Instrumentation

advertisement
Educational Research
Chapter 5
Selecting Measuring Instruments
Gay and Airasian
Collecting Data


The collection of data is an extremely
important part of all research
endeavors, for the conclusions for a
study are based on what the data
reveal.
As a result, the kind (s) of data to be
collected, the method (s) of collection
to used, and the scoring of the data
need to be considered with care.
“Data”



In this chapter:
Define data
Present several
types of instruments
that can be used to
collect data in a
research study

Different properties
that scores are
assumed to possess
Objectives: By the end of this chapter
you should be able to:



1)
Explain what is meant
by the term “data”
2) Explain what is meant
by the term
‘instrumentation”
3( Name three ways in
which data can be
collected by researchers



Explain what is meant by
the term “data-collection
instrument”
Describe five types of
researcher-completed
instruments used in
educational research
Describe five types of
subject-completed
instruments used in
educational research
Objectives


Explain what is meant
by the term
‘unobtrusive
measures” and give
two examples of such
measures
Name four types of
measurement scales
and give an example
of each
Name three
different types of
scores used In
educational
research and give
an example of each

Objectives:


Describe briefly the difference between
norm-referenced and criterionreferenced instrument
Describe how to score, tabulate, and
code data for analysis
Flow of Activities in Collecting
Data
Flow of Activities
Identify the variable
Operationally define the variable
Locate data (measures, observations,
documents with questions and scales)
Collect data on instruments yielding
numeric scores
Example
Self-efficacy for learning from others
Level of confidence that an individual
can learn something by being taught
by others
13 items on a self-efficacy attitudinal
scale from Bergin (1989)
Scores of each item ranged from 010 with 10 being “completely
confident.”
Data Collection

Scientific and disciplined inquiry
requires the collection, analysis, and
interpretation of data


Data – the pieces of information that are
collected to examine the research topic
Issues related to the collection of this
information are the focus of this chapter
Data Collection

Terminology related to data

Constructs – abstractions that cannot be
observed directly but are helpful when
trying to explain behavior



Intelligence
Teacher effectiveness
Self esteem
Identify Data Options: Specify
Variables






Independent Variables
Dependent Variables
Intervening Variables
Control
Moderating
Confounding
Identify Data Options:
Operationalize Variables


Operational Definition: The specification of how the
variable will be defined and measured
 typically based on the literature
 often found in reports under “definition of terms”
Sometimes the researcher must construct it
Some Times When Operational Definitions
Would Be Helpful
Figure 2.2
8
Which of the Following Definitions Are Operational?
Page 34
1. As shown by enthusiasm in class
2. As judged by the student’s math teacher using a rating scale she
developed
3. As measured by the “Math Interest” questionnaire
4. As shown by attention to math tasks in class
5. As reflected by achievement in mathematics
6. As indicated by records showing enrollment in mathematics
electives
7. As shown by effort expended in class
8. As demonstrated by number of optional assignments completed
9. As demonstrated by reading math books outside class
10. As observed by teacher aides using the “Mathematics Interest”
observation record
7
Data Collection

Data terminology (continued)

Operational definition – the ways by which
constructs are observed and measured




Weschler IQ test
Virgilio Teacher Effectiveness Inventory
Tennessee Self-Concept Scale
Variable – a construct that has been
operationalized and has two or more
values
WHAT ARE DATA?


The term "data" refers to the kinds of
information researchers obtain on the
subjects of their research.
The term "instrumentation" refers to the
entire process of collecting data in a research
investigation.
KEY QUESTIONS


An important consideration in the
choice of an instrument to be used in a
research investigation is validity:
the extent to which results from it
permit researchers to draw warranted
conclusions about the characteristics of
the individuals studied.
CONDITIONS




It involves not only the
selection or design of the
instruments but also the
conditions under which the
instruments will be
administered.
1. Where? -- location
2. When? - - Time
3. How often?- - Frequency

4. Who? --
administration of the
instruments

How you answer
these questions may
affect the data
obtained!
Good Instruments?




The data provided by
any instrument may be
affected by nay or all of
the preceding
considerations
If administered
incorrectly, disliked
Noisy or inhospitable
conditions
Subjects are exhausted


Every instrument if
it is of any value
must allow
researchers to draw
accurate conclusions
about the
capabilities or other
characteristics of the
people being studied
VALIDITY, RELIABILITY, AND
OBJECTIVITY




1) Validity
An important consideration in the choice of
an instrument to be used in a research
investigation is validity:
the extent to which results from it permit
researchers to draw warranted conclusions
about the characteristics of the individuals
studied.
Reliability and Objectivity


2) A reliable instrument is one that gives consistent
results.
3) Whenever possible, researchers try to eliminate
subjectivity from the judgments they make about the
achievement, performance, or characteristics of
subjects. That is, the researchers try to be objective.
USABILITY






Is it easy to use?
How long will it take to administer?
Are directions clear?
Is it appropriate for the ethnic group or other groups
to whom it will be administered?
How easy is it to score?
To interpret the scores?
Practical Questions





