Chapter 11
Single-Case Research Designs
Learning Objectives
Describe the different types of single-case designs.
Explain the strategies used in the single-case designs to rule out the influence of rival
hypotheses.
Identify the situations in which each of the single-case designs would be appropriate.
Describe the methodological issues that must be considered in using the single-case
designs.
Describe the criteria used for evaluating treatment effects with single-case designs.
Chapter Outline
Introduction
• Single-case designs
– use only one participant or one group of participants
– no random assignment and no control group
– single participant used most frequently
• History of single case designs
– not case studies
– research in psychology began with single case experiments
Pavlov, Ebbinghaus, Skinner
Fisher’s introduction of ANOVA
– single-case designs became more acceptable with the growth in research in behavior
therapy
Single-Case Designs
• Time series designs
– with multiple data points before and after treatment is introduced
– does not eliminate the history threat
– assessment of a treatment effect is based on the assumption that the pattern of pretreatment
responses would continue in the absence of the treatment
• Simplest type of single-case design is ABA
ABA Design
• Baseline (A)
– the target behavior of the participant in its naturally occurring state or prior to presentation
of the treatment condition
• Treatment (B)
– recordings of behavior after the treatment has been introduced
• ABA
– design in which the response to the treatment condition is compared to baseline responses
recorded before and after treatment
– baseline – treatment – baseline
– demonstration of treatment effectiveness requires return to baseline
,• Reversal
– change of behavior back to baseline level after withdrawal of treatment
ABA Example
• Walker and Buckley (1968)
– 9 year old exhibiting disruptive classroom behavior
– baseline
% time child spent on academic assignments
baseline recorded until DV stabilized
– treatment
points earned if no distraction occurred during a given time interval
points could be exchanged for model of his choice
– return to baseline
when child had completed three successive ten-minute distraction-free sessions, the
reinforcement of being able to earn points was withdrawn
ABA Designs
• Problems with ABA design
– ending on baseline not acceptable from therapist point of view because you are ending
with a denial of treatment
– solution – ABAB design may be used
– some DVs may not revert to baseline when treatment is withdrawn due to carryover
– solution – multiple-baseline design
– withdrawal vs. reversal design
reversal design – design in which the treatment condition is applied to an alternative
but incompatible behavior so that a reversal in behavior is produced
ABAB Design
• Disruptive classroom behavior example
– return to treatment condition after second baseline condition
– should see a return of DV to treatment levels
Interaction Design
• Tests the combined effects of two treatments
– e.g., concrete and verbal reinforcement
• Similar to factorial design in that you look at each treatment (IV) individually (main effects)
and combined (interaction)
• Must change only one treatment at a time
• Must use both sequences to test the combined influence over the effect of just one variable
• Disadvantages
– two participants may be required
– interaction effect can be demonstrated only if each variable does not cause a maximum
increment in performance
Multiple Baseline Design
• Design in which the treatment condition is successively administered to several target
participants, target outcomes (DVs), or target settings
– alternative to ABA or ABAB when history threat may be suspected
– no withdrawal or reversal involved
,• Design
– baseline data collected on several participants, DVs, or settings
– treatment successively administered to each target (i.e., staggered)
– Treatment effect demonstrated by a change in behavior only when treatment is given
• Requires independence of behaviors to demonstrate an effect
Multiple Baseline Example
• Van Houten, Van Houten, & Malenfant, 2007
– tested the effectiveness of a program designed to increase helmet use by middle school
students when riding their bicycles
– three schools were targeted, and baseline helmet use data were gathered at each school
– the treatment program was introduced at one school at a time
– increases in correct helmet use occurred when the helmet program was introduced in each
school
– when the campaign was introduced at the second and third schools, helmet use increased
but did not change at the schools still at baseline
– this fingerprint or pattern of change provided evidence of the causal efficacy of the helmet
advocacy program on helmet use by students
Changing-Criterion Design
• Design in which a participant’s behavior is gradually shaped by changing the criterion for
success during successive treatment periods
• Design
– baseline data taken on a single behavior
– treatment introduced with a criterion level of performance that needs to be met
– if criterion met, then second criterion level set
– target behavior increased with multiple criterion levels (at least two)
• Factors to consider in using this design
– length of treatment
long enough for the behavior to stabilize
– size of criterion change
large enough to notice a change
– number of treatment phases
at least two, but enough to demonstrate a treatment effect
