CH 11 Flashcards
Single Research Designs do not need…
random assignment or controls.
Who has used this design?
Pavlov, Skinner, Piaget, Ebbinghaus, Lorenz.
What was Lorenz famous for?
Species specific behavior of animals on his farm.
Piaget employed single case designs with…
developmental ideas based on his own children.
Growth of what helped them come back into fashion….
growth of Research and behavioral therapy.
What was Skinner’s argument for small end designs?
one major problem w large end designs is averaging, which may not be representative of any one subject. Averages can be misleading.
Skinner said that statistics isn’t necessary for analysis and potentially misleading due to averaging problem.
Instead, you should graph results, stare at the graph, look at pattern of data, and if there isn’t a clear pattern statistics won’t matter.
Critics said his results couldn’t generalize, but he said unwarranted because you could generalize results from a single subject to a population and even another species and did this.
What applications has Skinner’s legacy lead to?
behavioral therapy biofeedback autistic children mentally handicapped children use of teaching primates sign language etc.
Advantages to small end research?
developmental problems and as children grow are not great with large end designs
when you can’t find enough subjects
when researchers are interested in a more complete understanding of individuals
only need a small number of subjects which is advantageous for your time and resources, but doesn’t always pan out as an advantage. Trade amount for length.
Ethical concerns of these.
if using a control group w a placebo, how do you justify not giving effective treatment to the control group?
Comes up often, ex: AZT for treatment of AIDS.
Unethical to withhold potentially effective treatment for an experimental design.
Advantages again.
Individual performances are not disguised by averaging.
Types of single case/subject research designs.
AB
ABA (reversal designs)
ABAB
Explain AB design.
A (pre) B (post)
if treatment was effective ,you should see difference between A and B.
IF there is a change in behavior between A and B, can we really conclude treatment was cause of difference?
NO, you cannot be sure, because of potential confounding variables.
Explain ABA (reversal) design. Advantages?
Baseline, Treatment, withhold treatment, new baseline.
ex: Walker and Buckley. 9 year old. spending a lot of time not paying attention. used reinforcement. Treatment: points for toys. used ABA.
1) limits potential confounds.
Why would AB be ineffective?
1 )length of study: events change
history intervening as a confounding variable.
2) maturation of subject
3) instrumental decay (of researcher as well)
ex: boredom, fatigue, less diligent w time.
4) regression to the mean: child was selected because of high base rate, and it changes every day. tend to regress toward the middle level of behavior
5) testing issues: tend to do higher on next test only because of retaking test.
6) interactions of all of the above. composite confound.
Why would ABA design be ineffective?
1) ending on baseline is not acceptable from a therapist’s point of view.
2) some dependent variables may not revert to baseline. most obvious is carryover effects.
