5. Types of Quasi Experiments
One group pre-test post-test design
Non equivalent control group design
Interrupted time series design
Time series with non-equivalent control
group designs
6. One group pretest- post test design
represents a
pre- experimental design
May be called a bad experiment
O1 X O2
e.g Children’s attitudes toward drugs.
11. Manipulation of the independent variable
Pretest for all of the comparison groups
Posttest for all of the comparison groups
No random assignment to the comparison
groups (which as you can imagine is
going to cause some problems with this
design as compared to the strong
“randomized” experimental designs.
12. It was hypothesized that
Environmental changes associated with old
age contribute, in part, to the feelings of
lose, inadequacy, and low self esteem
among the elderly.
Quasi experiment was carried out in
nursing home
Two groups were used one experimental
and one comparison group.
13. Responsibility induced
condition.
Participants were given
a plant to take care of
it(if they decided to
accept it).
No responsibility
induced condition
Participants were given
a plant (whether they
chose to have one or
not), nurses have to
take care of the plant
14. Questionnaire were given that contained
the items related to “ how much control
they felt over general events in their life
and how happy and active they felt
(Langer & Rodin,1976)
Ratings by staff members on traits e.g
alertness, sociability and activity.
Posttest measure of social interest by a
competition
So there was a list of dependent
variables perception of control happiness
, activity, interest level and so forth.
15. Results
Residents in responsibility induced
group were more happier, active and
alert.
Movie attendance favored
responsibility induced group
10 residents from this group entered in
jelly bean competition but only one
from control group.
16. It was suggested that negative consequences of
aging can be reversed or reduced by giving the
elderly opportunity to make personal decisions
and to feel competent
17. Threats to the internal validity in non equivalent
control groups
(Langer and Rodin Study )
Non equivalent control group designs
generally control for all the major classes of
potential threats to the internal validity
excepts those due to additive effects of
Selection-maturation
Selection-history
Selection-instrumentation
Differential statistical regression
(Cook & Cambell, 1979).
18. External Validity
Replication of the study is necessary
One researcher must live with the fact that one
study is not likely to answer all questions about
a research hypothesis
20. Interrupted time series design
A time series is simply a set of
measurements of a variable taken at
various points in time.
e.g mood of students during semester
21. An interrupted time-series design, a time series
like this (the dependent variable) is interrupted
(usually near the middle) by the manipulation of
the independent variable.
O1 O2 O3 O4 O5 X O6 O7 O8 O9 O10
multiple pretests
a treatment (i.e., an intervention to be studied)
multiple posttests
(Gribbons & Herman, 1997)
23. Simple interrupted time series design
can be used to assess the effect of a
treatment in a situation such as when
A new product is introduced
A new social reform is instituted
Special advertising begun
26. Time Series with Nonequivalent Control Group
In a time series with nonequivalent
control group design, researchers make a
series of observations before and after
treatment for both a treatment group and a
comparable comparison group.
O1 O2 O3 O4 O5 X O6 O7 O8 O9 O10
------------------------------------------------
O1 O2 O3 O4 O5 O6 O7 O8 O9 O10
28. Easy to conduct
Lack of Randomization (no ethical concerns)
Can be performed in natural settings
Performed with groups
Efficient in longitudinal study
Provide comparisons ( treatment and control
groups)
29. Potential for Non-Equivalent Groups
Potential for Low Internal Validity
30. Researches
Echabe, (2013) used quasi experiment to analyze the
effect of the catastrophe of Japan on previous
stereotypes about Japanese people sample, entitled
Stereotype, emotion, and behavior: a quasi-experiment
on the effects of Japan’s earthquake.
Sample
N= 238 adults
Two phases of the study
126 adults (month and a half to a week before earth
quake
112 adults two weeks to month and a half after earth
quake.
31. Participants were asked to indicate the extent to which they thought
A)Japan was more powerful and influential in the world that their
own country
B) had incompatible and conflicting political and economic goals
and interests relative to their own country or
Participants were then asked to express their image (stereotype) of
Japanese people (warmth & competent)
Participants were also asked to express their overall attitude
toward Japanese people
32. the earthquake and its after-effects changed, at least temporarily,
their perception of the power of Japan and the compatibility
between its interests and those of participants’ own country.
As a result, the stereotype of Japanese people became warmer but
less competent.
The emotional reactions aroused by Japanese people became more
positive. After these events, participants felt more respect and
sympathy and less disdain and indifference.
Finally, the participants manifested a stronger willingness for
cooperation and lower disposition toward competition and
avoidance
33. Cont…
Impact of legislation requiring moped and
motorbike riders to wear helmets (Balart & Riba 1995).
Research design
Interrupted time series design with non equivalent groups to
investigate the effect of these measures on the number of serious
injuries and number of deaths.
The results indicate a significant decrease in the number of serious
cases in the treated group (smaller motorcycles and mopeds) ,
which does not occur in the control group (larger motorcycles).
34.
35. References
Ballart, X. & Riba , C. (1995). Impact of legislation requiring moped
and motorbike riders to wear helmets. Evaluation and Program
Planning,18 (4), 311-320
Campbell, D., & J. Stanley. (1963). Experimental and
quasi-experimental Designs. Chicago: Rand McNally.
Cook, T.D. & Campbell, D.T. (1979). Quasi-experimentation: design
and analysis issues for field settings. Retrieved from
http://books.google.com.pk/books/about/Quasi_experimentation.ht
ml?id=OKdEAAAAIAAJ&redir_esc=y
Echabe, A. E. (2013).Stereotype, emotion, and behavior: a
quasi-experiment on the effects of Japan’s earthquake. Journal of
Applied Social Psychology, 43, 2083–2089. doi:
10.1111/jasp.1216.
36. Referenes
Glass, G. V, Wilson, V. L., & Gottman, J. M. ( 1975). Design and
analysis of time-series experiments. Boulder, CO: Colorado
Associated University Press.
Gribbons, B., & Herman, J. (1997). True and quasi-experimental
designs. Practical Assessment, Research & Evaluation, 5(14).
Retrieved from http://PAREonline.net/getvn.asp?v=5&n=14
Shaughnessey, J. J., Zechmeister, E. B., & Zechmeister, J. S. (2012).
Research methods in Psychology. USA: McGraw-Hill company.