• Himadi, Osteen, Kaiser, and Daniel (1991)
– study to reduce the delusional verbalizations of a 51-year-old white male with
schizophrenia, chronic undifferentiated type
– the investigators first obtained baseline data on the number of delusional answers given
– the treatment session consisted of asking the patient a question that had reliably elicited a
delusional answer and instructing the patient to respond to the question “so that other people
would agree with your answers.” After the patient provided the appropriate answer, he was
given a reinforcer consisting of a cup of coffee
Phase 1 criterion, the patient had to provide nondelusional responses to two questions
Phase 2, nondelusional responses to four delusion-eliciting questions
Methodological Considerations
• Baseline
– must be stable before treatment implemented
– absence of trend or in the direction opposite of what is expected from the treatment
, – little variability
if variability in data, then track until stable or try to identify source of the instability
– must also consider reactivity when tracking baseline data
• Change only one variable at a time (cardinal rule)
• Length of phases
– no set rule – semblance of stability
– possibility of extraneous variables creeping in with long phases
– carry-over effect may require short phases
– cyclic variations maybe need to incorporate the cycle in all phases
Criteria for Evaluating Change
• Experimental criterion
– repeated demonstration of behavioral change should occur with treatment introduction
– nonoverlap of treatment and baseline phases
• Therapeutic criterion
– clinical significance of a therapeutic, or other psychological intervention for an individual
or group of clients
– researchers often use social validation – does it produce a change in the client’s daily
functioning
social comparison – compare behavior with nondeviant peers
subject evaluation – do others who interact with the client see a change
Multiple choice questions
1. Single-case designs, by definition, do not incorporate control groups. What is the standard for
comparison purposes to evaluate the treatment effects?
a. there is no comparison standard when using a single-case design
b. the researcher “gut” feeling
c. observations of a single control participant
* d. the pre-treatment observations
2. Single-case research is most closely associated, historically, with what area of psychology?
a. social psychology
b. cognitive science
c. personality psychology
* b. behavioral psychology
3. Single-case designs are closely related to which type of quasi-experimental design?
a. the non-equivalent comparison group design
b. the cross-over design
* c. the time series design
d. the multivariate design
Single-Case Research Designs
Learning Objectives
Describe the different types of single-case designs.
Explain the strategies used in the single-case designs to rule out the influence of rival
hypotheses.
Identify the situations in which each of the single-case designs would be appropriate.
Describe the methodological issues that must be considered in using the single-case
designs.
Describe the criteria used for evaluating treatment effects with single-case designs.
Chapter Outline
Introduction
• Single-case designs
– use only one participant or one group of participants
– no random assignment and no control group
– single participant used most frequently
• History of single case designs
– not case studies
– research in psychology began with single case experiments
Pavlov, Ebbinghaus, Skinner
Fisher’s introduction of ANOVA
– single-case designs became more acceptable with the growth in research in behavior
therapy
Single-Case Designs
• Time series designs
– with multiple data points before and after treatment is introduced
– does not eliminate the history threat
– assessment of a treatment effect is based on the assumption that the pattern of pretreatment
responses would continue in the absence of the treatment
• Simplest type of single-case design is ABA
ABA Design
• Baseline (A)
– the target behavior of the participant in its naturally occurring state or prior to presentation
of the treatment condition
• Treatment (B)
– recordings of behavior after the treatment has been introduced
• ABA
– design in which the response to the treatment condition is compared to baseline responses
recorded before and after treatment
– baseline – treatment – baseline
– demonstration of treatment effectiveness requires return to baseline
,• Reversal
– change of behavior back to baseline level after withdrawal of treatment
ABA Example
• Walker and Buckley (1968)
– 9 year old exhibiting disruptive classroom behavior
– baseline
% time child spent on academic assignments
baseline recorded until DV stabilized
– treatment
points earned if no distraction occurred during a given time interval
points could be exchanged for model of his choice
– return to baseline
when child had completed three successive ten-minute distraction-free sessions, the
reinforcement of being able to earn points was withdrawn
ABA Designs
• Problems with ABA design
– ending on baseline not acceptable from therapist point of view because you are ending
with a denial of treatment
– solution – ABAB design may be used
– some DVs may not revert to baseline when treatment is withdrawn due to carryover
– solution – multiple-baseline design
– withdrawal vs. reversal design
reversal design – design in which the treatment condition is applied to an alternative