How much does it cost?
Do equivalent forms exist?
Have any problems been reported?
Does Evidence of its reliability and validity exist?
Save time, energy and headaches!!!
Who Provides the Information



Research instruments can be classified in
many ways.
Some of the more common are in terms of
who provides the data, the method of data
collection, who collects the data, and what
kind of response they require from the
subjects.
Data Obtained



Research data are data obtained by directly
or indirectly assessing the subjects of the
study.
Self-report data are data provided by the
subjects of the study themselves.
Informant data are data provided by other
people about the subjects of a study.
Researcher Instruments



Many types of researcher-completed
instruments exist.
Some of the more commonly used are rating
scales, interview schedules, tally sheets,
flowcharts, performance checklists, anecdotal
records, and time-and-motion logs.
Subject Instruments



The types of items or questions used in
subject-completed instruments can take
many forms,
but they all can be classified as either
selection or supply items.
Subject Instruments

The types of items or questions used in
subject-completed instruments can take
many forms, but they all can be
classified as either selection or supply
items.
Subject Instruments


There are also many types of instruments
that are completed by the subjects of a study
rather than the researcher.
Some of the more commonly used of this
type are questionnaires; self-checklists;
attitude scales; personality inventories;
achievement, aptitude, and performance
tests; projective devices; and socimetric
devices.
Subject Instruments (con.t)


Examples of selection items include
true-false items, multiple-choice items,
matching items, and interpretive
exercises. Examples of supply items
include short-answer items and essay
questions.
Where Did the Instruments come
From?



1)
Find and administer a previously existing instrument of
some sort, or
2) administer an instrument the researcher personally
developed or had developed by someone else
An excellent source for locating already
available tests is the ERIC Clearinghouse on
Assessment and Evaluation.
Data Collection

Measurement scales

Nominal – categories


Ordinal – ordered categories


Rank in class, order of finish, etc.
Interval – equal intervals


Gender, ethnicity, etc.
Test scores, attitude scores, etc.
Ratio – absolute zero

Time, height, weight, etc.
Four Types of Measurement Scales
Figure 7.25
50
SCALE
EXAMPLE
Nominal
Gender
Ordinal
Position in race
Interval
Temperature
(in Fahrenheit)
Ratio
Money
An Ordinal Scale: The Winner of a Horse Race
Figure 7.27
51
Data Collection

Types of variables

Categorical or continuous



Categorical variables reflect nominal scales
Continuous variables reflect ordinal, interval or
ratio scales
Independent or dependent


Independent variables are the purported
causes
Dependent variables are the purported effects
Measurement Instruments

Types of instruments (continued)

Affective (continued)

Scales used for responding to items on
affective tests





Likert
Semantic differential
Thurstone
Guttman
Rating scales
Examples of Items from a Likert Scale Measuring
Attitude toward Teacher Empowerment
Figure 7.14
Instructions: Circle the choice after each statement that indicates your opinion.
1. All professors of education should be required to spend at least six months teaching at
the elementary or secondary level every five years.
Strongly
agree
(5)
Agree
(4)
Undecided
(3)
Disagree
(2)
Strongly
disagree
(1)
Disagree
(4)
Strongly
disagree
(5)
2. Teachers’ unions should be abolished.
Strongly
agree
(1)
Agree
(2)
Undecided
(3)
3. All school administrators should be required by law to teach at least one class in a
public school classroom every year.
Strongly
agree
(5)
44
Agree
(4)
Undecided
(3)
Disagree
(2)
Strongly
disagree
(1)
Example of the Semantic Differential
Figure 7.15
Instructions: Listed below are several pairs of adjectives. Place a checkmark (a) on the line
between each pair to indicate how you feel. Example Hockey:
exciting :_____:_____:_____:_____:_____:_____:_____:_____: dull
If you feel that hockey is very exciting, you would place a check in the first space next to the
word “exciting.” If you feel that hockey is very dull, you would place a checkmark in the space
nearest the word “dull.” If you are sort of undecided, you would place a checkmark in the
middle space between the two words. Now rate each of the activities that follow [only one is
listed]:
Working with other students in small groups
friendly
happy
easy
fun
hot
good
laugh
beautiful
45
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
:_____:_____:_____:_____:_____:_____:_____:_____:
unfriendly
sad
hard
work
cold
bad
cry
ugly
Measurement Instruments

Issues for cognitive, aptitude, or affective
tests


Bias – distortions of a respondent’s performance
or responses based on ethnicity, race, gender,
language, etc.
Responses to affective test items




Socially acceptable responses
Accuracy of responses
Response sets
Problems inherent in the use of self-report
measures and the use of projective tests
Criterion-Referenced vs. Norm-Referenced
Evaluation Instruments
Page 158
Criterion-referenced: A student . . .
•
•
•
•
•
spelled every word in the weekly spelling list correctly.
solved at least 75 percent of the assigned problems.
achieved a score of at least 80 out of 100 on the final exam.
did at least 25 push-ups within a five-minute period.
read a minimum of one nonfiction book a week.
Norm-referenced: A student . . .
• scored at the 50th percentile in his group.
• scored above 90 percent of all the students in the class.
• received a higher grade point average in English literature than any other
student in the school.
• ran faster than all but one other student on the team.
• and one other in the class were the only ones to receive A’s on the midterm.
48
Selection of a Test

