Lecture 2

advertisement
Lecture 2
Research Questions: Defining and
Justifying Problems; Defining
Hypotheses
Announcements:
SPSS installed on following PC’s:

Upstairs lab
Bisque
Citron
Crimson
Ebony
Ruby
Olive

Basement
Lavender
Sienna
2
Our Focus Today:
What makes a good research problem?


Research Questions for Theory Development
Research Questions for Practical Application
Turning research problems into testable
hypotheses
3
Purpose of Research
To increase knowledge within a discipline
or an area of study.
To increase knowledge as a consumer of
research and to understand research
within a discipline or area of study.
4
Increasing Knowledge Within a
Discipline or Area of Study
For Theory Development
Practical Application
Developing Research Tools
5
Defining Research Problems
What is a problem?



“an interrogative sentence or statement that asks:
What relation exists between two or more concepts?”
A problem can be restated in one or more ways to
produce testable hypotheses.
A good research problem often produces more than
one testable hypothesis.
6
Characteristics of good research
problems
Should state the concepts or variables to
be related clearly and unambiguously
Should be testable
Should be feasible, given resources
7
Three Specific Criteria for a
Research Problem
What are we going to learn as the result of
the proposed project that we do not know
now?
Why is it worth knowing?
How will we know that the conclusions are
valid?
8
The Research Question
Common mistakes in defining research
questions




Very broad area of interest
Too narrow
Cannot be measured
Problem is trivial or already understood
9
Problem: Too Broad
Very broad area of interest


“I want to understand how people use the
Internet”
“What factors influence the use of an
interface?”
Solutions?
10
Problem: Too Narrow
Too narrow
“Do Females Use Technology X more than
Males?”
Solutions?
11
Problem: Cannot be Measured
Cannot be measured
“Will this new information technology make society
better?”
Solutions?
12
Problem: Trivial or Previously
Answered Research Questions
This is why we actually use literature–
even in applied, business, or exploratory
research.
Bringing an ‘old’ problem to a ‘new’
discipline is not necessarily trivial.
13
So, what is a good research
problem statement?
“The research problem is to investigate the
presumed effect of A, B and C on X and Y
in (population).
14
Moving from General to Specific
“Could use of technology X affect society in a
positive way?”
“If we looked at two populations, one using
technology X and one not using it, would they
differ?”
“How is the use of technology X related to
productivity and work satisfaction in task groups
within population Y?”
15
Example from Week 1
Income
Job
Prestige
Socioeconomic
Status
Academic
Achievement
Academic Ability
Grades
Math skills
Level of
Schooling attained
Language
skills
16
Implications of Research Questions
for Statistical Analysis
17
Type of Research Question
Statistics
Descriptive
Mean, frequency
Complex Descriptive
Cross-tabulations, factor
analysis
Single factor difference questions
T-test, one-way ANOVA
Complex or multifactor difference
questions
Factorial ANOVA
Basic associational questions
Correlation
Complex or multivariate
associational questions
Multiple Regression
18
Justification
19
Justifying Research Problems
Explain what is not known about the problem.
Why does the problem matter?
Provide documentation that this is actually a
problem.


Available statistics?
Available literature that shows that this is a needed
area of inquiry?
20
What is not a Justification?
No one has looked at it before.
Literature has failed to address the issue.
You think its interesting.

If it is ‘interesting’ then there is probably a
justification buried in there, but you have to
spell it out.
21
Justification as Significance of the
Study (Creswell 2003)
What are the ways that the study will add
to the scholarly research/literature in the
field?
How does the study improve practice?
How might the study improve policy?
22
Turning Research Questions into
Testable Hypotheses
23
Inductive Logic of Research in
Qualitative Studies
Generalizations are made, or Theories to Past Experience
And Literature
Researcher Looks for Broad Patterns, Generalizations, or
Theories from Themes or Categories
Researcher Analyzes Data to Form Themes
Or Categories
Researcher Asks Open-Ended Questions of Participants
Or Records Field Notes
Researcher Gathers Information
24
The Deductive Approach in
Typical Quantitative Research
Researcher Tests or Verifies a Theory
Researcher Tests Hypotheses or Research Questions
From the Theory
Researcher Defines and Operationalizes
Variables Derived from the Theory
Researcher Measures or Observes Variables Using an
Instrument to Obtain Scores
25
Why not just rely on pure
observation?
Subjectivity

“group A is nicer than group B”
Recall

What did you say to me last week about topic
X?
Interpretations or conclusions that lack
convincing support

“most kids don’t care what their parents say”
26
Hypotheses
A good research question will produce one
or more testable hypotheses.
Testable hypotheses predict a relationship
between variables (not concepts).
27
Three Basic Kinds of Hypotheses
Descriptive Questions

Single variable descriptions
Central tendency, variability, percentages
Associational

Non-directional relationship between variables.
Difference

Group Comparison
28
Null hypothesis
Null Hypothesis:

H0: μ1 = μc
μ1 is the intervention population mean
μc is the control population mean
In English…

“There is no significant difference between the
intervention population mean and the control
population mean”
29
Alternative Hypotheses
Alternative Hypotheses:



H1: μ1 < μc
H0: μ1 > μc
H0: μ1 ≠ μc
30
Alternative Hypotheses
Non-directional hypotheses

Associations, not causal
Directional


Increase in A increases B
Decrease in A decreases B
Inverse


Increase in A decreases B
Decrease in A increases B
31
Conventions in Stating Hypotheses
Null hypothesis often not stated

Completely depends on convention in a given
discipline
Three basic approaches to using variables in
hypotheses:



Compare groups on an independent variable to see
impact on dependent variable
Relate one or more independent variables to a
dependent variable.
Describe responses to the independent, mediating, or
dependent variable.
32
Things to consider when stating
hypotheses
Know what you want to explain: dependent
variable

One common problem is under-specifying the key DV
or DV’s
The independent variable(s) should have
variation
Consider more than one independent variable,
especially factors for which you might want to
“control”
33
Exploratory versus Confirmatory
Exploratory Research

Often just testing to see if there are
associations between one or more variables.
Confirmatory Research

The more your topic has been researched,
the more likely that you will be engaging in
some type of confirmatory research.
34
An Example Model
Socioeconomic
Status
Academic
Achievement
Academic Ability
35
Next Week:
Causation, Validity and Reliability
Read over the online “Layman’s Guide” to
Research Methods
36
Download