,when another
~ express a difficult poinc, iris
)Wn argument. The ruli"co folcan he conveyed by a (;areful
riate credir of a citation, with
127
.
I ersuasive impact of the original, then don't quote. In almost
1
ht' clnrrY~n~ pbest for the propose r co speak directly to the reader. The
I
ii r
es it is
'k I
' 1 j,1sr11nc
' f ords from a third party should be reserved, h ..e 1eavy canJ1
·now
,nrllr.venno
battle, fOJ. those rare instances when the targets are spec1f1cand truly
of the contest.
riofl in
I outcome
· ·
·
I ·
·
I O c1e
·ririca
c
..
J
reclmique
fox studencs who rccogmze
t 1c1r own propensity
"
I7 1efic1a
,
..
A el
·ve quotatio n is to use the critical summary form of noted excessi
· a .fu 11c1tat1on,
· · eac I1 art1c
· Ie rs
·
·
format,
coWM
after carefully .recording
1
1· o In nis ·
f
d
·
I
d
'
r:ik J11:1
·
a mined and then paraphrased on re erence rnr s 111t 1.estu ent s
· ·c·1
· · 1s
· ma de on w I1eth er tl1e aest.he t crir•
' IIYdex, During note-ta k'mg, a dec1s1on
wor
s.
·
· JllStt
· 'f y tI1c
own
· g or the author' s importance. m terms o { aut I1onty
. 0 f phrasin
. .
'~ . ~£ direct quotation. Ex,cept in ra~e msranc~s, quoted material 1s r_10t
use ferred to t11e note cards . Thus ' direct quotmg becomes less tempting
cr1i~
s d1 ubsequent writi.ag phase when the stu<lcnt has recourse to notes.
during e s
·
·
I I · ·
. . J 1ique also preve.ncs unmtenr 1ona p ag.1ansm.
This cec11
.
I .. ·1
.
· a compu ter for note stora ge
If using
_ and . retneva , s1m1ar precrnt1ons
Id he taken. When retrieving intonnation from the computer, you
shou., . ake ccrtam· that cach item
·
, het· as yom ·
can dearyI be 1'd ent1·f'1ed e1t
s.ltouu 1n,
.
.
.
.
. .
.
,
'
c1uotat1on.
It 1s possible to lose thts mfoi:mat1on as
paraph ra. 5,...,or as a direct
.
.
'"'J·t,·h
back
and
forth
between
notes, . computer, and proposal •docuyou Sn
..,
'
.
.
. .
.
.
menc.Onc way to ensure 1dent1ficat1on, which can be used both on the computer an(l O n handwritten note cards, is to use
. quotation marks for .all direct
quotations, listing the page num~er on wh~ch the text was found m parentheses immediately after the dosmg quotauon mark. As you work betvveen
notes, computer, and writing of the proposal, transfer all of this to your
draft.
Clarity an<l Precision:
Speaking in System Language
The language we use in the commerce of our everyday lives is common language. We acq uired our common langu.age vocabu lary and gramm ar by a
prncessthat was grad ual, unsystematic, and mostly unconsci.ous. Our everyday language serves us well, at least as long as the iuevitable differences in
word meanings assigned by different peop le do not produce serious failures
of communication.
The language of science, spec ifically the language of research, is uncomm.on. The ongo ing conversation of science, for wbicb a research proposal is
a plan of entry, is carried on in system languages in wh ich each word must
mean one thing to both writer and n:adcr. Whcre small differences may
128
Writing the Proro sal
matter a great deal, as in research, rhere must be a minimum of sl·
1
between rhe referent objecr, the word used to stand for the object a PPa~
nd
images called forth by the word in the minds of listeners and read:rs_ th~
The rules of invariant word usage gjve system languages a high
of precision. Minute or subtle distinctions can be made with relativeOtd~,
Evaluative language can be eliminated or clearly segregated from ern/~se.
descriptive language. More important, however, the language of res lttc..1
affords the reliabiliry of communication that permits scientists to cr~atch
powerful interdependent research enrerprise rather than limited indepen~ea
investigations. When a chemise uses the system language of chemistr ertt
communicate wirh another chemise, the word "element" has one.and ~ ~o
one reference, is assigned co char referent on all occasions, is used forn Y
other purpose within the language system, and consistently evokes the sal\o
imagsin rhe minds of everyone, everywhere, who has mastered the lang~c
Various domains of knowledge and various research enterprises are cL-e,
11ar.
acterized by differing levelsof language development. Some disciplines,such
as anatomy or entomology, have highly developed and completely regular.
ized language systems, whereas ochers, particu larly the behavioral sciences,
employ languages still in the process of development. Irrespective of the area
of investigation, however, the language of any research proposal must, asa
minimum requirement, be systematic within itself. The words used in the
proposal must have referents that are clear to the reader, and each must consistently designate only one referent. When the investigation lies withina.
subject area with an existing language system, then, of course, the author is
bound to the conventions of that system.
Obviously, rhe researcher should be familiar with the system languages
that function in the area of proposed investigation. Reading and writing
both the specific language of the subject matter area and the more general
languages common ro the proposed methodology (sratisrics, experimental
design, psychometrics, qualitative traditions, ere.) are clear requirements for
aoy study. Less obvious, however, is the fact that research proposa ls, by their
exploratory nature, often demand the extension of existing language into
oew territory. Operations, observations, concepts, and relationships not
previously specified within a language system must be assigned invariant
word symbols by the investigator. More important, cbe reader must carefully
be drawn into the agreement to make these same assignments.
Advisors and reviewers r~isunderstand student proposals far more often
than they disagree with what .isproposed-:-i"he failure of communication ofren
occurs precisely at rhe poinr where the proposal moves beyond the
use of the existing system language. This problem involvesa failure of careful
invention cacher rhan a failute of mastering technique or subject maccer.The
Style and Form.in Writing the Proposal
e must be a minimum of sJ·
ed co stand for the object 1Pr>c111
,,
· d f 1·
'and Ill.
m s o . 1stenersand readers · t~e
ve system languages a hi
0
is can be made wit h rela:ve tl'Jer
· clear,Jysegregated froiu,
eitsc
empir· .
owever, the language of . tc~1
(esea
.
.
cbat permits scient ists to ere %
se rather than Umited inde ate a
h
nd
system language of chem~e eat
1stty
vord "e lement " has one.an d to
- ·
· used f 0nli•
·· on al l occasrons,
rs
.
Or l1o
an d1 cons1srent
ly evokes
•
____ th_e,s<1n1
~' w 10 nas mastered the lang~ c
tous research enterprises ar h&e.
e c ar.
. .
,1eIopmenr. Some chscrplines
, 'such
~veloped anJ completely reg I
. I
u ar.
·c1c:uar Iy the behavioral' ·.sci·ences
of tL,.
elopmenr. Irrespective
.
uc area'
any tescarch proposal must
, . j' .
> JS a
in itse f. 1 he words used in the
·o the rca<ler, and each rnt•st
.,. ,.
-.on11 the investigation lies within a
m, then, of course, the author is
iili:u- ~ith the system languages
~st1gat1on. Reading and writin<r
I:>
:itter area and the more general
·dology (statistics, experimental
, etc.) are dear requirements
for
that research proposals, by their
1sion of existing language into
:mccpts, and reJacionsl1ips not
:m must he assigned invariant
>rtant, the reader must carefully
;amc assignments.
,dent proposals far more often
failure of communication often
proposal moves beyon d the
lem involves a failure of careful
chniquc or subject matter. '[he
129
,s mav he of some help as you attempt to translate a personal
'
into the form of a carefully specified public record.
. ,,.rut.t
111
1101v
~ lw unknown
111
..
oft
fl
~J~Jll
· v·ent new words when the exi.sting svstem language is adequate. 1frhe
r-Jcveru1
.
f . r in established use has a label that excludes whar you do not want and
e c1cn
.
r. J des all that you do want, then 1t needs no new name.
1. ·
JJJC U
2.
.f we.reis reasonable doubt as to whether the word is in the system or the
1
doma in, provide early in the proposal the definition that will he ~
coni11100
throughou
t. Readers may give unnecessary . time and attentio n ro deci- y
h_
d
;:,:r ing the intended meaoing unless you put their minds at ease .
/'~
~
~4-r;..
Words that have been assigned system meaning should not be used in their.
J. common language form. for example, the word significant should not he
useJ co denote its common language meaning of "important" in a proposal
involving the use of stati.~tical analysis. The system language of inferential
statistics assigns invariant meaning to the word significant; any other use
invites confusion.
4. Where a system language wor<l is to he use<l in either a more limited or a
more expanded seme, make this clear when the word first is introduced in
the proposal. If local style requirements permir, rhis is one of the legitimate
uses of footnotes to the text.
5. Where it is necessary to assign i1wariam meaning to a common language
word to communicate about something not already accommo<lated within
the system language, the author should choose with gL"eatcare. Words with
strong evaluative overtones, words with a long history of ambiguity, and
words that have well entrenched usage in common language make pour candidates for elevation to system status. No matter how carefully the author
opcrationali1.es the new definition, it is always difficult for the reader to make
new responses to familiar stimuli.
6. A specific ddinirion is the best way to assign invariant meaning ro a word.
When only one or two words require such treatrnent, this can be accomplished in the rexr. A huger number of words may be set aside in a section of
the proposal devoted to definitions. The best. definition is one diat describes
the operations that arc required to produce or observe the event or object.
For example, note how the following words arc assigned special meaning for
the purpose of a proposal.
a. i\ common language word is assigned invariant use:
Exclusion will be deemed to have occurred when both of the following
happen: The .~tudent no longer is eligible to participate in extracurricular activities unJer any provision of school district policy, and the
student's name is stricken from the list of studencs eligible for extracurricular activities.
l 10
Writing the Proposal
b. A system language word is employed with limitarions nor
' Ot<fi
assigned:
ti~tilt
The curriculum will be limited co those after-sdtool acriviric
currenr School District Mam,al lisrs as approved for secondas t11ilt the
ry s~1<>uJ
l
srudenrs.
c. A system language word is operarionalized by describing a criter"
lncreased motivation will be presumed when, subsequent co a~On:
·
• spent m
· any exrracurncu
· Jar accrvity
•
Y tr"'
menr con d.
. 1tton,
t he rune
rises
·•t\ than LO% o.f the pi·evious weekly total.
niarc
d. A common language word is operationalized by describing a criter' .
D1·opout'sare defined as all partidpanrs who fail to attend three c ton:
0 nscc.
ucive activiry meetings.
e. A system language word is operat ionalized by describing procedur:c:
Rein(otceme1ztwill refer ro the procedure of lisring all club member .
the school newspaper, providing special hall passes for members.,and~ 111
st_
iragclub memberships on school transcr iprs.
f. A common language word is operationalized by describing procedure:
lnstruction will consist of five 10-minucesessions in which the clt,b spon.
sor may employ any method of reaching so long as it includes no fewer
than five attempts for each student ro complete the activity.
Editing: The Care aud Nurture of a Document
A proposa l is a working document. As a primary vehkJe for comrnw1ication
with advisors and fu11dingagencies, as a pla1.1for action, and as a contract,
the proposal petforms functions that are immediate and practica l, not symbolic or aesthetic. Precisely because of these impoi:tant functions, the pro.posal, in all its public appearances at least, should be free from distracting
mechanical errors and the irrirati.ngco11fusionof shoddy format.
At the privacy of your own desk, it is entirely appropriate to cross out
passages, add new ones, and rearrange the order of paragraphs. The series
of rough drafts is pan of the process tlu·ough which .is proposal evolves
toward final form. When, however, the proposa l is given to an advisor, sent
co a funding agency, or presented co a seminar, the occasion is public and
calls for an edited, fo.nnally prepared document. The document shoul.dbe
easy co read-for which a good printer and high-quality duplication are the
first essentials.
/
Every sentence must be examined and reexamined in terms of its clarity,
L grammar, and relationship with suxrnunding sentences. A mark of rhe neo·
phyte wcirer is the tendency to resist changing a sentence once it is written,
and even mo·re so when ir has been t)'ped. A sentence may be grammatically
correct and still be awkward within its surrou ndings. The cough test is che
Style anrl Form in Writing the.:Proposal
ployed with limitations
nor <>rd
·
ro
. .
l\~~
:o rhose after-school
1·
acriv1f
ists as approved for
res th
secondar <It'!it
.
. .
}' Sch
ronahzed by dcscribin
.
lliJ
.
d h
g a Crit..._.
,ume w en subscqu
~,10,1.
'
enc to
.
any extracurricular acri - ariy tr
I
:oral.
v ty rises,,,!)
~~,
.
ttionalized by describin
.
1'
g a crrt, .
Panrs w I10 fail ro arte d h cr,(>n.
. n
t
rec
.
>nal1zedb)' describing
.
Cor1s~
Procecf
,c~dure of listing all club me Ure:
:c1a/halt passes for memb . lllbersift
.
ers a0 d ·,
'
list.
nscn pts.
ionalized by describing
.
procedu
nure sessions in which :h
re:
·I .
.
t e clubs
img so long as it includes no Pon.
o comple_tethe acriviry,
fewer
, If in reading any sen ten ce, a colleague or reviewer hesitates,
,c:sr 11ere. ' co reread the sentence to understand the content, rhen the
!,(!!' 1,1es,
or habs examined for possib le revision-no
matter how dcgant,
~fl111l , n1usr e
.
th
I
.
ctJ1cc
·ecise 1r seems to e au t 101.
«11
· · g an d rewnnng,
· · t he most he..l p f1.1I provir1
1s1, iind p.1
. eciculous ca1·e 1·0 writin
"t, Asrde
. fron1
rn
.
.
.
.
.
, ial revis ion ts ro obtain the assrstance of colleagues co read
·n editor
· an d ma
· dequac1es
· o f confor rrred1a11i
ci:J.11f' 1
,ca I errors, 1ac I, o t·c Iarrty,
1
ihcproposahor can read rhe same erro r over and over without recognizing
11
,c11
1r. .A aur babilirv of discovery decl ines with each review. The same error
d hepro . .,
ti · on r
r once cothe attention of eve n the most c.:a~ualreader who is read111il)'lc:ip ~oposal for rhe first time. On e useful trick that may improve the
ing rh;.t~bility co spot mec hanica l erro rs i~ ro read the sentences in reverse
11u
cho
destroying the strong pe rceptual set created by the normal
order, n1 u5 .
enceof ideas.
seq11 houub format will be a matter of individual taste or departmental or
0
1
/\ c t·egulation, severaJ general rules may be used in designing the layout
agencyof rhedocument:
[.
