Quasi Experiment
Quasi Experiment
Quasi Experiment
CONCEPCION G. VELUZ
JUVARTA O. TIERRA
Methods of Research
Summer 2011
What is quasi experiment?
The word "quasi" means as if or almost, so a quasi-
experiment means almost a true experiment.
A quasi-experiment is an experiment in which the subjects to be
observed are not randomly assigned to different groups in order
to measure outcomes, as in a true experiment, but grouped
according to a characteristic that they already possess.
en.wikipedia.org/wiki/Quasi-experiment
Similar to true experiments. Have subjects, treatment, etc., but
uses nonrandomized groups. Incorporates interpretation and
transferability in order to compensate for lack of control of
variables.
writing.colostate.edu/guides/research/glossary/
Research design in which variation in the independent variable is
natural, or not manipulated by the researcher; participants are not
assigned randomly to treatment and control groups; also
referred to as natural experiment
highered.mcgraw-hill.com/sites/0073049506/student_view0/glos
Non-random assignment of subjects to conditions; the
experimenter lacks direct control over the independent variable;
also called a natural experiment
www.hettler.com/Direct/Assess/RESEARCH%20TERMS.htm
Quasi-experimental designs
came about because of:
1) difficulty of applying the classical
natural science method to the social
sciences
2) overemphasis on theory testing and
development
3) high cost of classic natural science
methods
4) development of new statistical tools
that allowed for statistical control
Some common characteristics of quasi-
experiments include the following:
matching instead of randomization
is used
time series analysis is involved
the unit of analysis is often
something different than people
Advantages
Can be led in non-experimental situations,
i.e. in "real" contexts
Can be used when the treatment may
become too "heavy", i.e. involves more than
2-3 well defined treatment variables.
Disadvantages
In quasi-experimental situations, you really
lack control:
You dont know all possible stimuli (causes
not due to experimental conditions)
You cant randomize (distribute evenly other
intervening unknown stimuli over the groups)
You may lack enough subjects
There are several types of quasi-
experimental designs, including:
Time Series Design
Control Group Time Series Design
Equivalent Time Samples Design
Equivalent Materials Design
Non-equivalent Control Group Design
Counterbalanced Designs
Separate Sample Pretest-Posttest Design
Separate Sample Pretest-Posttest Control
Group Design
Multiple Time Series Design
Recurrent Institutional Cycle Design
1. Time-Series Design
You make a few observations to establish
a baseline, do the intervention, and then
make a few more measurements. (Just
how many observations you make on
either side of the intervention is
determined by the stability of the data.
This will be covered in advanced
concepts.)
The major threat to the internal validity of
the time series is history. That is, a charge
that the results obtained would have
occurred with or without the experimental
intervention is difficult to defend with data
from the simple Time Series experiment.
2. Control Group Time Series Design
Another name for this quasi-
experimental design is the "Multiple
Time-Series Design".
This design is simply the Time
Series Design with the addition of a
comparison group. (Ideally, the
experimental treatment 'X' would be
randomly assigned to one group or
the other.)
3. Equivalent Time-Samples Design
In actual practice, you do not regularly
alternate X1 (intervention) with X0 (control);
rather, the selection of X1 or X0 should be
determined randomly.
This design is useful when there is reason to
suspect that the experimental intervention
produces results that are transient or
reversible in character. There is only one
participant group (or even just one
participant, in N of 1 studies). On one
occasion the experimental treatment is
applied. On another occasion an alternate
treatment, or no treatment, is applied.
4. The Equivalent Materials Design
This design is similar to the
equivalent time samples design, but
the different experimental situations
include different though equivalent
materials.
This design is useful where the
effects of the experimental treatment
are enduring, so that the different
treatments and repeats of treatments
must be applied to non-identical
content.
5. Nonequivalent Control Group Design
This design is identical to the Pretest-
Posttest Control Group design, with
the exception of randomization.
Comparison groups are naturally-
occurring but chosen to be as similar
as possible in all experimentally-
relevant aspects. If possible, the
choice of group that receives the
experimental treatment should be
made randomly.
6. Counterbalanced Designs
This is actually a family of
designs, generically referred to
as a Latin-square arrangement.
Other names for such designs
have been rotation experiment,
cross-over design, and switch-
over design.
Depicted here:
Four naturally-occurring groups
(i.e., you did not randomly select
participants from a larger
population nor did you randomly
assign them to the four groups).
Four different experimental
treatments, or variations on the
same treatment, to be compared.
Four different occasions, which
allows each group to receive all
four treatments, but in a different
order for each group.
8. Separate Sample Pretest-Posttest
Control Group Design
You should recognize the
previous design, the Separate-
Sample Pretest-Posttest design,
in the upper half. Added to this
are two comparison groups
minus any sort of treatment: one
group takes the pretest and one
group takes the posttest. If some
sort of historical influence has
led to a change in posttest
scores relative to pretest scores,
this design provides the
information to detect such an
influence.
The multiple Time-Series design simply adds an
equivalent control group to the Time Series
design. This provides some information about
what might have happened to your
experimental group had the experimental
treatment not been applied. Since naturally-
occurring groups are used instead of
randomization, you have less confidence in the
equivalence of the two groups; still, the
addition of the control group does add some
credibility to any positive results by helping to
rule out potential rival explanations based on
extraneous factors.
9. Multiple Time Series Design
10. Recurrent Institutional Cycle Design
This is actually more of
a concept than a
specific design. This
approach is appropriate
in situations where a
treatment is repeatedly
being applied, on a
cyclical basis, to a new
group of respondents.
What you basically do is
fine-tune the design on
subsequent cycles to
investigate questions
raised on previous
cycles.
Quasi-experimental thesis
example
Notari, Michele (2003). Scripting
Strategies In Computer Supported
Collaborative Learning Environments,
Master Thesis, MSc MALTT (Master of
Science in Learning and Teaching
Technologies), TECFA, University of
Geneva.
This master thesis concerns the design
and effects of ICT-supported activity-
based pedagogics in a normal classroom
setting
Target: Biology at high-school level
(various subjects)
Three research questions
formulated as 'working
hypotheses':
The use of a Swiki as collaborative editing tool causes no
technical and comprehensive problems (after a short
introduction) for high school students without experience in
collaborative editing but with some knowledge of the use of a
common text-editing software and the research of information
in the Web.
[Pedagogical] scripting which induces students to compare
and comment on the work of the whole learning community
(using a collaborative editing tool) leads to better learning
performance (as assessed by pre- and post-testing) than a
script leading students to work without such a tool and with
little advice or / and opportunity to make comments and
compare their work with the learning community.
The quality of the product of the working groups is better
(longer and more detailed) when students are induced to
compare and comment on their work (with a collaborative
editing tool) during the learning unit.
Method
The whole research took place in a normal
curricular class environment. The classes were
not aware of a special learning situation and a
deeper evaluation of the output they produced.
The researcher tried to embed the scenarios in
an absolutely everyday teaching situation and
supposed students to have the same
motivational state as in other lessons.
To collect data the researcher used
questionnaires, observed students while working,
and for one set up asked students to write three
tests.
The students asked about the purposes of the
tests. The researcher tried to motivate them to
perform as well as they could without telling them
the real reason of the tests.
This master theses concerns several quasi-
experiments, all in real-world settings. Here is
the reproduced settings for one of these.