LANDINGIN, Rossa Victoria M.

08-45456
OC109.2 (TFI)

QUASI-EXPERIMENTAL DESIGN
(QED)

Quasi Experimental Design DEFINED

Quasi dictionary definition (Cambridge Online Dictionary):

Used to show that something is almost, but not completely, the thing described

- Quasi-Experimental Design is a type of research design that looks similar to Experimental Design
but lacks one key ingredient: The randomization (random assignment) of subjects to the
treatment and comparison groups.
- Even without the randomization, the QED can still be useful in researching and analyzing cause
and effect types of relationship.
- The QED can be a very powerful tool, especially in situations where „true‟ experiments are not
possible or are hard to execute.

Example:
A quasi experiment constructed to analyze the effects of different educational programs on two groups of
children, for example, might generate results that show that one program is more effective than the other.
These results will not stand up to rigorous statistical scrutiny because the researcher also need to control
other factors that may have affected the results. This is really hard to do properly. One group of children
may have been slightly more intelligent or motivated. Without some form of pre-testing or random
selection, it is hard to judge the influence of such factors.

Matching

- Instead of randomization, MATCHING is used in the QED.

- This is the process of coming up with exposure and comparison groups
- The usual traits used to serve as criteria for matching are age, sex, gender, social status, and
educational attainment.

Experimental and Quasi-Experimental Design COMPARED AND CONTRASTED

Experimental Design Quasi-Experimental Design

- Random assignment of subjects to the - There is no random assignment
control and experimental groups - Instead, MATCHING is done to determine
which subjects are part of the treatment
and comparison groups
- The treatment is randomly assigned to - No treatment is randomly assigned
groups
- Purpose of the experiment is to research and analyze cause and effect relationships
- The Independent Variables can be varied or manipulated so its effects on the dependent variable
can be measured more effectively

Advantage and Disadvantage of Quasi-Experimental Design

Advantage
 - When pre-selection and randomization of groups are difficult, QEDs can be very useful in
generating results for general trends. such an act reduce the time and resources needed in
conducting experiments.

Disadvantage
- "Without proper randomization, statistical tests can be meaningless. These experimental designs
may not take into account any pre-existing factors that are necessary to the experiment being
made.

Research example for the QED‟s disadvantage

To study the effect of maternal alcohol use when the mother is pregnant.

Method:
- Ask people how much alcohol they used in their pregnancy and then assign them to groups using
the alcohol intake as the measure/criteria.

Disadvantage in the use of QED:

- An external influencer such as the reason behind the mothers' alcohol intake during pregnancy
which may have an impact on the results of the experiment may not be taken into consideration.
So the study may not be as meaningful as it could possibly be.

Internal and External Validity in relation to Quasi-Experimental Design

"Quasi experiments enable us to rule out some threats to validity because they include more data points
than the pre-experiments. The number of quasi-experimental designs that a creative researcher can
construct is limitless." (Schuester, 1963)

Internal Validity:
- Internal Validity is the "approximate" truth about inferences that regard causal relationships
(cause-effect). Hence, it is only applicable in studies that try to establish such a relationship.
Internal Validity is only relevant to the study you are doing and it cannot generalize like External
Validity. This type of validity illustrates how what you did in the study caused what you were able
to observe in the effect; it does not determine whether what you did was your goal or not.

Example:
A study to look at the effects of a new computerized tutoring program on math performance in Grade1
students.

Imagine that the tutoring is unique in that it has a heavy computer game component and you think that's
what will really work to improve math performance. Finally, imagine that you were wrong--it turns out that
math performance did improve, and that it was because of something you did, but that it had nothing to
do with the computer program. What caused the improvement was the individual attention that the adult
tutor gave to the child -- the computer program didn't make any difference.

This study would have internal validity because something that you did affected something that you
observed --you did cause something to happen. But the study would not have construct validity,
specifically, the label "computer math program" does not accurately describe the actual cause
(perhaps better described as "personal adult attention.”

External Validity:
- External Validity is the type of validity almost always related to generalizations. Basically, External
Validity is "the degree to which the conclusions in your study would hold for other persons in
other places and at other times." This type of validity uses the result of an experiment done to a
sample population as the result for the whole population.
List of Threats to Internal Validity

Regression
- When subjects are selected because of extreme scores on some type of instrument, there is
tendency for their scores to move more toward the average on subsequent tests

Instrumentation
- To overcome the testing threat to internal validity, a researcher develops a different form of the
test instrument, but it is not really equivalent

Testing
- Whenever you give a pretest, the students may remember the test questions, and get them
correct on the posttest

Mortality
- Some people drop out during an experiment. This may affect the outcome

Maturation
- People naturally change and evolve over time. This may cause the difference

History
- Another event occurs during the time of the experiment that might cause the difference

Selection
- The selection of the participants in the sample population may not always be the most effective
group for the experiment to succeed

Retrospective (Ex Post Facto) Design

- “After the fact”
- There is a control group or a comparison group
- Intact groups are used
- Simulated after-only experimental designs without randomized assignment
- Focuses first on the effect, then attempts to determine what caused the observed effect

Threats to Internal Validity: Selection, History, Maturation

Prospective Quasi-Experimental Design

- Similar to the retrospective design except that it has variations in the independent variable that
are measured as they occur
- Are said to be more persuasive than retrospective design “especially when the independent
variable occurred long ago” (Ellis, 1994)

Threats to Internal Validity: Selection, History, Maturation, Mortality

Time Series Design

- Time series QEDs have structures similar before-after no control group experiments
- They involve following a dependent variable over time in a single group of subjects/population
and observing whether the values of the dependent variable change in apparent response to
changes in an independent variable
- Reversal in a time series design refers to time frames in which the independent variable goes
back to its initial state
- Two different populations may be monitored over extended periods of time

Threats to Internal Validity: Selection, History, Maturation, Mortality

 Pretest-Postest Non-Equivalent Group Design
- The two groups have not been equated prior to treatment
- One can try to mitigate the problem by assigning subjects to groups (or selecting intact groups) in
ways that make it likely that the groups do not differ greatly prior to the treatment, but one always
worries about the unknown variables on which the groups might differ and which might affect the
criterion variable

N O X O
Threats to Internal Validity: Selection, History, Maturation, The design‟s
inability to assume that the populations being compared are equivalent on all N O O
things prior to the treatment since one has not randomly assigned subjects to
their respective groups

Regression Discontinuity Design

- This design looks a lot like the pretest-posttest nonequivalent groups design, but the groups are
nonequivalent by choice
- The ‘C’ the first column indicates that the subjects are assigned to groups based on their score
on the covariate (the pretest).

Example: During the report! C O X O

Threats to Internal Validity: Selection, History, Maturation, Testing,
Instrumentation C O O

Criterion Group Design (Causal Comparative)

- In this approach, the objective is to discover possible causes for a phenomenon being studied by
comparing the subjects in which the variable is present with similar subjects in whom it is absent
- It is seen as a design that bridges the gap between descriptive research methods on the one
hand and true experimental research on the other

References:

Cited on the cut and paste