>fa Document
mary vehicle for cornm11n1' .
f
.
cation
in <~ractwn, an<l as a contract
'
mediate
and l)f''('tJ.("ll
.
'-' · · , , nor svm~ important functions r)1n ,
I
.
, ...pros l~uld be tree from distracting
n of shod<ly formar.
tirclv- apprc
..
>pi.·.
t<tte to cross out
,rder ot
.
- paragraphs. Tl1e ..senes
!g1I which is
·
·.. Jn.oposa I evolves
13 l
Use double spacing, substantial margins, ,ind ample separation for major
subsections. Crowding makes reading both difficult and unpleasant. Always
number pages so that readers can quickly refer to a specific location.
2. ]_\,.fakeample use of gr:1phic illu,~tration. A chart or simple diagram
can improve clarity and case the difficult task of critical appraisal and
advisement.
J. Make careful and systematic use of headings. The system of headings recommended ir1 the Publication lvfonua/ of th(: i\meri.t'an Psychological Association
(2001) is particularly useful for the design of proposals.
4. Place in a11appendix everythi11gthat is not immediately essential to the main
tasks of the proposal. Allowing readers to decide whether they will read supplementary material is both a courtesy and good smncgy.
sal· is •giv<."l
d .
' I-0 an a nsor, sent
1
ar, d1e occasion is public and
em. The document should be
gh-guality duplication arc the
In Search of a Title: First Impressions
and the Route to Retrieval
1111ined
1·11re·1
· c larrty,
.
ms o f its
entences . .t\ mark of tl1e neoa sentence
· written,
· · once it· 1s
ltence may he grammaticallv
The title of the proposal is rhe first contact a reader has with the proposed
research. First impressions, be they about people, music, food, or potential
research topics, generate powerful anticipations about what is to follow.
~hocking the reader by implying one content domain in the title and following with a different one in rhe body of the proposal is certain to evoke a
st rong negative response. The first rule in composing a title is to achieve
idings. The tough test is th~
132
·wri ting the Proposal
reasona bk parity between the images evoked by the ritle and the
0 Pel\i'1Q
pagl:.) of the proposal.
For the gradu ate student, the propo sal title may well become the ti .
scir
disserta tion ti.tic an<l therefore calls for careful consideration of all th'es,
rions it m ust serve and the sta ndards by which it will be judged. l~ fti~,
fut1<.
:tion of the title: is ro iJentify contenr for the purpose of retrieval. { 61'\i
and di.sscrtation s are much more retrievab le rhan was ooce the case 1le 3e\
1
they have become a par.t of the public domain of the scho lar. The i~ct f~ct,
use of the Internet has made the circulation of unpublished documents ~ 111
&
rimes faster and far broader in geographi c scope. TitliJ1gresearch has becrllan}
ther eby, an important fac.:ro
r.in shari ng research.
on,~,
fo less sophisticated rimes, titles could be carelessly constructed and the d
~menrs _woul~Istill_be discovered by diligent researchers who cou~d take:~
tm1e ro 1nvest1gate items that appeared only remotely related to their interes
Toda y, scholars snigger under the burden of sifting through enormous and co~;
staml y increasing quantities of material apparently pertinent to rheir dorn/ ·
There is no recourse other than to be increasingly selective in choosing Whi~
documents ro aettially retrieve and inspect. Hence, each tide the researcher
scans rnust presenr. at least a moJe rate probability of being pertinent on th,
basis of the title alone, or it will not be included on the reading list for review
ln short , the ck.-grecto which rhe title communicates a concise, thorough, and
unamb iguous picture of tl1ccontent is the first factor governing whetl1era given
n:po rr wi.11
enter the ongoing dialogue of the academic community.
Word selection should be governed more by universality of usage than by
personal aesthetic judg ment or peculiarly local considerations. Some computer retrieval systems classify tides according to a limited ser of keywords.
As we discussed in Chapter 4, researchers construct search plans that will
identify all studies categorized by keywords known to be associated wich
their area of interest. Thus, both readers and writers o( research repo rts must
describe the research in similar tcnns or, in too man y instances, r.ht:ywill not
reach each ot her.
The tirle shou ld describe as accurately as possible the exact nature of the
ma.in elemerm; in the study. Although such accuracy dema nds the nsc of spe·
cific language, the title shoul d be free of obscure techn ical tl'r ms or jargon
that will be recogn.iied only by small groups of resc<1rchers who happen to
pursue similar quesrions within a narro w band of th e knowledge domain.
Components Appropriat e for Inclu sion in th e Title
What is included in a sound tide will depend first of all on the type oi
research involved. The elements most commonly con sidered for inclusion in
Style and Form in Writ ing the Proposal 133
the images evoked by the citl
e ao d th.e
rirnental st udy, for example, are th e dependent and inde-
,
· 1 e)(pe
b
'
t he proposal title may
rlt' ol :.ii
the performance component represented by the criterion
11
I
c
we ecome tL
11(II
b cl . .
1
1 ri·ib es,
lls
•re ca ro1·carefu l cons·d
. of 11etk.
1 1.nr v: ' 1 e treatment or treatments to e a mm1stered , t 1e model ]
J,
eration
,,,l
ks
ri
.
.
.
.
.,A,
.,
st~ndards by which it will b .
a 11~~ 1"110 r ~:is ' 5•cudy, the _eurpose of the stu dy (pred1ctmg, establishing rela -
efiudged. 1\1 1 rJ'Jcrl>'jng rhe. , ining differences, or describing a setting ), a special ized envi1 11cr1 • • dereJn
. ..
.
1pose o retr ·
10
re retnevabl e than w
ieva1_
'h
~h1rs, h"ch rhe research was conducted (e.g., health care fac 1hcy), and
'I tin11
.1n w i .
.
as once the ,
case, 1.
, n 1cJ1C
·ontribution of the study. In contrast, titles for qualitative field
,e public domain of the 1
·/1 . I .
sc 101ar. The 1. 'fl r111 is1ia1c
.
d i. d
.b
. .
.
e cu·cu at1on of unpubJ· 1 d d
llct;., ,rtl 11nt l
give pronunence ro wor s ttlat escn e part1c1pant s, physica l
cl
. uJ
h
··
1s1e ocum "t • . 0 rcen
geographic scope. Tier .
enrs n1d1t!
S.
ncext, and 1e par.tic ar researc rradmon that frames the
1 co
n shar1· .
mg ieseard1 has b II 5 11I ~oc1t1
ng research.
~ .11 • f daca.
O
ties cou ld be carelessly
11n,
t1>'51s . auchor can, by carefu l selection 0£ words , prov ide information in
. d b ..
constructed a, 1d
\ dt:vei '
.
.
.
.
.
e
Y diligent rese.u·che . h
th~ . 1 hat a theory LS being tested by usmg a word that often JS assoqatecl
O
>peared only remote ly J.lswd could tak/1hrr,cle r theory. For instance, the title "Genera lizability of Contingency
re ate to the· . 1.
·rh r 1e
.
.
.
.
te burden of sifting t!u·ou h
It ntert\vi , .,.ement and Rernforcemcnr
m Second-Grade Special Educauon
1
11 t ·a1
g euormousa d '·1111• 1:>
t1 h ·
·
·
· h
L" bT
fb I · 1
roaben. appa1·endy perrineot to their dn O ~ l:isse.'i" implies f_1art e 1n~esttgMator1s cest1bngc e app 1ca_1 1tydob ~ 1av1d~i:a
1
1
1
0/b,
ro a speci 1cpopu at 1on.
uc 1 ms een communicate
y me 1u mg
e mcreasinn-ly se/ , t· .
1
o
tc 1ve tn choos·
ihcorv
111g
11{
,-'
ngle
word
"generali:zabil
icy"
in
the
tit
le.
~nd inspect. Hence e h . I
d
, ac tit e the
r11~ ~1
) erace probability of b .
. resea,~ The ulrjmate purpose of the study in terms of predicting, estab lishing relalOt be included on the . endil~
pe,~tinent OQ,
iships determinmg d ifferences, or descnbing a setting can be expressed
nOJ
,
·I
rea no ist fo
r re~ 1·chout providing an exp licit statement. Fo r example, when variables are
tit e communicares a
. t>
h
w
.
concise
tho
1 t JS the first factor gover .
'
roug ,~ expressed in a series, such as "Anthropometrics, Swimming Speed, and
1
>gue of the academic co ning ': ler/,er ag,1 Shoulder-Girdle Strength," a relationsh ip generally is implied. If the same
·rned more by univcrsa:nu;ity.
.:,wdy were titled "Anthropometrics and Shoulder-Girdle Strength of Fast
•eculiarly local cons·d ty_o usage thanj and Slow Swimmers," the reader would anticipate a study in which differl erat 1ons Some
,
d
. cl
I
es according ro a lini · d
·
C(X cnces were etermme .
ite sec of ke
.
.
.
.
Y.WOtf Any aspect of the srudy that 1s parncu la1·ly unusua l m terms of methodolsearchers construc r .. ,_ I
k
seatca pans th
at II: ogy, or that re resenrs a unique contribution co the literat ure sho uld be
· eywords known t b
aders and writers of
e _a~sociated 111 included in the tirJe. A treatment
at is unusua lly long or of great magnitude
ns or, in too many i searc reports mu(e.g., "longitudina l Analysis of Human Sh.on-Term Memory From Age 20 to
.
.
.
nstauces tl1ey w1'l/ \ 80"}
I d f
'
IX f ge
, a met 10 o · observar.10n that 1s creat ive or unusually accurate (e.g.,
ra tely as po 'bl 1
"Hand Preference in Te lephone Use as a Measure of Limb Dom inance and
•l
ss1 e t 1e exact nature of lb
Lareral
iry"), a samp ling teclrnique tbat is unique (e.g., "Inte lligence of
' l such accuracy demand ti
·ee of obsc .
. s . le use ofsp: Children Whose Parents Comm un icate wit h Hand -Held Devices" ), and a
u1e recJ1n1caJ tet·o1s .
11
or Jarga particu.lar site for measurement that sets the study apart from ochers (e.g.,
gro ups of researchers who I
rrow band f h k
·
1appen 11 "Perceprnal Judgme nt Lna Weightless Environment: Report From the Space
. o t e ·now/edge doma in.
Shuttle") are examp les of such aspects.
r
itify comenr for the pu.
4
r:
Inclusfon in the T'tl
1
e
J
Components [nappropriate
for Inclusion in the Title
~ill depend first of all on rll.c type o'
commonly
·J
Stich factors as population, research design, and instrumentation should
· consi ercd for inclusion in not he included .in the title unless they represent a substantial departure from
134
Writing th(\ Proposal
similar studies. The o ulation for instance shou ld not be noted u
is a population never sampled before, or is in some way an unusuat"1essit
, __ gJ:OJ.:t
12,,.Inthe title "lmbedded Figures Acuity in World-Class C ess Mastat~Ct
the popu lat!oo of th~ subjects is critical to the rationale for the stud/ei:s,
populat ion m "Running Speed, Leg Strength, and Long Jump Perforl'tl1 ht
of High School Boys" is nor in1portant eno ugh to occupy space in the / 11'4/
S. il 1
·gn an cl msrrum.entat1on
·
· are not appropr iat ttle·
101 ar y, researc h des1
inclusion in the title unless they represent a n unusual approach to rneae fQr
menc or ana lysis. The type of research method expressed in "Physiolos~te.
Analysis of Precomperitive Stress" is common in st udies dealing With st&lc~r
and surely some other aspect of the study would make a more infonnaress
.'
comrib .tuion to the title. The approac h in "Phenomenological AnalysistJve
recompetitive Stress," howeve.r, is unique and signals the reader that
eport contafos information of an unusua l kind.
e
tr
Mechanics of Titling
Mechanically, the title :;hould be conc.iseand should provide comfortable
reading, free from elaborate or jarring constructions. Excessive length should
be avoided 1->ernusf.
it dilutes the impact of the key elemems presented; two
lines generally should be adequate. Some retrieval systems place a word lirnj.
tation on titles, thus enforcin~ brevity. Redundanc ies such as "Aspects of,''
"Comments on," "Study of," "Investigation of," "Inguiry Into," and ''A,n
i\nalysis of" are expendable. ft 1sobvious that a careful investigation of a topic
will include "aspects of" rhe ropic, whereas the research report has as its entire
purpose the cornnrnnication of "commenrs on" the findjng:s o.f a study. lt is
pointkss tO stare the obvious in a ritle.
Attempts to incluJe all subtopics of a study in rl1e title sometimes rcsulr
in elephantine l'Librics. The decision co include or exclude menti.on of a
subtopic should be made less in terms of an absnacri on, .such as complete
coverage, and more in terms of whether incl usion actually will facilitate
appropri,irc retrieval. One usdul way ro construct a title is to lisr all the
elements that. ~eern.appropnar.e for ineh1sion, and then to weave them inro
various permmations until a tirlc appears tlrnt satisfies bor.h rechnical and
aesthetic standards.
?peated measuresNlANOVAs
factor and time as the within.
lucted co examin e changes in
icacy and fear of falling, anct
s a function of the exercise
PROPOSAL
I be conduc ted co investigate
r stre ngth and balance with
:>tbeses 4a-d), and ADLs and
::m analyses wi11be employed
relation ship between muscuin ADL performance, fear of
han ge scores fo r th e psychoing the pos t-pro gr am scores
; pre -po st ch ang e scores are
· po ssess high reliability , Sllch
2
QualitativeStudy
Teachers Who Make a Diff erence :
Voices of Mexic an Americ an Students
;e over time in psychological
Note to the Reader
well as the appendi
ces noted
=orms
iving and the Record Formfor
The propo sal that follows involves use of th e qualitative paradigm for research . If yo u
are unfamiliar with this kind of inquiry, a prior read ing of Chapte r 5 will help you to
unde rsta nd t he part icular prob lems that w1~re confronted by the autho r, both in planning the study and in prep aring the proposal document . Also. if you are familiar o nly
with experimental and quasi-experimental designs, Chapter 5 will explain what may
seem to be unorthod ox or, at least, unexpected ways of handling some elem ents in the
prop osal.
As you read , it will becom e increasingly clear that the autho r's assumptions abou t
the nature of such matters as reliability. objectivity, validity,replicabilicy,and gener alizability differ from those made by investigator s using trad itional models of quantitative
:e Confidence(ABC}Scale,and
AUTHORS' NOTE : The o riginal of this proposa l, used with perm1ss1on, was
prepared by Belinda .J. Minor und er the dir ectio n of Profes sor Linda C. Wing, in
partial fulfillmenr of the requirements for the Ed .D. in the Graduarc Schoo l of
Educat ion at Harvard University . The ens uing do ctora l dissertati on was comp leted
and accep ted by the Graduate School in 1997 . Dr. Belinda J. Minor, rhe author ,
now lives in Ca lifornia wher e she is a school admini strato r and pare-time lecturer in
administration and teacher developmem.
