5. Establish what research design is appropriate for
this study…
•Problem: A researcher has three different groups of employees
(engineers, administrators, sales personal) which are not
performing at the expected level. Data shows three different levels
of current performance: high, average, and low.
•Purpose: To determine the effects of different types of feedback
on employees’ performance
•Research Question: What is the effect of different types of
feedback over the performance of employees?
• Values of the independent variable: positive, negative, no
feedback
6. Establish what research design is appropriate for
this study…
•Problem: Faculty morale is low at a university according to
the results of a questionnaire applied.
•Purpose: To establish the effect of sensitivity training on
faculty morale
•Research Question: What is the effect of sensitivity training
workshops on faculty morale?
• Values of the independent variable: treatment, no
treatment
7.
8. Establish what research design is appropriate for
this study…
•Problem: A new calculus program has been suggested to
improve students performance
•Purpose: To establish the effect of the suggested program
on students’ calculus scores
•Research Question: What is the effect of the suggested
calculus program on students’ scores?
• Values of the independent variable: treatment, no
treatment
9. Now do the same with your own study…
•Problem:
• Research Question:
• Research Design:
•Research Method:
10. Sharing Your Experience
1. What was the most clear point?
2. What was the least clear point?
3. What would you like to say about the experience?
4. Any specific exercise you want to talk about?
12. Ex Post Facto Research
• Description: Variables are studied after the fact has occurred without
interference from the researcher. Possible relationships and effects are
searched. It is applied in contexts in which it is not possible or acceptable
to manipulate the characteristics of human participants or in situations in
which it is not practical or ethically acceptable to apply the full protocol of
a true experimental design. They are widely used in social sciences.
• Types:
– Causal-Comparative Design
– Correlational Design
13. Causal-Comparative Design
• Description: Researchers try to establish the possible cause-effect
relationship that already exists between variables in a
nonexperimental setting. There are comparison groups rather than
experimental and control groups.
• Basic approaches:
– Retrospective
– Prospective
14. Correlational Design
• Description: It determines whether or not two variables are correlated,
which means that an increase or decrease in one variable corresponds to
an increase or decrease in the other variable. It is very important to note
that correlation does not imply causation, just an association.
• Types :
– Relational Design
– Prediction design
15. Survey Design
• Description: It intends to describe a situation using the answers
to questions
• Types of Designs:
– Longitudinal Design
• Trend Study
• Cohort Study
• Panel Study
– Cross-Sectional Design
• Types of Surveys:
– Questionnaire
– Interview
16. Difference-in-Difference Estimation
The simplest set up is one where outcomes are observed for two groups
for two time periods. One of the groups is exposed to a treatment in the
second period, but not in the first period. The second group is not
exposed to the treatment during either period. Then, the researcher
estimates the “difference of differences”
An estimator, not a design. Used to study policy questions by estimating
treatment effects with nonexperimental data
17.
18. Example
A study about the effect of the higher minimum wage in fast-food restaurants
in Pennsylvania (where the wage is constant) and New Jersey (where the wage
has been increased). These are some of the results:
State Before Increase After Increase Difference
New Jersey
(treatment)
20.44
(0.51)
21.03
(0.52)
0.59
(0.54)
Pennsylvania
(control)
23.33
(1.35)
21.07
(0.94)
-2.16
(1.25)
Difference -2.89
(1.44)
-0.14
(1.07)
2.76
(1.36)
Note: Standard errors in parentheses
Taken from: Albouy (n.d)
24. Statistical Regression
Description: An effect caused by a tendency for subjects
selected on the basis of extreme scores to regress
toward an average performance on a subsequent test. It
is also known as “regression to the mean”
25. Selection
Description: An effect due to the groups of subjects
not being randomly assigned to groups; a selection bias
is operating such that the groups are not equivalent
26. Mortality
Description: An effect due to subjects dropping out of
the experiment. The subjects that stay may be more
motivated or capable and that affects the results
27. Selection- Maturation Interaction
Description: An effect of maturation not being
consistent across the groups because of some selection
factor may lead to confusing results and an erroneous
interpretation of the effect of the treatment.
29. Interaction Effect of Testing
Description: Pretesting interacts with the experimental
treatment and causes some effect such that the results
cannot be not generalized to an unpretested population
30. Interaction Effects of Selection
Biases and the Experimental
Treatment
Description: An effect of some selection factor of intact
groups interacting with the experimental treatment that
would not be the case if the groups were formed
randomly
31. Reactive Effects of Experimental
Arrangements
Description: An effect that is due to the artificial or novel
experimental setting. It may also threaten internal
validity
32. Multiple-Treatment Interference
Description: When the same subjects receive two or more
treatments, there may be a carryover effect between
treatments and therefore the results cannot be
generalized to single treatments
44. Now you know
1. The pre-experimental designs
2. What difference-in-difference
estimation is
3. Different types of threats to validity
45. Reference
Albouy, D. (n.d). Program evaluation and the difference in
difference estimator. Retrieved from
http://eml.berkeley.edu/~webfac/saez/e131_s04/diff.pdf
Wiersma, W. & Jurs, S.G. (2009). Research methods in education:
An introduction (9th ed.). Boston, MA: Allyn and Bacon
46. What’s Up at MN & DN this summer?
1. VERY IMPORTANT – backwards map your summer work
2. 30 day writing challenge, 30 day work-life balance challenge, and 365s -
keep you in touch with your work
3. Group work – RLC writing OR
Lingerers
47. Avoid this one
1. VERY IMPORTANT – backwards map your summer work
2. 30 day writing challenge, 30 day work-life balance challenge, and 365s -
keep you in touch with your work
3. Group work – RLC writing OR Lingerers