Chapter 11- Single Case, Quasi-Experimental, and Developmental Research
There are 3 types of special research situations:
o Single case experimental design- form of experiment with only one participant. Was
derived to see if an experimental manipulation had effect on single research participant.
The subject’s behaviour is measured over time during baseline control period. Then
manipulation put in during treatment periods and subject observed again for changes.
Only controversy is that the reason for the change could easily be something else than the
manipulation hard to tell with one subject.
o Reversal design- form of experiment where the reversibility of the manipulation is
tested. This is done by observing the baseline (A) , then the treatment period (B), then
observing the second baseline again after the manipulation is withdrawn again, aka
ABA design : A(baseline period) B(treatment period) A (baseline period)
A way to improve it is to have a ABAB design or a ABABAB design where you
could test the manipulation once. Just testing the reversibility once doesn’t
validate anything because it could have been a random fluctuation
o Multiple Baseline Design- effectiveness of treatment is demonstrated when a behaviour
changes only after the manipulation is introduced.
o Single case designs are good when you are trying something new and want to see if its
effective. Like a teacher trying a new teaching technique. But grouping your findings
from a whole bunch of single case designs can be misleading and inaccurate. Its valuable
for someone who is applying some change technique in a natural environment.
o Program Evaluation: research on programs taht are prposed and implemented to
achieve some positive effect on a group of individuals
In school the DARE( Drug Abuse Resistance Education) designed to reduce drug
abuse is an example
o There are 5 types of evaluations:
Needs Assessment- this study asks if there are any problems that need to be
addressed in a target population
Program Theory Assessment- once the needs are set then you ask what
program should be made to address the needs. It should be based on valid
assumptions about the causes of the problem and the rationale of proposed
program. Process Evaluation- aka program monitoring. Determines whether its research
is reaching the target population, whether its attracting enough clients and if the
services are getting provided as planned.
Outcome Evaluation- aka impact assessment. To see if the intended outcomes
being realized or if the goal being achieved.
Efficiency Assessment- once intended effect is shown by the program to should
be analyzed to see if its actually worth it. Cost-benefit analysis should be made.
Quasi-experimental designs- the study the effects of the independent variable in settings in
which control features of true experimental design cannot be achieved.
o There are many but 6 ones usually used are:
One group posttest only design- aka one-shot case study. Has no control group
or pretest comparison; very poor design in terms of internal validity. Example of
that is information session followed by a test at end. People do well and
concluded information session helped but don’t really know since you don’t
know what the score was before the session
One group pretest-posttest desing- effect of an independent variable is inferred
from the pretest-posttest diff in a single group.
History effects- As a threat to the internal validity of an