|
Method & Designs |
Description |
|
Experimental Designs |
|
Pretest-Posttest randomized control-group
R O X O
R O X O |
Participants are randomly assigned to either experimental/
treatment group or control group – can assume groups
equivalent prior to treatment. Pre- post- tests are
administered for both groups.
Strengths:
Controls for virtually all threats to internal validity.
Weaknesses:
The main problem with this design is that it improves
internal validity but sacrifices external validity to do so.
There is no way of judging whether the process of
pre-testing actually influenced the results because there is
no baseline measurement against groups that remained
completely untreated.
|
|
Two-group
posttest-only randomized design
R X O
R X O
|
Designs
do not include a control group or a pretest
Strengths:
Simple. Posttest only,
so it is relatively inexpensive.
Because of random assignment can assume equivalence prior to
treatment. Testing maybe expensive or difficult so
eliminating pre-test may make the evaluation more feasible.
Weaknesses:
no baseline so cannot assess changes within group, can only
makes comparisons across groups.
|
|
Solomon
four-group pretest-posttest randomized design
R O X O
R O O
R X O
R O |
This design
includes involves two
experimental groups, E1 and E2, and two
control groups,C1 andC2. All four groups
complete posttest measures, but only groupsE1 andC1
complete pretest measures in order to allow for better
control of pretesting effects
Strengths: Detects
potential changes in the dependent variable due to some
interaction effect between the pretest and the treatment
Weaknesses:
Twice as many groups. Increased complexity. Doubles the time
and cost. |
|
Quasi – experimental designs |
|
Pretest–posttest non-equivalent-control-group
N O X O
N O X O |
Designs
include a pretest and a posttest but do not require a
control group. It is structured like a pretest-posttest
randomized experiment, but it lacks random assignment
Strengths:
makes it possible to determine if groups were equivalent
prior to treatment
Weaknesses:
prior differences between groups may affect the measure of
the dependent variable. |
|
Posttest-only
non-equivalent control-group
N X O
N X O |
Same as
above, but design does not include a pretest
Strengths:
can be used to examine the impact of a treatment after the
fact.
Weaknesses:
not very robust. Susceptible to numerous threats to validity
such selection bias. |
|
Time Series
O O X O
O |
The researcher
makes several observations of behavior over time prior to
and then immediately after introduction of an IV.
Strengths:
This establishes a baseline
prior to the treatment and examines the stability of
behavior after the treatment.
Weaknesses:
history is a potential threat to internal validity |
|
Single group
pre-post test
O X O
|
A group
of participants takes a pretest and then receives the
experimental training. Afterwards a posttest is given.
Strengths:
pre-test establishes a
baseline to measure change.
Weaknesses:
single group designs are susceptible to numerous threats to
validity including history, maturation, testing, regression,
etc. |
|
Single group
post test only
X O |
A group
of participants receives the experimental training and
afterwards is given a posttest.
Strengths:
Can assess criterion
performance level however cannot make attributions of
performance to the treatment/training.
Weaknesses:
same as for single group pre-post test. |