253
2:i•I
Sped men Propos,,b
science. Altho ugh those differences som etim es are mo r e appa rent than re<1I,th
exist. and they are vital to an underst anding of qualitative rese arch .
ey do
The special and uniquely valuable powers of the qualitative paradigm resid
cisely in how It is different from o thar forms of inquiry. Changing the starting e Pre.
assulll
t ions allows research questions of a very different order. In the following Pro P·
for example, the auth or lays o ur a plan for askinga group of senior high school st Posa1,
Uden
what they think (and fee l) abou t the role played in their school careers and lives b ts
ticular teach er s. The research questions served by that process demand much y Par.
than just reconstr uction of events, o r the simple collection and tabulation of
ions. The author is pro posing to ask the participancs to reflect upon and discuss Pn,.
per ceived influence of t eachers on their motivation, academic achievement, perslste tlie
in school , and aspirations . Quest ions of that kind, and the data collected to serve thnee
are quite unlike what one expects to encount er in a quantitative study.
el\1,
To che exte nt , r.hen, that qualitative researchers start with different assumption
about some aspects of inquiry, their proposals likewise will display some distinctivet
diffe rent characteristics .At bottom, however, there must be rules of thought and pr:.
ced ure that e nsure chat q ualitative designs represent research chat is systematic , trans.
parent, rigor ous. and faithful to che demands of Its own paradigmatic assumptions.
Proposals for qualitative rese arch must allow a clear discinccion between what is science
on the one hand, and what is no more than careful reportage, thoughtful observation:
or connoisseurship, on the ocher.To have different ru les for inquiry is not the same as
having no rules for inquiry. For that reason, some investment in prior study will make
your reading o f the following proposal both much easier and far more profitable.
Finally,we want you to take par ticular note of the fact that the following document
was selecte d for· use here beca use it represents a relatively early stage in proposal
deve lopme nt .Authored by a well-prepared and highly motivated doctoral candidate,
and guided by th e efforts of a comp etent and helpful graduate advisor (and thesis
comm ittee ), che draft see n he re was subsequently revised into a final form that
received prompt appro val. For our reader s, however. the particular advantage In this
do cument lies in t he opportunity co look ove r the shoulders of th e author and her
advisors during the pro cess of tweak ing the propose d stud y t owa r d a finished level of
polish.A s they did, you can work at th e task of con sideri ng additio ns and alteratio ns
that would furt her str engthen t his st rong initial effort.
a
:~re
Teac her s \.Vho ~1akc a Differen ce:
Voices of Mexica n American Students
fitle for Retrieval. Granted,thisis no more thc'.l
n the tent;.itivetitle used for ihe proposal,
but this is the place to start crafting one that will best ser11e its primary function-retrieval of the study by potential readers. The present title clearly identifiesthe two
Q\.lalitativc Study
·e more apparent chan
· .
real h
1ahcat1veresearch.
·t ~ d
:he qualitative paradigm
.
Cl
. C
resrd
e Pr11
.
1wry. hanging the Startin
lt order. In the follow · g asst,'n"
Ing Pro
.,.
:roup of senior high sch
Posa1
h .
OoI Stt,d •
err school caree rs and IIYes ents
0
· that process demand
YPar.
rnuch
: collection and tabulat·
more
IOI) Of O .
its to reflect upo n and d'
Pin.
.
rscuss •L
aca dem1c achieve ment
. ,..,e
, persist
d the data collect ed to s
=ence
erve th
3 quantitative stu dy.
em.
; start with different assurn .
·
.
Pt1on
'llrse w, 11display some disc· . s
rnct1ve1
y
nusc be rules of thought and
re.search that is systemati
Pro.
c, trans.
s own paradigmatic assumpc·,
· . .
ons
istrnctron betwee n what is sc·
.
1ence
·eportage, though tful observati ,
lies for inquiry is not the sameon,
•estment in prior study will m kas
.
ae
srer and far more profitable
fact_chat the following docu~ent
·elat,vely early stage in proposal
y motivated doctoral candidat
ful graduate
advisor (and the s~
~·
.
revised into a final form that
the particular advantage in this
houlders of the author and her
study coward a finished level of
dering additions and alterations
ts
t'
.
.ivetitle usedfor theproposal,
serveits primaryfunction:it/edearlyidentifiesthe two
255
o(JpS
thatwill beof centralinterestin thestucly:(a) Mexican
Arneri<:an
studentsand
ceachers.
Furthermore,
althoughlessexplicit,thewar.ding
wouldleadmostreaders
to assumethat thestudy':""illinvo!ve.quotedmate_rial
of somek!nd fromstudents,as
wellas an effort to describethe influenceof their teachers.Brief.nonpedantic,
and
nicelyphrasedto seemintriguing,the title maywell serveto attractinterest-but can
dosoonlyforthosewho readit!
It is importantto remember
thatin mostinstances
titlesarereadonlywhentheyare
retrieved
fromsometypeof index,whetherfroma data-based
computerretrievalsystemor thepagesof a hard-copy
volume.
Whenindexingis basedon keywords
assigned
bytheauthor(ora reviewspecialist),
anartistictitlesuchastheoneusedherecanserve
perfectly
well.When,however.
theassignment
of thestudyto indexcategories
is done
on chebasisof thewordsactuallyappearing
in thetitle,whatwasaesthetically
attractivemaybecomefunctionally
inadequate.
In an indexbasedon title wordsalone,personssearchingfor qualitativestudies
of studentlife in secondary
schools,investigations
of studentachievement
in high
schools,researchemployingphenomenological
interviewingwith adolescents,
or
evenstudiesthat identifyvariablesrelatedto teachers'influenceon studentswill
be unlikelyto retrievethis study.Research
method,socialcontext,schoollevel.and
primaryvariablesmay be inferredby the presenttitle, but they are not actually
specified.
Byall means,
writegracefultitles.Besure,however,
to usethekeywordsthat might
be importantin placingyourstudyinto indexcategories
whereit canbe noticedand
retrieved
by peoplewhowill find it valuable.
Oursuggestions
in Chapter6 concerning
thedesignof titlesshouldhelpyouto achievethatgoal.
A standardtitle pagewasomittedhere.
r~)
Abstract
This qualitative study will describe and analyze the perspectives of Mexican
American high school seniors regarding the influence of their teachers on
their learning, school careers, and lives. The theoretical framework undergirding this study is "effective teacher" research. The research design will
involve "in-depth phenomenological interviews" (Seidman, 1991), focus
groups, and observations involving a primary sample composed of twelve
Mexican American seniors. In addition, a second sample.:of six will be interviewed in a separate focus group as a "member check" (Guba, 1981). The
research will have implications for educators who want to make a positive
difference in Latino students' learning, achievement, an<l persistence to high
school graduation.
256
Specimen Proposals
Theoretical
Framework
. In termsof the definitionwe haveusedthroughoutthl b
a bodyof research
is not,in Itself,a "theoretical
framework."
Theauthorappat 00~.
usingthetermin its moregeneral
senseto Indicatethesourceof theoretical
conently
1s
andempirically
basedassertions
thatformedthebackground
forde-signing
thestu~t
ru1:1s
developing
theproposal.
It is ourstrongpreference,
however.
to reserve
useof th/ and
theoretical
framework
to designate
a particular
(andformal)theory,or networkof relterni
problemandi afed
theories,
usedas thesingularvantagepointfor defininga research
pretingthe ~esults
of anensuingstucjy
. In thatsense,
then, a theoretical
framewo,tercoherent
setof postulates
aboutthenatureandfunctio
ningofsomeaspectof thewo~~d~
In light of de.mographic changes in the United States-distinguis hed b .
large, young, and quickly growing Latino' popula tion-coupled with a~;~
ahl.eand historically per~ist·emachievement gap between Lat ino and Ang!
students, the education of Latino students is a critical cha llenge facing ou~
nation. The life chalienges confronti ng most high school dropouts, including
higher unemployment and lower wages, as well as the costs to society are welldocurnemed (CatteraJI, l98.5; hne, 1991; U.S. Department of Educarion
199.1).
'
Since the pubiicarion of A Natio n at Risk by the Nationa l Commission
on Excellence in Education in 1983, nationa l leaders have exhorted
schools to eliminate rhe achieveme nt gap and to educate ALL children
to high standards. I {owevcr, many teachers today question their ab.iJity
to rncct this goal because they locat e the source of students' academic and
rnorivational problcrns in their famil ies, peers, SES, cul ture, or personality and believe that, in relation to those external factors, their influence is
of little consequence. (Clark, 1990; Grant & Secada, J 990; Harris, 1991;
Hidalgo, 1991; Ladson-Billings, 1994; Paine, 1989}. ·rragically, teachers'
self-perceived lack of "cfficacy" 1 may actually contribute to the low achievement of srudcncs (Ros~, 1945).
Research is needed to help educators appreciate and understand rhcir
influence on I ,atino sn,dcius' achievemeut. Despite the numerous factors
r.hat are, or seem to be, beyond their conrrol, teachers must understand how
they can use tliat influence in ways that make a positive difference in their
students' lives.
This study will contribute to meeting this need by responding to the following research questions: Wlrnc do Mexican American 3 high school seniors
report as the sal.ient qualities and behaviors of teachers who have made a differen.::ein their lives? How do the studenr.s d1aracrerizc r.heinfluence of these
teachers 011 theix motivarion, learning, ac.1dernicachieverneru, persistence in
Qualitative Study
? haveusedthroug
hout
th·
15
nework." Theauthorap book,
Pare
nt1
y ..
he source of theoretica
l
rs
constru
Jroundfor designingthestud ~ts
owever.to reserveuseof
and
,rmal) thec:,ry.
or network terrn
.
,
oI related
mg a researchproblemand.
r.
then,a theoreti.cal framew Linte
•
.
o~ ~
irnng
of someaspectof thewor
. l.d.
a
________
th/
:ed States-distinguished b
pular.ion-coupl ed Wit! ya
.
l as~
tp betwee n La tm o and A
1 critical
cha Uenge facin nglo
g our
pI scb oo.I dropouts, includin
~s the costs ro soc iety are wel
,. Depanm ent of Education
'
>y the National Co mmi ssion
onal lea ders have exhorced
rd to educat e ALLchildren
toda y question their ability
e of students' academic and
;, SES, culture, or personaltal factors, their influence is
;cca<la, 1990; Harris 1991 ·
'
'
19 89). Tragically, teachers'
>ntribute to the low achieve!ciate and understand their
·spite the numerous factors
chers must understand .how
1 positive difference in their
:d by responding to the folnerican3 high school seniors
ichers who have made a dif:terize the influence of these
achievement, persistence in
257
and aspirations? 4 In what other ways do pivotal reachers influence
11
00'Is
sc d o.rs' Jives?
;tLI
e.
bstract
. Notall graduateschoolsrequirean abstractin a prescribedformatfor dis-
Atation proposals(abstractsinvariablyare requiredin grant proposals).Because
s~:paringan abstractfor the proposalprovidespracticeon a vital skill (see our
~rscussion
of abstractsin Chapter9). we are muchin favorof the requirementand
suggest,that you developone whether or not it must be includedin the formal
document
,
It wouldbe inappropriate
to attempta critiqueof the proposedstudyas presented
in the abstract
. Technical
detailscanbe judgedonlyin thecontextof theirexplanation
.
It is possible,
however,to considerthe adequacyof the abstractas a mechanismfor
accomp
lishingits primarypurposes:
(a) concisecommu,nication
of the study's purpose,
method
, andjustification;and (b) suggestionthat the findingswill containsomething
of specialinterest.
In this abstract,the writing is crisp,graceful.and a modelof efficiency.Thefirst
paragraphtoucheson purpose.method.and justificationwithout use of jargon or
technicaltermslikelyto be unfamiliarto reviewers.Useof footnotesand citationsin
an abstractwould not be our preference.but they do servea clearpurposein this
instance.
Threeof the five paragraphs
areemployedto developfurtherthe argumentfor the
importance
of the study(and they do so very effectively).
That,however,leavesonly
theopeningand closingpardgraphs
to dealwith a formalstatementof the research
questions-andmethodology
. Necessarily.
a greatdeal mustbe left out. Twoof the
missingbits of informationare amongthosethat mostreviewerswill be lookingfor:
(a) methodfor selectionof participants
and(b) m!:!ans
of dataanalysis.
Althoughwe can
anticipatethat thosewill be specifiedin detailin the subsequent
bodyof theproposal,
mostreaderswill expectsomebriefindicationin theabstract.Evenif thereisa wordor
pagelimit on the lengthof the abstract.that smallexpansionshouldnot be difficultto
achieve.
Theopeningsentences
of thethird paragraph,
for example,mightbe sacrificed
withoutseriouslossasa wayof providingspaceto identifythe typeof participantselection anddataanalysisbeingproposed.
Headings.
Notethat the useof an underscored.
centeredheadingis not allowedin
eitherAPAor Chicagostyle.Unlessotherwisespecifiedby the graduateschool(asmay
havebeenthe casehere)or fundingagency,it is best to employone of the heading
stylescommonlyusedin the disciplineor professionalfield of the study.Thecorrect
numberandtypeof headingsin theAPAformat.for example,caneasilybe established
hyconsultingthe "Levelof Headings"sectionof thePublicationManualof theAmerican
(2001). Headings
forsubsequent
sectionsof thisproposalhave
Psychological
Association
beenrevisedto conformto APAguidelines.Pagenumbershavebeenomittedin the
tableof contents.
Sp~cirnen Propos~ls
258
Table of Contents
Abstract
Table of Contents
Introduction
Background
The Problem
Literarun; Review
Metho<l
Site
Sample
Access and Sample Selcaion
111
tervievvs
Observ:nions
Focus (;roups
T rusrworthi ness
Ethical Conccrus
I,imita rions
Data Analysis
Conclusion
References
Appendices
Appendix
Appendix
Appendix
Appendix
Appendix
Appendix
Appe ndix
A-l: Latino Demograpbics
A-2: Latin0 Academic Achievement Statistics
A-3: Tables Showing School Achievemenr Gap
B: The Latino Achievement Equation
C: EffecriveSchools/EffectiveTeaching Research
D: 85 Variables Associated With Effective Teachers
E-l: "Ca ltui-ally Relevant" v. "Assimilationist"
Teaching
Appendix E-2: Defpit's Summary of Atrribures of Good Te.1chers
Appendix F: SJmple Select.ion
Appendix G: Sample and Methodology
Appendix H: Qw~srionnaire for Prospective Participants
Appendix I: Interview Guide
Appl'ndix J: Observation Gui<lelincs
1\ppcndix K: Sample Codes for Analysis of Data
Qualitative Study
2.59
Contents.
