Doing Research in Behavior Modification

1
Doing Research in Behavior Modification

• Chapter 23

2
Introduction

• Phases of Behavior Modification Program
• Screening phase
• Clarifying the problem and determining who should
  treat it
• Baseline phase
• Determining the initial level of the behavior
• Treatment phase
• Intervention strategy is initiated
• Follow-up phase
• Evaluating the persistence of desirable
  behavioral changes after the termination of the
  program
• A behavior modification research project attempts
  to demonstrate convincingly that it is the
  treatment, rather than some uncontrolled
  variables, that was responsible for the change in
  the behavior in question.

3
Reversal-Replication (ABAB) Research Designs

• Baseline (A) is followed by treatment (B), return
  to baseline (A) condition, and then treatment
  again (B)
• Allows for replication of treatment effect
• Replication makes it clearer that treatment
  caused change in behavior

4
Reversal-Replication (ABAB) Research Designs
5
Reversal-Replication (ABAB) Research Designs
6
Reversal-Replication (ABAB) Research Designs

• Considerations
• How long should the baseline phase last?
• Until see stable pattern or trend opposite that
  is expected from treatment
• Shorter baselines in studies of behavior
  previously researched
• Availability of time may shorten baseline time
• Ethical considerations
• Some behaviors are dangerous, and it may be
  unethical to leave them untreated for long
  periods of time
• How many reversals and replications are
  necessary?
• Less replications if large effects are observed
  and a lot of previous research exists in the area
• Limitations
• Withdrawal of treatment may not lead to return to
  baseline
• Withdrawal may be undesirable or unethical

7
Multiple Baseline Designs

• Conduct more than one AB design concurrently with
  treatments beginning at different times
• Useful when reversals cannot be introduced

8
Multiple Baseline Designs

• Across behaviors
• Baselining several similar behaviors within an
  individual
• Across subjects
• Applying the same treatment to the same behavior
  problems of two or more individuals
• Across situations
• Baselining one type of behavior for a single
  individual in more than one setting

9
Multiple Baseline Designs
10
Changing-Criterion Designs

• Change over time the criterion for success and
  look for relationship between criteria changes
  and behavior change
• Can increase or decrease
• Frequency requirements
• Rate requirements
• Duration requirements
• Etc.

11
Changing-Criterion Designs
12
Alternating-Treatment Designs

• Compare effects of two or more treatment
  conditions considerably more rapidly than in ABAB
  design
• Applied at alternating times within the same time
  period
• Also known as multielement design
• Does not require reversal
• Several treatments can be evaluated at the same
  time
• Disadvantage treatment effects interaction

13
Data Analysis and Interpretation

• Data typically analyzed without control groups
  and statistical techniques used in other areas of
  psychology
• Behavior modifiers interested in understanding
  and improving the behavior of individuals, not
  groups

14
Data Analysis and Interpretation

• Evaluate treatments on two basic criteria
• Scientific
• Guidelines used by a researcher to evaluate
  whether or not there has been a convincing
  demonstration that the treatment was responsible
  for the changes
• Judgment made by visually inspecting the graph of
  the results.
• Guidelines for inspecting data - There is greater
  confidence that a treatment effect has been
  observed
• the greater the number of times that results are
  replicated
• the fewer the overlapping points between baseline
  and treatment phases
• the sooner the effect is observed following the
  introduction of treatment
• the larger the effect is in comparison to
  baseline
• the more precisely the treatment procedures are
  specified
• the more reliable the response measures
• the more consistent the finding are with the
  existing data and accepted behavioral theory
• Practical
• Changes to client, other significant individuals
  in clients life, and society in general

15
Data Analysis and Interpretation

• Social Validity
• Behavior modifiers need to socially validate
  their work on at least three levels (Wolf, 1978)
• Must examine the extent to which target behaviors
  indentified for treatment programs are really the
  most important for client and society
• Must be concerned with the acceptability to the
  client of the particular procedures used
• Must ensure that the consumers are satisfied with
  the results