but incompatible behavior so that a reversal in behavior is produced
ABAB Design
• Disruptive classroom behavior example
– return to treatment condition after second baseline condition
– should see a return of DV to treatment levels
Interaction Design
• Tests the combined effects of two treatments
– e.g., concrete and verbal reinforcement
• Similar to factorial design in that you look at each treatment (IV) individually (main effects)
and combined (interaction)
• Must change only one treatment at a time
• Must use both sequences to test the combined influence over the effect of just one variable
• Disadvantages
– two participants may be required
– interaction effect can be demonstrated only if each variable does not cause a maximum
increment in performance
Multiple Baseline Design
• Design in which the treatment condition is successively administered to several target
participants, target outcomes (DVs), or target settings
– alternative to ABA or ABAB when history threat may be suspected
– no withdrawal or reversal involved
,• Design
– baseline data collected on several participants, DVs, or settings
– treatment successively administered to each target (i.e., staggered)
– Treatment effect demonstrated by a change in behavior only when treatment is given
• Requires independence of behaviors to demonstrate an effect
Multiple Baseline Example
• Van Houten, Van Houten, & Malenfant, 2007
– tested the effectiveness of a program designed to increase helmet use by middle school
students when riding their bicycles
– three schools were targeted, and baseline helmet use data were gathered at each school
– the treatment program was introduced at one school at a time
– increases in correct helmet use occurred when the helmet program was introduced in each
school
– when the campaign was introduced at the second and third schools, helmet use increased
but did not change at the schools still at baseline
– this fingerprint or pattern of change provided evidence of the causal efficacy of the helmet
advocacy program on helmet use by students
Changing-Criterion Design
• Design in which a participant’s behavior is gradually shaped by changing the criterion for
success during successive treatment periods
• Design
– baseline data taken on a single behavior
– treatment introduced with a criterion level of performance that needs to be met
– if criterion met, then second criterion level set
– target behavior increased with multiple criterion levels (at least two)
• Factors to consider in using this design
– length of treatment
long enough for the behavior to stabilize
– size of criterion change
large enough to notice a change
– number of treatment phases
at least two, but enough to demonstrate a treatment effect
• Himadi, Osteen, Kaiser, and Daniel (1991)
– study to reduce the delusional verbalizations of a 51-year-old white male with
schizophrenia, chronic undifferentiated type
– the investigators first obtained baseline data on the number of delusional answers given
– the treatment session consisted of asking the patient a question that had reliably elicited a
delusional answer and instructing the patient to respond to the question “so that other people
would agree with your answers.” After the patient provided the appropriate answer, he was
given a reinforcer consisting of a cup of coffee
Phase 1 criterion, the patient had to provide nondelusional responses to two questions
Phase 2, nondelusional responses to four delusion-eliciting questions
Methodological Considerations
• Baseline
– must be stable before treatment implemented
– absence of trend or in the direction opposite of what is expected from the treatment
, – little variability
if variability in data, then track until stable or try to identify source of the instability
– must also consider reactivity when tracking baseline data
• Change only one variable at a time (cardinal rule)
• Length of phases
– no set rule – semblance of stability
– possibility of extraneous variables creeping in with long phases
– carry-over effect may require short phases
– cyclic variations maybe need to incorporate the cycle in all phases
Criteria for Evaluating Change
• Experimental criterion
– repeated demonstration of behavioral change should occur with treatment introduction
– nonoverlap of treatment and baseline phases
• Therapeutic criterion
– clinical significance of a therapeutic, or other psychological intervention for an individual
or group of clients
– researchers often use social validation – does it produce a change in the client’s daily
functioning
social comparison – compare behavior with nondeviant peers
subject evaluation – do others who interact with the client see a change
Multiple choice questions
1. Single-case designs, by definition, do not incorporate control groups. What is the standard for
comparison purposes to evaluate the treatment effects?
a. there is no comparison standard when using a single-case design
b. the researcher “gut” feeling
c. observations of a single control participant
* d. the pre-treatment observations
2. Single-case research is most closely associated, historically, with what area of psychology?
a. social psychology
b. cognitive science
c. personality psychology
* b. behavioral psychology
3. Single-case designs are closely related to which type of quasi-experimental design?
a. the non-equivalent comparison group design
b. the cross-over design
* c. the time series design
d. the multivariate design
