PSYB01 - Chapter 11
Single Case , Quasi-Experimental and Developmental Research
Single Case Experimental Designs
Single Case experimental designs:
Effect of independent variable must inferred from experiment with only one participant .
Need to determine whether experimental manipulation had effect on single participant .
1) Subject’s behavior is measured over time during baseline control periods .
Baseline: Subjects behavior during control period before introduction of experimental
2) Manipulation is introduced during treatment period .
3) Behavior of subjects is continues to be observed .
Change in behavior from baseline to treatment period is evidence for manipulation
Have issue in how to determine that the manipulation of independent variable had an effect .
By reversing manipulation in reversal design .
Reversal design: Demonstrate reversibility of manipulation to show effect in how it’s effect in variable.
Known as ABA design, or withdrawal design .
Requires behavior be observed during Baseline control period and again during treatment period .
And also during second baseline period after experimental treatment has removed .
Design can be improved by extending to ABAB design .
Experimental treatment is introduced a second time .
Provide opportunity to observe second reversal when treatment is introduced again .
Control group also may be used in reversal design .
Multiple Baseline Design
Multiple Baseline Design: Use in irreversible treatment .
Effectiveness of treatment demonstrated when behavior changes only after manipulation is
To demonstrate the effectiveness of treatment:
Changes must be observed under multiple circumstances to rule out possibilities of other
Several variation of multiple baseline design:
1) Multiple baseline across subjects: Behavior of several subjects is measured over time .
For each subject, manipulation is introduced at different point in time .
Can rule out explanations based on chance/historical events .
2) Multiple baseline across behavior: Several different behavior of single subject are measured over time .
Same manipulation is applied to each of behaviors at different times .
e.g. reward system applied to different behavior at different times .
3) Multiple baseline across situation: Same behavior is measured in different setting .
At home/work, manipulation is introduced at different time in each setting .
Replications in Single Case Designs
Procedure for use with single subject can be replicated with other subjects .
Greatly enhance the generalizability of results . Single case research present the result from each subject individually, rather than group data/ overall
Useful for studying many research problems, not require complex statistical analyses .
Valuable for someone who applying some change technique in natural environment .
e.g. teacher trying a new technique in classroom .
Program Evaluation: Research on program that are proposed and implemented .
To achieve some positive effect on group of individuals .
e.g. Drug abuse resistance education (DARE ) program in school .
”Outcome evaluation” : The initial focus of evaluation research .
e.g. asking did program result in positive outcome for which it was designed .
Rossi identifies 5 types of evaluations: as five phase of evaluation process:
1) Needs Assessment: Evaluation of need:
Needs assessments asks whether there’s problem need to addressed in target population .
2) Program Theory Assessment: Question addresses program theory .
Program must be based on valid assumption about causes of problem and rationale .
Assessment of program theory: Involve collaboration of researchers, service provider etc.
To determine proposed program do address needs of target population in appropriate way .
3) Process evaluation: or program monitoring
Monitors to determine it’s reaching the target population when program is under way .
To determine whether it’s attracting enough clients etc.
To obtain assurance that program is doing what it’s supposed to do .
4) Outcome Evaluation / impact assessment:
To determine does intended outcome of program being realized .
Must devise a way of measuring outcome .
experimental design with random assignment will carry out to answer question about outcome .
5) Efficiency assessment:
Researcher must determine whether it’s “worth it” .
Cost of program must weighted against its benefits .
Quasi-experimental designs: Address the need to study the effect of independent variable in settings .
Which control features of true experimental designs cannot be achieved .