A glancethroughthe tableof contents
will remindyouthat when
fable01. refullydone,as is thecas.e here,tablesof contentsservenot only asa way
weyare
laparticulartopicswithin the proposa
l documenf
but alsoasanexcellent
outof10Gadn!t1,eauthor hasdevelopedthepresentation
.The14appendicesindica
te that
li~e
oft~~rhastriedto kee~the maintextas~ompa~~~d tightlyfocusedaspossible,
t~eaucing
supportive
mo.tenals,complex
technical
def1n1t1ons,
forms,andprotocolspec
releg~s to attachments
thatcanbeconsulted
asneeded.
Because
virtuallyall readers
ifi~ano~t
to readthemateria
l on sampleselect
ion,howeve
r, we thinkit inappropria
te
w\l~
wae
thernto flipbackandforthbetweenappendices
(FandG}andthemaintext.
.
10,ore
ndices
.
~pp_e
____·arefor non-essentials
__
_ _only
_________
_ __
_
lotroduction
Background
cs
:hievement Statistics
ool Acl1icvemcnt Gap
nt Equation
tive Teaching Research
·d Wit11 Effective Teachers
H
"A .
v.
ss1rnilationist"
Artributes of Good Teacl1ers
gy
>ective Participants
sis of Data
This qualitative study will c.Icscribe and analyze the perspectives of
Mexican American high school seniors regarding the influence of their
ublic school teachers on rheir learning, school careers, and lives. The theo~ecicalframework undergirding this study is "effective teacher" research.
The researd1 design will involve "in-depth phenomenological interviews"
(Seidman, 1991), focus groups, and observations involving a primary sample composed of twelve Mexican American seniors. ln addition, a second
sample of six will be interviewed in a separate focus group as a "member
check" (Guha, 1981). The research will have implications for educators who
want to make a positive difference in Latino students' karning, achievement,
and persistence to high school graduation.
The Problem
In light of demographic changes in the United States-distinguished
by a
large, young, and quickly growing Latino population-coupled
with a sizable and historically persistent achievement gap between I ,atino and Anglo
students, the education of Latino students is a critical challenge facing our
nation. Latino school enrollment has doubled over the past twenty years.
Historically, Latino students ha vc attained lower standardized test scores 1
and have suffered lower high school completion rates and substantially
higher dropout rates than Anglo and African Americans (supporting data are
presented in Appendix A). The life challenges confronting most high schoo.1
dropouts, including higher unemployment and lower wages, as wdl as the
costs to society are well documented (Cattcrall, 198.5'; fine, 1991; lJ.S.
Department of Education, 1993).
260
Specimen Propos~I.,
Since the publication of A Nation at Risk by the Na tional Con, i .
Excellence in Educarion in l98 3, nati .onal leaders have ex horted 01
scfs1Qi1 (Ill
diminate the ad1ievement gap, a11d to educate ALL children to hi 1~Qlsto
dards. However, many teachers today question their abi lit y to meet~ - sta11.
because they locate the source of stud ents' academic and motivationat'S &oar
lems in their families, peers, SES, cultu re, or personality a nd believe thProi).
relation to those external factors, th eir influ ence is of little conse at, iri
(( ·1 k · 9c . c··.
0 1
· ' 199'L
'd a 1go, J 9·91; Lads
%enec
: '.1r , 1 .90; .)f~nt &··, Secal ·!~' ·t
. -~~7 0 ; H ams,
, ; H 1.
l3rllmgs, 1994; Pa,ne, I 989). 1 ragrcall y, teachers lack of self-perceived,, 0ncacy" may actually contribute ro th e low achievement of students (/ffj _
1994).
0 ss,
Research is needed r.o help ed ucat ors appr eciat e and understand th,·
influence on Latino srudents' achi evement and to und erstand how they Cit
f
· wavs ro ma f,c a pos1t1v
· · e d'ff
· stu dents' 1· can
use t 1at ·m fl uence m
l .erence m
. ,
1~j
despite the numerous tacrors thar. ar e, or seem to be, beyond th eir conrr
(see Appendix B).
0
This study will contribute r.o meetin g th is need by respondi11g to the following research questions: What do Mexican American l,jg b schoo l seniots
report as the sali.cnt qualities and behaviors of teachers who have made a difference in tlie.ir lives? 1-low do the stud ents charactecize rhe influen ce of these
teachers on rheir motivation, bu-ning , academic ac hievement, per sistence in
school., an<l aspirations? In what oth er ways do pivotal teac hers influence
srudcnrs' lives?
lntroducrion.
/\t f:rsi glance,thi~ iniroduoory sectio;·appearsto be J repeatof the
abstrr.ic~.
" tactictl:at, in itself,is perfectly,icceotableso Ion()as both texisµerlormthe
functionsneeded,it e.ich locat;on. CloserreadincJ.
however,will revealtr.at thereis
,l<1ditio11a:
derail here,nlon~1with Chi:lnge5
in wording and reierencesthat lead the
reaoerto supp!e1rP.nta1y
11:ate1
ial in tt,eappendices--ausignalsthat Wf! now arein the
•nainbodyof the propnsai.
RPsea
tch Questio11s
. BeGlcJSi?
researchquestions,ire sucha centrJ/elemefltin a proposal.\/•.'C'would preferthat theyalwc1ys
be listedand numberedin serialorder,rather
than!ie embeddedin the text of a paragraph
. fhat helpsthe readertreal the,nasseDarate i'ntities,each1..vithits O>Nn
denw1dson methodology.
5ampll'. Althoughtile main exposilionis yet to come.this introductionwill ,iiuse
most·eadersto postca11tio11
flaqsat threepoints:(a) Howwill rhe samplegroupsbe
selected,(bl Will l 2 pc1rticio.ints
be sufficientiq numberto servethe relativelyambi,rnd (c) Howcan ,ix differentpartici:iousgualspresumedby the researchouesticms?
pants~e -:sec!asa ''membercheck"whenuseof ihat termgenerullyis takento m!?a'I
a confirrnat1on
of data by the original sou;ces7
Qualitative Study
261
sk by the Natio nal Corn .
"'11"1
1ss1
I Ieade rs have ex hort ed
0,\,.
S
ch
0
Iucace ALL children to h. olsult
t
. h .
1ity to "'e tgh St-ah0
st1on t eu:
abi
.
,
"·' et th. .,,
academic an d motivati'
IS&0,1
.
.
onal ,.
01 persona!tcy and belie
Pto!)_
. f
Ve th
m luence is of little co
at,i
.
nsequ 11
ns, 199 l; Hidalgo 199 ·l· L enco
I
> ac1s
tchers' lack of self-perceived,, on.
' achievement of st udents effi,
(Ross
cions.
Finally,thereis an opportunityat this pointto testyourownabilityto
~um~rougha problemin studydesign.In formulating
theresearch
questions
above,
think
;thorhasmadea veryimporta
nt (thoughtacit)assumption
aboutherparticipants
tb~a bouttheirexperience
of schooling
in particular.
Wedo not referhereto theusual
anu~ptionsaboutmethod(participantswill volunteer,will not deliberatelymislead
ass·nierviewer,
etc.),but to howthe studentsunderstand
what happened
to themin
1~~ 1 1.Aswith Poe'sfamousPurloined
Letter,whattheauthorhasassumed
is soobvi0
sc 0(and takenfor granted)thatit easily becomes
invisib
le.Exami
nethe research
ques~uss ,arefullyandwatchfor furtherdues as you read-and leaveour "Posts
cript"
):i1owing
theproposa
l untilyouhavefinisheda complete
readingof thedocument.
'
appreciate and understa nd th.
and to w1derscaod how tJ
e,r
. . d'f:c
,ey ca
:rive l .lerence in stude nrs' 1· n
.
,eern
ro be, beyond their co ives•
ntrol
is need by responding to ti r
.
.
le tOJ.
an Ameacan high school se1llors
.
o f teachers who have made d'f.
b
a 1aracterize the influence of th
.
h'
ese
em 1c ae. ievement, persistence in
ys do pivota l teachers influence
:ionappearsto be a repeatof the
le so longasbothtextsperformthe
, however,
will revealthat thereis
·dingand references
that leadthe
signalsthatwe nowarein the
-au
aresucha centralelementina p,0 .
nd numbered
in serialorder,rather
ielpsthereadertreatthemassep·
dology.
come,this introduction
will cause
a) Howwill the samplegroupsbe
mberto servethe relativelyambi·
d (c) Howcansix differentpartici·
3t termgenerally
is takento mean
Literature Review
Sincethe turn o.f the century, educational researchers have conduc ted many
investigations in an attempt to find ways co define and measure attr ibut es of
"effective" education. Nevelitheless, no consens us has emerge d concerning
either criteria for or qllant ification of t hat elusive con·struct.
Three lines o.f inquii-y have dominated such studjes : "effective schoo ls"
and "effective educational practices" re.search (Appendix C), and investigations of "effective teachers" (Appendix D). As explained below, the study
proposed here will contribute to the latter.
Cruickshank (1990), in his rnmprehensive review of the effective teacher
research from the 1890s through the 1980s, notes two historically distincr
periods of research that differ theoretically and methodologically. Research
on "effective teachers" conducted prior to 1960 used subjectively derived
items and vague terms created by administrators to describe teacher attributes
and behaviors that they presumed to be effective. That line of study proved
to be of little value.
Subsequent to the publication of the so-callc<lColeman R.eport6 researchers
turned their attention to identifying specific teacher behaviors present or
operative when pupils were succeeding (Cruickshank, 1990). Conducted in the
1970s and 1980s, these quantitative studies used a host of new instruments
to record classroom behaviors, which were then analyzed systematically. The
goal of this methodology was "to determine how and to what extent teachers
performa group of precise actions and the extent to which 1,erformingthese
actions is related to othe1· desirable attendant classroom events and/or pupil
learning" (Cruickshank, 1990: 68, emphasis in original).
Crnickshank's table of 85 "promising teacher effectiveness variables"
(Appendix D), culled from his review of ten other reviews of roughly 2.U0
ZG2
Specimen Proposal.~
studies, shows few areas of agreem ent about which variabl,
closely associ<1tedwith effective teaching. Only eight of the 85 va~a:te 1))
011
ciar~·d with effL:ctiveteachers appear i~ at least fom of the reviews. L'~es<lss()
clarny, au:end111ghnomtonng; behavi or, use of more pupil Parf _eseatt
. . . structure, ttme-on-tas
.
I< persistt<:tt}af
equitao eness o f pup1·1part1c1pat1on,
' lon,
efficiency, use of ieedback, and crit icism (the latter is negatively enl.ce illld
re ated to
effectiveness).
One of the main lirnitations of the quantitative methods used in th
studies is that, because they arc h;lsed primarily on observations, th: above
not explain why cerrain teacher qu alities or behavi.o.rs appear to be y can.
·1·1
.
between what stud
ass0ci•
ate d wit,·1 Stu( Icnr outcomes.
rJe connections
perceive teachers ro be doing and the students' motivation and achieve ents
remain invisible to r.hf.observer and must be specu lated by hiiu or her.Il1Cnt
Researchers who study "effective teachers" do not disaggregate data
examine what kind of teachers anJ teaching are effective for students ft to
differcnr ethnic or cultural background s. Instead, when they differenri~;1
among student groups, they use tf.rms such as "low SES," "disadvan taged;
or ''at-risk,'' w refer to poor student s of all ethnic backgrounds. The exte~r
ro whid1 these or other variables are salient for effective teachers of Latino
srudems rema.ins unclear. The proposed study will focus on "effective teachers~
for one somewhat heterogeneous (Matutc-Riand1i, 1991; Suarez-Orozco &
Suarez-O;ozco, 1995) cultural group: :'.'v1exicm,Americnns.
One large, 4w111titativestudy not mentioned in Cruickshank's review
(possibly because ir. involved teacher ineffec:tiueness) was conducted by the
U.S. Comin.ission on Civil Rights (1973 ), ,rnd focused specifically on the
behaviors of teachers with Mexican American studenrs. The investigators
used the Hmders htterncrion Analysis (Amidon & Fl.anders, 1%3) to code
observations of 494 classrooms in 4.,0 southwestern schools. They found
that teachers were "failing to involve Mexic;:in 1\merican children as active
particip8nts in rhe classroom to the same extent ;is Anglo children." The
i.,
study reported:
Comhini11g all types of approving or ,Kcepting te,icher behavior, the teacher.;;
t·espondcd posirivdy to Anglos ,,bout 40 percent 111orethm1 they did to Chirnno
students. Teachers also directed qttestions to i\ng.lo stucicnr., 2.l percent more
oite.11than they directed chem to :Vlexic,111s
{U.S. Conimi.ssion on Civil Rights,
1.973: 43).
Clearly, several ot the behaviors of "effective teachers" were withheld
from lVIcxican American swdcnts. Low Mexican American achievement
was ,1ssociated with 1·ead1ers' inequitable r.reatmcnt of the students, but r.he
researchers inferred tbe c.1us<ilconnection betweeu reacher behavior: and
26-1
Spe(irllP.11Propo.~.ils
This body of qualitative research, while stiU in its infancy, describes
effective teacher belrnv.iors-induding listen ing to students, caring, and ; 0 1l'le
.1ng respect tor:
.. stu dems ' cu Irnres-r hat were not note d 1.11
· pnor
· res8 10\1,,Indeed, they are entirely absent from the list of 85 "teac her effecti;arc.h.
variables" enumerated by Cruickshank. The four main foci of the qua(·en~s
.
n:scarch mcluded:
ttan..,
'...
,,
·,
L:stsof Ctatior1~
~avebf!Pno~itted in rhe foJr l,Hegoriesuelow.
-------------
--------~--------
.
.)
I
1) Studies of dropours. These studies reveal, primarily through intervie
with dropouts, how dropouts am ;ibute much of their loss of motivation aw;
lack of achievement to ineffective reaching practices . The students common~
ident ify these practices as: lack of caring and preju .dice against stude-n:
s
of. co Ior.
2) Studies of mfooriry teachers' m.emor ies of influ entia l teachers. These
stu.dies find. qua lities such as: caring that students learn, connection of material co "real life," not moving on in the cnrricu lum until the students have
master ed a lesson, pushing stL1denrsto think, and getting to know students
and their families, to be amibutes of effective teachers for students of color.