Designing you own tests


Get help from others with experience developing
tests
Item writing guidelines






Avoid ambiguous and confusing wording and sentence
structure
Use appropriate vocabulary
Write items that have only one correct answer
Give information about the nature of the desired answer
Do not provide clues to the correct answer
See Writing Multiple Choice Items
Selection of a Test

Test administration guidelines



Plan ahead
Be certain that there is consistency across
testing sessions
Be familiar with any and all procedures
necessary to administer a test
Identify Data Options: Select
Scales of Measurement


Nominal (Categorical): categories
that describe traits or characteristics
participants can check
Ordinal: participants rank order a
characteristic, trait or attribute
Identify Data Options: Select
Scales of Measurement


Interval: provides “continuous”
response possibilities to questions with
assumed equal distance
Ratio: a scale with a true zero and
equal distances among units
Record and Administer Data
Collection: Locate or Develop an
Instrument



Develop your own instrument
Locate an existing instrument
Modify an existing instrument
Record and Administer Data
Collection: Obtain Reliable and Valid
Data


Validity: the ability to draw meaningful and
justifiable inferences from the scores about a sample
or a population
Types of validity



Content (representative of all possible questions that could
be asked)
Criterion-referenced (scores are a predictor of an outcome
or criterion they are expected to predict
Construct (determination of the significance, meaning,
purpose and use of the scores)
Record and Administer Data Collection:
Develop Administrate Procedures for Data
Collection




Develop standard written procedures for
administering an instrument
Train researchers to collect observational data
Obtain permission to collect and use public
documents
Respect individuals and sites during data
gathering
Illustration of Types of Evidence of Validity
Figure 8.1
52
Reliability and Validity
Figure 8.2
53
Methods of Checking Validity and Reliability
Table 8.2, page 180
VALIDITY (“TRUTHFULNESS”)
Method
Content-related evidence
Criterion-related evidence
Procedure
Expert judgment
Relate to another measure of the same
variable
Assess evidence on predictions made
from theory
Construct-related evidence
RELIABILITY (“CONSISTENCY”)
54
Method
Content
Time
Interval
Test-retest
Identical
Varies
Give identical instrument twice
Equivalent
forms
Different
None
Give two forms of instrument
Equivalent
forms/retest
Different
Varies
Give two forms of instrument, with time
interval between
Internal
consistency
Observer
agreement
Different
None
Identical
None
Divide instrument into halves and score
each or use KR
Compare scores obtained by two or more
observers
Procedure
More About Research: Threats to Internal Validity
in Everyday Life
Box 9A, page 199
Consider the following commonly held beliefs:
• Because “failure” often precedes “suicide,” it is therefore the cause of “suicide.”
(probable history and mortality threat)
• Boys are genetically more talented in mathematics than are girls. (probable subject
attitude and location threats)
• Girls are genetically more talented in language than are boys. (probable location and
subject attitude threats)
• Minority students are less academically able than students from the dominant culture.
(probable subject characteristics, subject attitude, location, and instrumentation threats)
• People on welfare are lazy. (probable subject characteristics, location, and history
threats)
• Schooling makes students rebellious. (probable maturation and history threats)
• A policy of temporarily expelling students who don’t “behave” improves a school’s test
scores. (probable mortality threat)
• Indoctrination changes attitude. (probable testing threat)
• So-called miracle drugs cure intellectual retardation. (probable regression threat)
• Smoking marijuana leads eventually to using cocaine and heroin. (probable mortality
threat)
55
Illustration of Threats to Internal Validity
Figure 9.2
Note: We are not implying that any of these statements are necessarily true; our guess is that some are and some are not.
56
*This seems unlikely.
†If these teacher characteristics are a result of the type of school, then they do not constitute a threat.
General Techniques for Controlling Threats
to Internal Validity
Table 9.1, page 202
Threat
Subject characteristics
Mortality
Location
Instrumentation
Testing
History
Maturation
Subject attitude
Regression
Implementer
57
Obtain
Obtain More
More
Choose
Standardize Information Information Appropriate
Conditions on Subjects
on Details
Design
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Technical Issues

Validity (continued)

Consequential – to what extent are the
consequences that occur from the test
harmful


Estimated by empirical and expert judgment
Factors affecting validity



Unclear test directions
Confusing and ambiguous test items
Vocabulary that is too difficult for test takers
Technical Issues

Validity (continued)

Factors affecting validity





Overly difficult and complex sentence structure
Inconsistent and subjective scoring
Untaught items
Failure to follow standardized administration
procedures
Cheating by the participants or someone
teaching to the test items
Technical Issues

Validity – extent to which
interpretations made from a test score
are appropriate

Characteristics




The most important technical characteristic
Situation specific
Does not refer to the instrument but to the
interpretations of scores on the instrument
Best thought of in terms of degree
Technical Issues

Validity (continued)

Four types

Content – to what extent does the test
measure what it is supposed to measure



Item validity
Sampling validity
Determined by expert judgment
Download