3) Studies involving interviews and direct observation of teachers whom
parents, adm inistrato rs, or colleagues deem "effective." The se studies find
that ~ among other attributes, effective teachers of students of color use "cultural ly relevant" pedagogy. 9
4) Studies invo lving etboog raph.ies and/or int erviews with students, individuaUy or in focus groups . These studies, like the studies of di:opouts, point
to the importance of teacher "cari J1g" and "respect" demonstrated through
instructional practices that elicit and val.ue all students' learning.
My study will fir into ch.is fourth category, as it artemprs w understand
Mexican American students' experiences with and perceptions of. rhc influence of te-achers on their motivation, achievement, persistence in school, and
personal lives.
Num,mAiffiburion. ',Nedrawyo11r
at~enlionto the phl'asin[J
in ~heseniencPimmPniately
would lead the readerto beli~vet11a1
the study,an inan·
alJovc.A litc-r,;Ii·llerprC'la,;on
ini,;ie Pntity,i$ ~oing to do somethingthat only an animareentity (a human,in this
uise) canachieve--thc~ti\ to under~,andsornPthing.
We conf~ssthJl this hasbecome
Qualitative Study
: stilJ in its infan cy d
.
.
, escrrbes
'1mgto students car itlg
so,h
·'
, and ••1:
.rere not noted in p.rior
Sho\v
,
. list of 85 ''tea cher ff. research
'
.
e ect,v
.
te rour main foci of the
. e11e_~s
qttal1taf
'"r.1tegories
below.
eal, primarily rfuough i . .
nterv 1e\
:h o f rh cir loss of motiv .
vs
.
at1on a11d
ract1ces. The st udents c .
. .
omn,onf
and prqudICe against st d }'
u ents
es of influential teachers lh
.J
• •
ese
.ients learn, connection of
.
ntate.
rn.:u1um unril the studenr ·s have
:, and getting to· know stud
enrs
c teachers for s tudents
· of' co1or.
: observation of t~achers wh
" f.l'. • ,,
om
e rect1vc. These stu<lies find
"S of students of color use "cul-
interviews with sn,dents, indi~ the studies of dropouts, point
espect" demonstrated through
students' learning.
, as it attempts to under.stand
1 and perceptions of the influent, persistence in school, and
ising_in
thesentence
immediately
l believe
that thestudy,an inanlnimateentity(a human,in this
/1/econfess
that thishasbecome
265
mmonplace
of construction
in·writing aboutresearch,
andit is likelyto remainso.
0
a' ertheless.
we preferto writeas thoughit weretheinvestigator,
or thereaderof the
Ne~arch
report
, who is described
as performinghumanfunctions
- understanding,
re~foring,
theorizing,
<1nd
thelike.Thus
, we wouldmuchprefer:"My studywill fit into
fourthcategory,
and throughit I hopeto betrerunderstand
MexicanAmerican
1 5
~dents'experiences
. . .. " Asyoucansee,this is not a matter of correct
.or incorrect
s~amrnar
(whichmightbe resolved
byapplication
or theappropriate
rule),but rathera
9roblemin the logicof word usage.We think you will serveyour readersbe.tterby
~ttributing
things attemptedto theresearc
her andnot theresearch.
e:.
The only study that has been directed exclusively at Latino students' perspectives on the influence of teachers on their learning was conducted by
Gladys Cappella Noya for hc.:rHarvard Graduate School of Education doccoral thesis, completed in 1995. Noya\ sample includes eleven a<lolescents,
Puerto Ricans and Dominicans, whom she engaged in "collective discussions,
writing, and individual conversations." She found the following themes to be
salient:
... participants' perceptions that teachers play a determinant role in the
demarcation of the <lepth to which students are willing to hring themselves
into the classroom; students' needs to express their doubts, questions and
disagreements; and young people's appreciation for relationships among
students and teachers which acknowledge human dimensions beyond
their respective reacher/student .~tatus, and which transcend hierarchical
boundaries (vii).
Noya found the students in her sample described effective teachers as
teachers who demonstrate caring and respect for students. Participants indicated that teachc.:rs demonstrate respect by listening to student comments in
class without cutting them off or humiliating thc.:rnwith criticism. Students
particularly value teachers who can create a class climate in which everyone
feels free to ask questions and think aloud. These teachers convey a sense
that the srudc.:nts are intelligent people, capable of doing challenging work
and contributing to the construction of knowledge in the classroom. They
enable students to bring their whole selvc.:sinto classrooms and reciprocate
respect.
Furtherreviewof Noya'sfindingshasbeenomittedhere.
2GG
Spec.irncn Proposals
Although my research questions are similar to Noya's, and th
be some similarities in our met hodo logy, the proposed stu dy w ill dif;re \\:>ill
methodologica lly a nd theoretica lly from Noya's in some impo.1.·
tan:r both
First, Noya's met hodology emp loys wha t co uld be considered "ev '-"ars.
.
·
.
.--tre
samp Jmg (Patton, 1990: 169). Her samp le mcludes on ly students wh ~ti.e~
been rejected from traditional America n pub lic schoo ls and stude n~ ave
attend a_relative~y affl~ent scho? J on the !sland of P~ert:0 _Rico. M.y n:i.e
t~ 10
ology will use " 1ntens1ty samp lmg." "Using the logic of mtensity sam . done seeks exce llent o.r .rich examples of the phenomenon of interesr bp1tng,
unusua l cases" (Patton, 1990: 171). My samp le wil l be dra w n from'sru~ not
who attend a typical, n.rban public high schoo l.
ents
Second, Noya's ana lysis does nor consider how students perce·
teachers' responses
the ir ethn icity o.r socia l class. I intend to look for
relevance of these qualit ies for Mexican America n st udents .
e
Thi.rd, and 6nally, since Lat inos are not a unita.ry pop ulation, and Mexica,
Americans may have different experiences and perspect ives from Pucrt~
Ricans, due to differences in regio n, history, and culture (Suarez-O rozco
1995), when contras ted wit h Noya's findings it ls likely that there will be dif:
fercnces in my participants' percept.ions of teachers.
to
;~e
Prior,qcscarrhHere:h:i ,iuthor has introducedil keyrefP.rence-onewith ·Niiich1,er
,idvisorsa;e likely to be iJrniliar.Not only doesshe makeskillful use of the resu,ts to
supporther proposal,but shealsocarefullyshowsilov,;her own wo:'k will inaKc>
a distincti-,ecomribL'lionthat movesbeyondthe prior i11vestigat1011.
l\t1etho<l
The purpose of this research is to describe aod analyze the perspective.~ of
i'vlcxican American adolescents regarding the influence of their teachers. The
ccnmil. questions arc: What do ,\t1exican American high school seniors report
as the s<1lirntqualities and behaviors of teachers who have made a differem;e
in their lives? .How do the students characterize rhe influence of these
teachers on their motivation, learning, academic achievement, persistence in
school, and a~pi1\1tio11s?
In what orher ways do pivotal tead1ers influence
studenr.s' lives?
'fhcse phcnomenologicil w questions will be inv est igated th roug h a qua l.itative ~esearch design that will consist initially of interviews, focus groups ,
and obscrv,nions of eighteen stu<lcms. Lincoln an d Guba (1985: 225) caution that "the design of a nawralistic study ... cannot be given in advance;
Qualitative Study
similar to Noy a's and h
th.
'
t ·ere .
e proposed stud y will d'fc \V11J
, ,
I cer b
N oya s Ill som e import
0th
ant w
lt cou l<l be con sidered «
ays
. I
exrre .
c me udes only students
h 111<.'~
.
w oI
bl
pu 1c scl1ools and stud ·
•ave
ents '1•l
sland of Puerto Rieo M.y
' lo
.
,
· · n1erh0 d
ng t I1e logic of mtensity sa
: mpf1n
e ph enomenon of interest b
g,
·11
b
,
ur
nor
I
imp e w1 e dr awn from s cl
tu ents
.·h00. J
~onsider how students p ..
. r J
.
erce1ve
,c1a c ass. J mtend to look f
.
or the
mencan students.
l unitary populat ion and tyr ·
'
...e.x,can
~s and perspecti ves from p .
uerto
:Jry, and culture (Suarez-Oro·
' ' J'k
zco,
gs it 1s 1 ely tha t th ere will be difteachers.
1 keyreference-onewith whichher
he makeskillfuluseof the resultsto
s how herownwork will makea disinvestigation.
and analyze the perspectives of
e influence of their teachers. The
erican high school seniors report
hers who have made a difference
·acterize the influence of the.~e
~mic achievement, persistence in
is do pivotal teachers influence
be investigated through a quali.lly of interviews, focus groups,
)In and Guba (1985: 225) cau... cannot be given in advance;
2(i7
st e.roerge, develop, unfold ... " (emphasis in original). Thus, where it
it J11Llreasonab le to modify the design, J will do so and wi ll "re p ort fully on
5
secJ11
was done, why it was done, an d what the impli cat ions are for the find~µar
~ s" (Patton, 1.990:62).
iflSA qualitative approach seems to be the most reasonable approach for
weriog these research questions, since qualitative r:esearch is especia lly
1115
• rop!'iate
11PP
for studies where little empirical research ex ists (Patton, 1990).
Site
The site for this study will be a rypical mban Ca lifornia public high
,hoof with a predo.rujnantly Latino student body. l have chosen a predom~naotlY
Latino school because more than two thirds of the Latino smdents in
:he United States are educated in racially iso lated schools (Appendix A-1.).
'fhe site is in California because California has the largest Mexican
Americanschoo l population. I have selected a typical, rather than an "exemplary" school because I want to learn how individual teachers can make
a difference to Latino students in average schoo ls, under typical circ urnsrances. I chose a high school as the site for the research because I am interested in interviewing seniors who are on the brink of "making it" through
chcsystem. Their insights concerning the influence of teachers on their motivation, learning, and achievement may have implications for. supporting
other students in ways that will lead to the same end result.
One site is sufficient for this study because a number of schools "feed"
into the high school and the district has a voluntary school choice program.
Thus, f can select students who previously have heen exposed to a wide
range of teachers anti pedagogy while conducting my study at this site.
Repetition.
Is it reallynecessary
to repeatthe researchquestionshere?Probablynot,
somereviewersturndirectlyto the meth·
but we thinkit a goodidea,if onlybecause
odssection.
Flexibility.Is the secondparagraphnote about necessary
flexibilityappropriate?
Indeedit is in a qualitativestudy,thoughwith a noviceresearcher
the licenseto make
suchadjustments
shouldbe usedwith greatcaution.Theapt quotationfrom Patton
notwithstanding,
in mostdissertation
proposals
it is customary
to makeit explicitthat
changes
will be madewith the adviceandconsentof the primaryadvisor,or,for major
changes.
all members
of the thesiscommittee.
WhyQualitative?
Theclaimfor appropriateness
of the qualitativeparadigmin the
thirdparagraph
mightbetterbe madeby notingthatonlya qualitativestudywouldbe
(Continued)
268
Specimen Proposals
(Continued)
appropriatefor answeringthe questionsposed.Theargumentthat qualitative .
arebestwherelittle is knowncantoo easilyleadto theassumption
that theyattlldies
able only for "exploratoryresearch
." Thatv.iew weakensthe positionof qua~tvalu.
designsas viablealternatives
for anyquestionthat theyfit-irrespective of the •ltlve
of existingknowledge.
Slatus
DefininqTerms.Basedon our ex:Jerience
with manythesiscommittees,W'<'can
diet thai if a distinctionis madesuchas "tyoica,~ ratherthan ''exempli!ry."soi>ite0
',Nillasktor a definitionof what theauthorconsid
ersto be "typical." It mightbl' b ne
.
.
e1ter
to oealwith that here,ratherthan Ir, d'1 externpo
raneousexchange
with adviso,s.
Variability.Finally,we think that ~he authorintendsto assertthat one school.
1
enoughbecausethe substantial
varfetyof teachingpresentin feederschools will pr ~
ducea partkipantsample within whicha wide rangeof educational experiences
a~
represented
(that is an importantconditionfor the study). Although the argumen~
seemsplausible enough,wewouldpreferthat it bemadeevenmoreexplicit.It isimpor
tant to avoidany implicationthat diversityof studentssomehowmakesthis "typical"
schoolsitP11:orereµre:;2ri1a1ive
ul urb;;r.iliyh s(1100I~
(a cautionthe authorcloe~
rnake
in ,1 sub~equenr
sectionoi th!?p,-oposal).
Qualirat
ive ~tufliP\"f rhis typeco not huve
strongclaimsto externalvalidlty(transfcrabiity of I esul1~)
,is a conse~uence
of 5am
piing(althougi1
Tiley.TIayaddressthe matterof gc'lcralizabilityrhrougnothr:>r
n~eans).
Appendix.As illus:rationforour earliercriticismof placingthe explanationof sampleselectionin the appendife5.
yournuslnow t,Hr periodicallyto F aod G whileread(andrlo so if youare ro under'standOld comments
on tl1i1!section)
ing the follnwin<J
Sarnple
Thi.: primary sampk will he twdvi.: Mexican Amerkan
high school
intensivi.: focus of the study; a second sample will be six additional students who will participate in a focus group serving as a "member
check" (Cuha, 1n1) at the end of the study. I will purposefully select participants from a pool of volunteers in order to include a range of diverse
characteristics including: males and females, studenrs from various academic
srudents of varying grade point average, and first and second gen·
progr,1111s,
erntion Mexican Americans (see Appendix fo).The primary sample of twelve
students will each be interviewed three times. from these twelve, two wtll be
selected for obse:rvatron mid six for participation .in ongoing focus group
interviews (sec Appendix G).
I will interview twelve students in order r.o select a cross-section of
studCilts who have been exposed to diverse school and life experiences.
While this is not a study specifically about gender, tr.ick, grades, or generation, prior research indicates that students' experiences and pe1:ceptions vary
srniors-thc
Qu111itativ~ Study
269
dca lly a long chose dimensions (Gilligan, 1982; Oakes, l995; Suarez)'srcJl1a t9 95). Th us, chose factors will serve as "sensi tizing concepts"
. Theargumentthat qualitat'
d to the assumptionthat the1vestudies
AJ weakensthe positionof y arevaru
.
hat they flt-irrespective of QhlJalltam,e
t e status
1 manythesiscommitteesw
" rather than "exemplary
· • ,,e canpre.
.I
t b ,,
• sorneon
Jerso e typical.,.It mightbeb e
>ra
neousexchangewith adv'
etter
.
isors
Intends to assertthat one h.
SC ooj ·
ng presentIn feederschoolsw·u is.
0
rangeof ed.ucationalexperien:e/' •
the study). Although the argu are
? madeevenmoreexplicit It Is
.• ment
1mpor·
dentssomehowmakesthis
"ty· .
.
P1cal"
JOIs (a cautronrheauthordoesm ~
tfvestudiesof this typedo not h:v:
f results)as a consequence
of s •
neraliza~ilitythroughothermea:sT.
) of placingthe explanation of sam.
1
periodicallyto F and G while readnd our comments on that section).
1I
exican American high school
;econd sample will be six addi1sgroup serving as a "member
· I ~ill purposefully select parto include a range of diverse
tudcnts from various academic
rage, and first and second genThe primary sample of twelve
From these twelve, two will be
ltion in ongoing focus group
to select a cross-section of
school and life experiences.
td~r, tra(;k, grades, or genera·
enences and perceptions vary
o,ozcO,l99l: 391) in my dara ana lysis. By including a cross -sect ion of
(P:ittO~; a long these dimens ion s, I hop e to discover perceptio ns of effect ive
r11~e~n
g rhat are either wide ly shared or that differ categor ica lly (Patton,
,ea .;'. 172). By selecting six ?ch_er stud ents fo~ focu_s groups, using these
9
19rne
crireria, 1 can test my fmd,qgs from the interviews and seek further
stl
.
.111
forni::1oon.
.,
.
.
Since fine (1991) suggests that man y stud ents of co lor · make 1t" despite
herrhanbecause of teachers and th e system, this samp le of "ac hievers" w ill
,~caskedto locate the influence of thei r teachers in particular. Acknowl edging
l;catother people aod inscirurions impact th eir lives, I will be ca utious a nd
;
presuppose a causal relations hip between teache r influence and student
1
successiJ1 schoo .i.
:t
HowMany?Unlikethe situationin manyquantitativestudies,thereis no reliablemechanismfor definitivelyansweringthe question"How manyparticipantsare needed?"In
theend,establishingthat numbe1will be a matterof judgment.Thereare,nevertheless.
somesimplerulesof thumb(discussed
below)that applyto conditionslike thosein the
proposed
study.
ComplicatingConditions,As the autho1skillfullyargues,one of the desiredconditionsis representation
of particularcharacteristics
in the samplegroup.Hercitation of
priorresearch
to sustainthat point isexactlycorrect.Froma readingof AppendixF,however.we alsoknowthat maintaininga degreeof balanceamongthosecharacteristics
is
an additionalrequisite(for example,a finalsamplewith onlyoneout of twelvestudent
participants
to representthe "secondgeneration"conditiondearlywould be unacceptable).Thisis furthercomplicatedby the fact that the authorobviouslywants to maintainan evengenderbalance(or closeto it) as a priorityconsideration.
A Ruleof Thumb.
Theapplicablegeneralization
hereis that as the numberof desired
characteristics
increases,
and as you add provisionsabout maintaininga minimum
balanceof characteristics
in the total sample,the numberof participantsalso mustbe
increased.
Findingparticipantswith exactlythe right characteristics
simplyis too difficult;with a small,all-volunteerinitial poolto drawfrom,it maybe impossible.
Giventhe
specifications
as proposed,we think it is highlyunlikelythat the desiredsamplecanbe
obtained.Eitherthe total numberof participantsmustbe increased(at leasttwofold
wouldbe our roughguess),or the numberof characteristics
decreased
(or,ideally,both
miyhtbe adjusted).
BwJdingthe Sample.Althoughit preemptsthe laterdiscussionof selectionprocedures,it may be helpful hereto insertour adviceabout how the desiredsample(size
and nature)might be obtained.We would urge the author to proceedwith a pilot
(Continued)
270
Specimen Proposals
(Continlied)
rern,itr.1c."lt
ai :·1e s,udy site. Usingonly the :!1reeoriginal criteria fo,· :Jrei· .
I I111
selection(seniorciasss«i;i:s,Mexica:,American,volunwcr),the initial qrou~cif _rr
a,y
.
.
·
ll!Cr ·
thus9er~rilted can thenbP rnspec~,;>j
for represer1tatIon
and balanceof otherch Uits
;ersti~s.Given a pooi of s11Hicicnt
size,it rnayhe possibleto makenece'.;sary
a tac.
11e,icsbv p11rposefL,
selectio11
of candirfotesfor lhe final (nowmuchic1nJer'J
riart·CJtist.
.
.
. ..
K~
s;irnnle
. II lira: does,101 :Hoducea s,implPwith rt1edes,recchar,1Ctenstic~
the . nt
•
.
' ' 0111
r-'.IOn
I
,i furrt,cr "1c1
rgeted" rec;ui,menti?ffortmJy be avaiabJe.BP.yond
those~trateqi
e,s Ii
lessdesirable,al:.ernative~
ui si1'l~lyvvorki'lgwith whatcwr the call for volunteer:
prov:deoor a; ·110vIng
on '.O d d1;fe1c11t
s,1efor 3nothe1try.
J:'
ha:
Access and San1plc Selection
] have negotiated access to the district and schoo l where 1 plan to con.
ducr the study between ,lanuary and Nfarch, 1996 . In January, I will solicir
panicipants by posring announcemen ts and by staffing a sign-up table
whei:e !-itudents can meet me, ask questions, and fill out questionnaires.
l will perform the contact work myself because, "b ujldin g the interviewing
rclationshi p begins the mom~nt the potential pa.rticipanr hears of the study•;
(Seidman, 1991: 37). I will select the participants from the pool of studenrs
who complcre the demographic
information questionnaire
(copy in
Appl~ndix I-!) and whose parents co-sig11the ''informe d consent" agreement
(Kimmel, 1988: 67-76).
Kecausc Latino sr.udents are nor dispersed equally among rhe differenr
academic tracks_, and because the sample size is so small, the sample will
not be "repr.csc11t,1tive" of the school population, nor is it intcndt:d to
be (Patron, 1.990: 185). !Jsing the logic of qualitative inquiry, I am more
interested in understanding, in deprh, the experiences and perspective.~
of diverse individuals than being able to generalia those experiences to a
hrger population. To be considered for participation, students must be
sf.niors of Mexic,1n origin who have ;Htended schools in the U.S. for ar.
bi.st fivr- years (so that they have l'xperience with teachers .in at least two
U.S. schools).
G1m11Hiwo1k
. It is cleilrthat i!le at:thorhasdoneher homework.conipletingpreliminary
negotiationslor acce>ss
10 the si1eancldr.JftinginformedconsentforIlls and oche
r study
ciocu1ne'1ts
•.A!Iof tha~is powerfullyreassuring
to advi~ors.
Again,readersarc cautioned
Qualitative Study
ie threeoriginalcrite.
. rra for .
111
, volunteer).
theinitial r Pre1,llli
9
·esentation
and balance
of rec~'~
f be possibleto maken Other
chil:t
r thefinal(nowmuchlaecessary
a~Ju.
.
rger)p . st.
Ihe deslfedcharacteristi . art,cipalJI
· theOptio~
3vailable.
Beyondthosescsr
rateg· • 01
Ithwhateverthe<:allfor vol
ies1,e1/Je
anothertry.
unteers
ha.s
t'P
°
271
assumption
that thesampleis intendedto represent
thepopulationat the
st~nyorthestudentpopulations
again
in urbanschoolsmorebroadly.Planning to han51
sd1o01
e.ruitrnent
tasksherse
lf (onsite)is particularlyimportant
, for reasons
that she
dieihe rec
111ake
~~~x /.It isessentialthatAppendixI, "InterviewGuide,"be readin conjunction
•
following.
Again,thedecisionto relegatean itemto theappendixhasworked
,w1rh disadvantageof the reader
.
10 t e
5
:f:e
--------------------
JritervieWS
and school wh ere
I I
P an to c ·
.
·h, .1 996. In January r . .
011.
.
, w,11s 1·
anJ by staffin g a s·
o icit
tgn-up tabl
)ns, and fill ou t question . e
.·au..~e, ''b u1·1d·
rng rhe inte na1res
. . .
.
. .
rv1ew111
1
·1 p~rt1c1pant hears of th
g
.
•
e stud ,,
1pants from the pool of
d Y
·
.
Stu enrs
iauon questionnaire
(
.
"' f,
copy in
~ In ormed consent" . .
agi.ccmcnc
~d ~qual/y among the differem
ize i.~ so sma!J, the sample will
,ulat1011,nor is it intended to
qualitative inquiry•
·
, I am more
exp~ricnccs and perspectives
lera hze those experiences to a
irticipac·
1011, stu Jems muse be
·
led schools in rhe U.S. for at
~ with teachers in at least two
omework,
completing
preliminary
?drnnse~tformsandotherstudy
ors.Again,readers
arecautioned
f will use in-depth, phenomenological interviewing with cwelve Mexican
erican seniors as my primary method of data collection. Consistent with
~~~man's (1~91) mode l, each participan~ wi ll be intei:vie.wed three tirn.es.
rerviewing 1s important because, as Seidman (1991: 4) asserts, "If the
1
s:archer's goal ... is to understand the meaning people involved in educarionmake of their experience, then interviewing people provides a necessa.i:y,
if not always completely sufficient, avenue of inquiry."
In or<ler to provide context for understanding the participant's perspective, the first interview in Seidman's model (1991: 11) focuses on the person's life history. Using an "interview guide approach" (Patton, 1991: 288)
consisting of open-ended questions (sec Appendix 1), l will focus on the participant's school life history, particularly on experiences with and perceptions of past teachers. The second inrcrview focuses on the "concrete details
of the participant's present experience in the topic area of the study ... upon
which their opinions may be built" (Seidman, 1991: 12). In this interview,
I will ask students to describe their lives and the impact of current teachers.
"ln the third interview, participants a.re asked to reflect on the meaning of
their experience" (SeiJman, 1991: 12). In this interview, I will ask students
to make connections between pivotal K-12 teachers and the students' motivation, achievement, persistence in school, and aspirations. I will also ask
students to reflect on how teachers can support students in ways that will
help them achieve at high levels in school an<l beyond.
Whenever possible, I will conduct interviews at least one day, but not
more than one week, apart with each participant, in order to give them a
chance to reflect on their experience and to build rapport over time. I will
tape record each interview. A bilingual transcriber will assist me in transcribing half of the interviews. l will trnnscribe the other half and will listen to all
the tapes as I proofread and annotate the transcripts. Immediately following
each interview, I will record field notes (Patton, 1991: 239).
2 72
Sp~cirncn Propos,~ls
Seidm_an's
P:otocol.Althoughphenomenol~gical
in~erviewin_g
In the format . .
by Se1~man
1sa complexprocess
that requires~ons1derable
interpersonal
SkitOli1t1eo
of purpose(aswell asa · ~ reflJI
attentionto detail.anda thoroughunderstanding
guidedpractice),
thebriefdescription
givenhereisentirelyadequate.
TheSeidPeriodot
provides
a definitivesource,
andmostadvisors
arelikelyto befamiliarwith th;an_t~~
ologyinvolved.
Wherethatis not thecase,priorconversations
andprovision
njel,IQ(I.
0 refer.
encematerials
canserveaspreparation
.
Transcribing.
Eventhoughtheauthorisbilingual
{apointmadedearina laterseai
rheproposal,
alongwithherIntention
rogiveeachparticipant
thechoice
of language
fo0110f
interviews),
theu~eof a bilingual
transcriber
fora portionof thatdifficulttaskIsa sign•~,lheif
addition.
Providing
foranexternal
checkonherowntechnka
l skillreflects
theauthor~
ICclht
cemformaintaining
a highstandard
ofaccuracy
in thedata-againa signa
l heradviso,;'O~.
interpret
asindicative
of soundpreparation
andreadiness
forthelongtaskahead. WIii
CuingParricipants.
As someof youmayalreadyhaveconcluded
froma reactingf
AppendixI. however.
therernaybesomereasonfor concernaboutthe author'spla;
forcolledingdatafrominterviews.
An interviewer's
choiceof wordscansignaltheinter~
vlew.ee
as ro whatsortof responses
areanticipated
. Giventheprobabilitythatln the
contextof a friendly,nonthreatening
setting,mostparticipants
alreadywill h_
avesome
inclinationcosaythingsthatpleasetheinvestigator,
suchsignalscanserveasa pow.
Thatbeingthe<rase,
advisors
will belookingforcJ~ar
erful sourceof datacontamination.
assurance
that theauthorintends(initially,at least)to avoidgivingherparticipants
any
indicationthatteachers
shouldbeidentifiedasinfluentialfiguresin theirlives.
Whetherteachers
are called"memorable,
" "influen
tial," "pivotal.""effective,"
or
simply "good,"thosedescriptors
constitute
a dearsigna
l aboutwhatis expected
when
students
talkaboutteachers
. Evena seemingly
neutralword like"memorable"
cangive
moredirectionthanintended.
Stimulat
ingparticipants
to reflectuponandtalkabourthe
cat,
behaviors
of teachers
whomtheyperceive
ashavingfallenintoanyof thosepositive
egories
willbeanappropriate
(evennecessary)
partof theproposed
srudy.It wH/beso,
however, only after studentshave,of their own unpromptedvolition,indicatedthat
teachers
weieamongthepeople"whomadea difference"for chem.
Jnotherwords,It~
essentiaf
to firstestabli5h
the "whether"of things,beforemovingonto the"how"I
It is easyto understand
how the author.committedto the vitalsignificance
of the
resear
.ch questions,
steepedin the supportive
literature,
andeagerto moveaheadto
conductingthe study,mighthaveoverlooked
the possibility
of <:uing
her participanis
.
Happily,thecorrecrionIs.easilymade-and thereis clearevidence
in the proposal
that
pointsalreadymadein herpresenta·
she knowsprecisely
howto do it. Toparaphrase
tion: Don't rush,leadwith comfortable,
nonpremlptive
questions
that invitestudents
to share theirstories,listen intentlyandshowinterestin theparticipants'
perceptions,
be preparedto usefollow-up probequestionswhenneededto clarifyer extenda
response.
and movethe conversation
onlygradually(perhaps
in the laterinrervlews)
The
towardthe explkrt questionsthat addresssuchmattersas teachereffectivenes's.
readershouldnote.thata modestpilot trialof theinterviewguideml9hthave sensitized
theauthorto thismatterbeforepresentation
of thispreliminary
draftfor review
.
Qualitative Study
gical inteNiewingin th ,
"-S cons,
'derablelnterpee ,orrnat p, ,.
f
rsona/sk· 0 •1cfe,,
.
ndrngo. purpose(as ... II
"
. .
vve as 111,C<1
teru,
•re,sentirelyadequate
. TheS _a P'eriou
·arelikelyto be familiarw· herdrna,i
k ..of
·
It th
"UOk
·1orconversations
and pr I . e rne1nO'r'
ovsron r ...
o,,ref~(
.
ual(apoint madedearina
1aters"""'
..
thechoiceofIan .""onof
' parncrpant
for1he;
portionof thatdifficulttask
1
;Jlintechnical
ski/Ireflects
th
srgnffica
111
e author·
t1ie data-againa signals..e
d . scon.
.
,
11rav,so .
d
:1 111ess
,orthelongtaskahead '5 w,11
?adyhaveconcluded
froma ·
r
readin
n ,or concernabourtheaurs..• 9 of
'
h
.
11or
s Pia
•rsc o,ceof wordsc;ansign 1 . 115
.
·
a
the
rm
ated• Giventhe probabilityth . er.
· at rnth
>stpartrdpantsalreadyw,11
haveso. e
ator,suchsignalscanserveas me
I ca,ced .
.
a POW., • a visorsw,11
belookingfor I
·ct
·
·
cear
·t) r ·
a
' o ~vo1 g1vrng
herparticipants
1_fluent1al
figuresin theirlives. ny
1nlluentlal."
"pivotal" .,effecti •
. .
·
ve, or
signalaboutwhatis expected
wh
J_tralwordlike "memorable"
·can
ntsto reflectuponandtalkaboutthe
1fallenintoanyof thosepositivecatt of theproposed
srudy
. It willbe
t:g~
g;;
so
unprompted
volition,
indicated
that
'rence,,for them.In otherwords,it is
eforemovingon to the "how"1
medto thevital significance
of the
rture,and eagerto moveaheadto
•ossibili~of cuingherparticipants.
evrdence
in the proposalrJ1at
_clear
,~tsalreadymadein herpresentative ques_tfons
that invitestudents
st in the participants'
perceptions,
=nneededto clarifyor extenda
· (perhapsin the later inteNiews)
te_
rs asteachereffee:tiveness.
The
rviewguidemighthavesensitized
ireliminary
draftfor review.
273
,..vations
obse to observe ar least rwo of the participants from when they rise .in the
I P_i
an tJntiJ they prepare to retire in the evening . By spending a day wi th the
,,,vrr1tng expect cogee a sense of how the st uden ts negotiate their home~schoo l,
0 1
5111de
'\ cultures, which may prove helpful in comexmalizing the .information
.,od~:i~~s
chat they shat·~about r_he relati:e influence of teachers_in their Jives. 11
1d I ing che observaaons, I will take field notes (see Appen dix J) 12 and w ill
111
J)Ll
l:em soon after, adding reflections. I will use the data from the o bse r1YP~
t to infor m my follow-up questions in the interviews an d to generate
vsiuons .
u
jve hypotheses .
cei1rar
Rationale.
Wearenot convinced
that the proposed
observations
will contributesignificantlyto the data collectedfromothersources.
Furthermore,
givenwhat is provided
hete(<1nd
elsewherein the proposal)concerning
observations.
we are not sure the
authorhe,selfhasfullycometo gripswith the howandwhyof thispartof theresearch
plan.Withan alreadydemanding
schedule
of datacollection,
andthe arduoustasksof
transcription
and analysisbeyondthat, unlessmorepersuasiveargumentscan be
advanced,
we are inclinedto advisedroppingthiscomponent
of the proposedstudy.
Focus Groups
"(;roups are not just a convenient way to accumulate the individual knowledge of their members. They give rise synergiscically to insights and solutions
that would not come about without them" (Brown et al., 1989:40). Both
Hidalgo (1.991) and Noya (1995) conducted focus groups with Latino adolescents and found the sessions fruitful. In this study, I will facilitate two groups.
One will consist of six of the twelve primary interviewees. We will meet as
a group three times for fifty minutes to discuss the attributes of influential
teachers. The six participants in the member check group will meet once for
ninety minutes, near the end of the study, to critique the developing analysis.
CitingSources.
Therationalefor obtainingfocusgroupdata makesgoodsensein the
contextof thisstudy(althoughfor somereadersa muchmorethoroughdefinitionof
thetermmighthavebeenrequiredto makethilt apparent).
Thatconclusion
is furthered
bydocumentation
of the factthat the technique
haspreviously
provedto be a helpful
supplementto interviews.Becausethe formationand facilitationof focusgroups
requiresconsiderable
knowledge,
skill,andplanning,we wouldnormallyexpectto see
bothcitationsto standardworkson the methodology
(seeChapter5 for our suggestionsin that regard)andindicationof successful
pilot work by the author.
274
Specimen Proposals
Trustworthiness
I have made several choices to increase " trustworthiness" (L·
(;uba, 1985) and to minimize comm on threats to validity. Firs:ctn 8<_
sel.ectcda school where neither srudenrs nor teachers know me ro ~- _hav
e
·
1111
. . f rorn persona I b.1as towar ds participants
t hreats to va l.d.
I ay ansmg
.nf1~1:
viduals named by rhcm. Second, because I am bilingual, I wiJJ inor 111di.
studcnrs in the language they prefer. The bili11gualtranscriber will a:s~tvie\v
in checking the accuracy of both direct quotes and trans.larions. Thir;t Ille
final focus group will provide a member check on the analysis . Fottnh t?e
discuss the possible influence that my own life experiences-for exam~( Wilt
a former teacher and administrator who worked for ten years in sour~' as
( ..a 1I·forn1a
. w1t1
. j many ,v'·{ ex1can
.
A· men.can stu denrs-may have on ll'ly 1ter11
.
ccprions and thinking. Finally, by conducting mult ip le interviews over ti~er.
as well as by using multiple data collection strategies, I will miniltlize :·
chances that the findings will be based on idiosyncratic data .
e
1
Getting11Right.It is impossible
to defineherethe complexof criteriaandproced
uresthat
qualitativeresearchers
assignto the word "trustworthiness."Advisors will knowwhat'is
inter;ded,
however.
andil full discussion
isnot requiredin theproposal.Atbottom,trustwor
thinesshasto dowith thequalfty(goodness)
of qualltatlyeresearth.
Thus,in practicalterms,
it hasto do with QSlablishing
andmeetingcriteria thatwill leadreade
rs of the studyreport
to concludethatthe investigator
"got it righi"1 Put anotherway,the termdenotesa revi~
inrernal,mdexterrnlvalidity,andobjectivit'j,
to
sior of traditionalcriteriasuchasreliability,
maketllem moreappropriate
to the ass1J1nptions
mildein thequalitativeparadigm.
Whattheauthorpresents
hereareprocedures
that will servethe endsof trustworthiof that constructbe
ness,thoughwe vvouldhavepreferredthat the severalcomporients
madeexplicit(theusualfist of criteriaincludescredibility,transferabilily,
dependability,
and
wnflrmability),andthat pmposedstrategies
for meetingeath thenbe.discussed
in turn.
ComingClean.Wewereparticularlypleasedto note that the authorplansto write
a thorough·explicationof the natureandpotentialinfluenceof herown life experiente
s
as theymightbe expeqedto shapethe researthprocess
. (:)fcourse,that discussio11
will
be necessary
for an informedappreciationof the study'sconclusions.
ConfirmingData.We continueto be concernedabout the proposeduseof a secondarysampleof students,who werenor interviewed,for the purposeof conductinga
"mernbercheck."Theword "member"hereordinarilydesignates
a lull participantwho
has beenselectedto read throughtranscr
ipts to ensureaccuracyof what has been
recorded,o, to reviewproductsof dataanalysissuchas thematicstatementsor individ1Jalprofiies.Thepurposeis to detectanysubstantialdiscrepancies
betweenthe perceptions of the participants(whetheras individualsor as membersof a group)and the
investigator's
reconstructions
and representations
. If !hilt is the intention,then it is the
participantsin the primarysamplewho will haveto do the checking.
Qualitative Studv
275
C oncerns
as_e "trustworth iness"
.
~ltnc
th rea ts to vaJidit
01
y. F1
ll R~
1or teac hers know rn r sr, I h ..._.
e to .. ~vc
. .
bJas
towards parci· ·
rn1n11
1, .
..
c1pan.ts . , 1~
e TI am bil11
1gual, l W1·u rn
, Or 'tid·,_
b, mgua l transcribe•· ·11 tervj<.'\
, WJ a .
V
uotes a nd tra nslations ,- h~Slstn,c
. .l lt d
iec I< on t he ana lysis F
., th,,
.
. OUrtl1
'
life experiences- fa c. exa , 1Will
1111
Pe, as
worked for ten years ·
10 SOut j1
stu dents- may h ave
ern
.
. .
on my
ing mul trple interview
Per.
s over ·
>n strategies I wi ll rn · . . tirne,
. .
,
in1m,2e ·1
td.rosyn.cratic data .
tie
omplexof-criteria
andprocedures
·h
' at
• " A.d •
'rth
. iness. visorswill knowwhatis
d ,_
n the proposal.At
bottom,trustwo,
.
:atN~research.
Thus
, in practical
terms
3t will leadreaders
of thestudyreport
3nother
way,thetermdenotes
a revind.externalvalidity,
andobjeaivity,
to
1e thequalitative
parad
igm.
at wrll seNethe endsof trustworthi·~l components
of thatconstruct
be
'.'.ty•
transferability,
dependability,
and
,,ngeachthenbediscussed
Inturn.
notethat theauthorplansto write
ifluenceof herown lifeexperiences
thatdiscussion
will
cess.Ofcourse,
rdy'sconclusions.
aboutthe proposeduseof a seea
d, fo~the purposeof conducting
Ydesignates
a full participantwho
1sureaccuracy
of what hasbeen
asthematicstatements
or individ·
liscrepancies
betweenthepercep3s members
of a group)and the
:hatis the intention,thenit is the
o thechecking.
m
~
1
tPiCl¼
che nature o.f the research questions, the confidentiality
provided
0
. . d
..
,
d
DLleht carefu lly marnrame
part1c1pant anonylllity, an . my status as a
rlJfoLIS stu den t with no connections to the teac hers or sc hoo l admjn is tra~tndL~~::re
is minim~! ~hre_at to the weU-being of the smdents in the stu dy as
11
r1°' _leof rbeir partrc1pat1on .
~ r:;i seance of "empathetic neutrality "'" should minimize the threat of
1
•
\ewing as exploitat ion"-a process that turns othe rs into subjects so
.,111
ref'
dieir words can be appropriated for the benefit of the researcher"
rhac
· l pro 'blems in
· some stu dies.
..,~Jan 1991:7) th ac presents eoI ,ca
(Se1lu•
protecting
Participants.
Theauthor'scarefulreassurances
notwithstanding,
hard.experi
ences
havetaughtus thattherealwaysis the lurkingpossibility
of harmto participants
in a naturalistic
fieldstudy.Anonymity
nevercanbeguaranteed,
if onlybecause
it is not
entirely
underthe researcher's
control.In a typicalschool,everyone
will knowwho the
participants
wereandexactlywhatwasasked1ntheinterviews
anddiscussed
in thefocus
groups.
Furthermore,
if memberchecking
is done,whateveris checked
hasthepotentia
l
tobecome
publicand,correctly
or incorrect
ly,will beassignedto everyparticipant.
Wethinktheauthor'sarguments
aboutmeetingherethicalresponsibilities
arereasonable.
Certainly,
shehasgivenusgoodreasonto believethat shehasgivenserious
consideration
to theproblemandwill protectherstudentcollaborators
in everypossibleway.Wewantto warnourreaders,
however,
thatethicalconcerns
areneversimple.
Carefulforethought
in planningandwatchfuldiligencein execution
areresponsibilities
thatcomewith thedecisionto do qualitativeresearch.
Limitations
As with all exploratory research, the findings of this study will be tentative. It is important to note: 1) the sample size and procedures for participant
selection, while appropriate for a qualitative study, will not support generalization to a larger population of Mexican Americans, and 2) the relationship between students' perceptions of dfective teachers and empirical
measures of "teacher effectiveness" has yet to be defined. The exploration of
that relationship is beyond the scope of this study.
Data Ana]ysis
Because this is an exploratory study, I will begin coding data from interview transcripts, field notes, and focus group transcripts using coding
2 7o
Sp~.::imcn Proposal;;
cat~gorics (Patton, 1991:402). s11ggcsted by_the literature (~ppendix {(
l will acr.ivelyseek p;mcrns ot darn suggest ing new categories as Welt. ~lit
<lata organized and displayed in i-hat fashion, l wilt return f irst to an s111g
sis of interviews from individual participants, as separnte cases, ant;:aly.
then consider the wider maner of cross-case anaJysis . Such a seq nl},
of analytic seeps would conforrn broadly to the suggestions of G!ase~encc
Strauss (1967), who have argue<l !'hat an understanding of individual 3tlcl
1 in
. any way) 1s
• t he best guarantor .for theoreti
cases
(he fore r.hey are aggrcgar.ec.
assertions that arc grounded in specific contexts and real-world patter ns caJ
i\'ext, bl'causl' '\\ phenomenoJ.ogist assumes a commonaJity in th.
human experiences and tnLlst· use rigorously the mecl1od of bracketingOse
search for those cornrnonalir;es" (Eichelberger, 1989: 6), I wi.ll read acroto
interviews, noi:ing simibrit.ics and differences. I will then use pattern coclin
ss
(Miles & I lubel.'man, 1':>'.i4)ro identi fy common themes. While I will ex.alll~
ine the entire sequence cf responses a bout teachers by each participant, 1 Will
also disringuish between r.hosf. categ ories generated by the stud ents them.
selves and those discussed in response to probes . For examp le, if the students
mention nothing about the relevance of the instructional materials to thei.r
lives, 1 might inquire about r.hcconne crion, but will indicate that f needed to
probe ro elicit thar information.
l
Finally, I wi!I prepare the data analysis by briefly des1.:ribinge<1cbof the
twelve p,Hti-:ipants, and, by using quor.es from their interviews, will illlJstrate
comrnon themes as wci l as atypical responses. Transcripts from die focus
groups will be utilized primarily as a check for confirmation of the salience
of themes across the sample, although I wil.l indude in rny r.hcsis excerpts of
dia logue that illustrate important points oF consensu~ or disagn:·c'.lllcnt.The
observations wiU be used both ro describe two of the participanls more fully
<1ndto fill in descriptions of the school.
KeepingHone;t3ooks.Again,the rrom11Jl'es
ind;c,,leo'here(ccoinq,catego1y
developmentand sorli1gof d2:,1.id0rnificauo11
of paiterns.constructionof th~rnaticstatements,
appro,01
i,ite to theresearch
task.
anowriiing 9('oundl!dde~ciptionsof eart1ciuc1nts)
are ,111
WecanacidS<'verdl
suqqestiori\tha1rnign1wcn~then the nlar.for analysis.
First.;t would iJ0wise to consio'r.1
useol oneof the sofiwc1re
flrogrDmsfor comp111erDasedanc:1lysis.
Tl1is\(;Jdy will 9er.c1;itesuchan enorrnousvolumeof dat2 that hardcopyn1anipuLilio11
will bl' rliffiwJt. Sernnd.it will Lie important to maintaina recordof
lhe c!xau~ourcefor ec1ch
category. Nol only .;ire,1flvisnrsoften curiousabout suchori9ins. but the fin,1! :rpm( wiJJr::-quircwritin1;about calegorydevt:1lop111en{
long after
irnmcdiat::n1emoryha~dir·1med.fhird. 1fat ail possible,do sornepilot trialsof the proposed<Jnalysis
usir;gin1erviewd2t;1obtrined ;rom .1sourceOll{sidethe studysi{e (an
Qualitative Study
by the literature (A
!sting new categ . Pp.endi~ v
ones as
,,) b
.
I11011, I wiUreturn f. ' Welt·u' ."t
.
itst to
s,n
pants, as separate
ana It
case
• n"t
ss-case analysis S
s, and Y·
· uch a
Q111
Y to the suggestions of sequel\Y
understanding of i d' ? laser,..11-t
O lVtd
"IJ(l
s t be best guarant o.1 f
Ualca
or ti
Ses
>ntexts and real-world
leorerfca1
~ssumes a common 1· Patterns
a tty i
·
usJy the method of b . n thosh
.
racke ·
~
erger, 1989: 6}, r will . ting to
tead ac
1Ces. 1 WJ·11
then use p attei:
toss
11
mmon them.es. While 1 will Codin&
teach ers byeach pare· .
e)(a,11
.
lClparu· I .
generated by the stud
' w,11
. b
en.ts th
o es. For example if b
en,.
· .
' t e stude
le rnsrrnctiona! material
nts
. . .
• s to the·
, but wdl mdtcate chat [' d tr
1ec cd to
by briefly
each of the
. . describi.noo
:>mcl1e1rinterviews ' will t·1·1usrrare
nses. Transcripts from -1, f
f .
.
.
· u e ocus
. oi: confo-mat 1on of the saJ·
d .
1ence
.
1}oc 1u e m my thes is excei;pts of
consensus or disao-reeme t 1·1
f
"'
n.
le
,vo o the participants more fully
~tedhere.(coding,categorydevelopconstruction
of themaHc
statements
~ allappropriate
to theresearch
task'.
entheplanforanalysis.
hesoftwareprograms
forcomputer10~mous
volumeof data that hard~.rmportantto maintaina recordof
11sors
oftencuriousaboutsuchorit categorydevelopment
long after
ble,do somepilot trialsof thepro:1 sourceoutsidethe studysite(an
277
dixshowingthat ~hishasbeenaccompli:hed
successfully
~lwaysis reassuring
to
appen) finally,make11an absoluteand 1nv1olate
rulenot to ignoreor discarddata
.s'dvl50~ · contraryto initial expectations.
or that appearnot to fit comfortablyinto
t~at
'~ng
categories
for
analysis.
enier
__
,________________________
..,
9
co:ndusion
. country cannot afford, ethica lly or economically, to fail to eJucate
'fh•:her generation of Mexican Americ an students. While many factors out·
0 .~: rhe contro l of teachers may contri bute co underachievement
of Latinos
~, rhe [Joited States, evidence abom1ds to confirm that teachers can make a
~fference, although many teachers que stion their ability to do so. ln the pro·
osed scudy I will explore Mexican American high school seniors' perspec~ves on what constitutes effective teaching and their perspectives on the
influence that teachers have on their learning, school careers, and lives. lt is
rn)' hope that teachers who are act ively seeking ways to support the achievement of their Mexican American students will find, in reading this thesis,
useful ideas that will support them in their important work. Jc is equally my
hope that teachers who hav e doubted their ability to make a difference in
their Mexican America.n students' ach ievement may reconsider their thinking and pedagogy upon contemplating this research .
TheBadNewsandtheGoodNews.Weweredelightedto findtheauthorstillenthusiastic
andconvinced
of the potentialutilityin herproposed
study.Sadto report,hardlyanyone
readsthesesanddissertations-except
othergraduate
students!
Theencouraging
flip side
of thatrule,however,
is thatpublications
basedon thesesanddissertations
canbeaimed
at anyaudience
theresearcher
desires
to reach---0ther
scholars,
professors.
policymakers,
andpractitioners
at anylevelof education-including,
of course,the teachers
specified
above.
All thatis requiredis preparation
of a soundmanuscript
basedon the dissertation
(orsomepartthereof)andselection
of anoutletthatreaches
theintendedreaders.
Good
proposals
leadto goodstudies,
andwhenappropriately
translated
anddisseminated,
the
findingsfromgoodstudiescanmakea positivedifference
in anyprofessional
field.
A Postscript to this Proposa l
The Restof the Story.This proposal, already well along in its development, was reviewed
by a committee of thesis advisors, the author was given feedback and suggestions for
278
Specimen Propos~1ls
improveme nts, the revisions were executed. and with the full suppor-i: of th
co mmittee the st.udy was approved by the Graduate School. In that proc e advis0 1)
tho ugh not all, of the points in our critique above found their way into th ess, so!\'11!
.
. the first draft Was
e final Pro.'
po sal. Nota bly, tt,e observation
component suggested 1n
(as was the "member check" procedure), the samp le size was expanded (to : roPPed
3
ipants), the specifications for representation and balance of characteristics Wer Pat·t1c.
tained in somewhat less demanding form, and discussion of teacher lnflu e ll'laln.
ence was
placed at the end of the interview sequence.
As for our earlier note about the author's assumptions concern1ng the role
by teachers in the cosmo logy of influences that shape student school exper· Played
this po int most readers will have detected that her unspoken but pervasive ience
ex ' bY
t io n was that some (ff not most} of her participants would nominate (and d~~~taparcicular teachers as pivotal individuals In their school careers. From the ou ibe)
we found ourselves asking, "What ff none (or very few) of her participants nonftser.
teachers as influential In making it through to graduacio.n?" At the least, that out~:ate
would limit a search for the attributes of effective teache rs-although it might w~~
open the door to other important insights into students' perceptions of what and
who were he lpful in their schoo l careers.
Given her own life experiences, the personal values that are evident in the preposa l, and the nature of her continu ing professional development, the expectation that
teachers might be sa lient was perfectly reasonab le. Indeed, In the case of several of her
pa r dcipants , the assumption was verified by direct and persuasive testimony. Virtually
every one of the study participants indicated that they had encountered one or more
teachers whom they considered to have been excellent. Furthermore, they unanimous ly affirmed the proposition that the quality of teaching makes an Important difference in school .
N eve rthe less, the main direc:Uon of llnding; ran -:ontr,iry co some of the assL1mpt lo ns apparent In the first.di-aft of the ;irJcho1·'sproposal. The mcyorityof swdents perceived their education and livc:sto /;ove been shapr:xlmor.~profi>und/y
by lheir inrcmetions
with parents, fricr,ds,and peers than witil te()che;-s.That her study was sens it ive w just
such a pOS$ibility,and that she was n=:adytu make that unexpected and (the reby) particularfy valuable finding \"he cr~nter of .m c11·ticubteand powerful report, were the consequence of thoroughprepart1tion,
;irongadviseme:ni;
and developmentof a sound research
plan throughsuccessiverevisio11s
o( her proposal.fhe lesson ;:c be taken here is that effort
expanded on construction of ;i ,:ar~fully r;onsidered propo~«I for qualitative research
may well be c1simportant as Ccildoct of the study itself.
Reforenccs
Qualitative Study
with the full support of th
e <tdv;
duace School. In that
Proc:
sol',.
·e found their way int
ess, sa ·'
.
o the fl11 Ille
ested 111 the first draft
1a1Pr '
.
Wasd1
a•
.pIe size was expanded (
·app
to 34
ed
,a rance of characterist ·
Part; .
.
.
ICS Wer .
C:,
facuss1on of teach e r 1•n ftUene t11a1i
i.
Ce \\-as
rmpcions concerning th
e role I
hape student school ex .. Payed
Pei ienc
e, Ii)>
=r unspoken but pervas·
1ve e)(p
nts would nominate (and d ect,i.
· school careers. Fro m h esc;·ibe)
f
t e outs
1 ew) of her partic ipants no , . et,
'uacion?" At the lease t h
'l'linate
' at Outc0
e teachers---although it m· h fl'le
students' percepti o ns of ight Well
w at and
ialues chat are evidenr. in th
d
.
epro
. evelopment, the expectation tha~
ndeed, in the case of several of h
and persuasive testimony v· er
· rrtuaUy
1ey had encountered one or rn
ore
.
..cellent.. Furthermore, the Y unani.
teaching makes an important difconcrary to some of the assump.
posaf. The majorityof studentsper.
wre profoundlyby their interactions
it her study was sensitive to just
at unexpected and (thereby) parnd powerful report, were the con-
endi ces
M'P
j\pPendix F: Sample Selection
. rend to select a cross-section of students for participation in both samsample ot 12.an~ the "m~mber_check" group. Ideally, the
~amplewill assume a form approximately hke rlHS:
(:;s,the primary
composition of Primary Sample
J\ll mLtSt be seniors of Mexican decent who have attended US schools for
at ]east five years.
1 male in honors classes
·1female in honors classes
1 male in "regular" classes who is getting good grades (i\'s & B's)
'.I
inalc in "regular" das.~es who is getr.ing average grades (C's & F's)
1 female in "regular" dasses who is getting good grades (A's & H's)
1 female in "regular" classes who is getting average grades (C's & F's)
1 male in "hasic" or special c<lucation classes
J female in "basic" or special education classes
l male first-generation Mexican American
d developmentof a sound research
1 female firsr-generar.ion Mexican American
;on to be taken here is chat effort
proposal for qualitative research
elf.
I male second-generation
__
)
Mexican American
l female second-gencrnr.icm Mexican American
Tora]=
-~
279
·12.
Composition of Secondary Sample
All must he seniors of M.exican decent who have attended US schools for
at least five years.
280
Spe.cimen Proposals
l male firsr-gc.:neral'ion,'vlexican American
J fcm:~lefirst-generation Mexican i\mcric:rn
1 male second-generation
1\!kxican American
l female sei::ond-generntion 1\1cxica11American
l male or female honors studenc
1 male or female student in "basic" or -~pecial education dasses
Total=
6
Appendix G: Sample and Methodology
Table 1
---------------------------------
l?.irtid/)mit
Number
Acti11ity
-------------------------------l
2
3
4
5
6
7
8
9
lO
lI
12
13
14
15
]6
17
lS
3 in-depth, individu,1I interviews + one day ohservation.
3 in-depth, individual interviews + one day ohservatiou
3 in-depth, individu:;il interviews + 3 focus group interviews
Jin-depth, inclividual interview&+ 3 focus gnmp ime1views
3 in-depth, individual interviews + J focus gmllp interviews
3 in-depth, individual intcrview.s + 3 focus group interviews
3 in-depth, individual ir1tcrvicws + 3 focus group interviews
3 in-depth, imlividual interviews + 3 focus group ir1terviews
Note: participants 3-8 will all he in the same focus group
r.hat will meet r.hrcc times.
3 in-depth, individual interviews
3 in-deprh, individual inrervic\.vs
3 in-deprh, individual interviews
•~in-deprh, i11dividual interviews
"member check" focus group interview
"member check" focus group interview
"member check" focus group interview
"member check" focus group interview
"member check" focus group interview
"member check'' focus group interview
Note: parricipanrs 13-J S will all be in the same focus group
rhat will meet once for ninery minutes.
282
Specimen Proposals
school, and a.,pirations. lf rhcy have not already done so, I will ask,th
dcs.::1·ibeto me the 4ualities and at tributes of an effecti ve teac:het e~l\t()
based on son1cone they know, or based on how they imagine one c:o: ,: 1tbe-r
1
bl:,)
Notes
1. fn rhis propos;1l, the terms "Latino" and "Hispanic" refer interchan
to people, or their-descendants, who hail from diverse Spanish-speaking r!e~bty
such as Mexico, Cuba, PL1ertoRico; rhe Dominican Republic, and parts of Cc~; 11s
and South America (Suarez-Orozco & Suarez-Orozco, 1995).
taJ
2. Teacher efficacy measures rhe extent to which reache1:sbelieve their
. . efCrccc on sru denc acrnevement.
L'
0 rt.s
w1·11I1ave a posmve
3. I will focus rhe study on Mexican Americans, rather. than on all Latinos
Mexican Americans constitute more than half of r.he Latino population. I have
sen to use the term Mexican American, rather tl1an Chicano, Mexicano • a
' 01 fly
other terms because many people prefer it (Nieto, 1992).
4 . These terms ("motivation," etc.) will be defined by che students .
5. The validity of these tests for Lati110students is qucstionab Je (Valencia
en
ch:~
1991).
'
6. U.S. Uep;1rtme"ni of Health, [du,:ation, and \V/r[forc ( I 966).
7. Th,: ,·dacionship between st udcnl~' perc::ption.s of effective te,11:hcrs and
empiric.ii rne:1sures of tc,1cher die,:riven,:.s.~ is 1,11kil.Ownam/ is b(:yond the sc:op~oi
this 5tudy.
8. l'ht'Se in~csri3,1tions focus primarily on "111.Jcro~yste111facror," (f..eCompre
& Dworkin, 199 i: Si) ,;nch as tamily var;:1bles and ~choo] .,rn1..:t1ire(see Appendix
Bi,, r,1ther than tea,:l1cr eff'c,:tiv'.'ness per se, l'irns they will nor be discussed further
here:.
9. Definitions of "culturally relevant pedagogy" differ, but often include
involving students' families and connecting subject matter wi.d1studen ts' Jivesand
experience in meaningful ways (see Appendix E- l ).
10. Phenomenological inquiry focuses on the question: "What is the structure
and essence of experience of this phenomenon for these people?" (Patton, 1990: 69).
11. My experience visiti11ghomes of Mexican Americans while working for
Migrauc Education leads me to believe I will not have difficult)' finding participants
who are willing to bring me inco their homes.
·12. I will ~ssign pse1.1cionyinsto tl1e sd1ool :rnd ro re:lchns rr11:111io11ed
by
participanr.s.
'13. The foc1.1sof this .;ri1dy is on ,it!.denr.,· per,:eptions: rherdo re,
ob.serve teachers.
1.
will nor
14. \...rnp,~theti.: 1ieutrality, "Fn1pa,hy .. , t3 a [caring I .sr;H1cc toward,; the people
one en..:.ouutrrs, whilt: ncuiraliry is :1 .stance toward d ie f.indi,13s"{Parr.on, ·1990: 58),