Chapter 8 Changing-Criterion Designs

Chapter 8 Changing-Criterion Designs

By:

Hallie Steinberg & Grete Christianson

Limitations/concerns

What we like/don't like about the design

Discussion

Rationale for the use of this design

Explanation

• no need to withhold intervention
• helps distinguish if intervention is working (if the criterion gradually gets met, performance relates to intervention)
• subphases help make & test predictions

• Allows for gradual changes, which is especially important in behavior changes such as quitting smoking, increasing exercise, etc.
• It's flexible... Researchers can decide how to set each criterion. Each step does not have to be equal!
• Limitations on showing causality, especially since behaviors rarely ever meet criteria perfectly

• If you have any, please share your experience with a changing-criterion design
• If not, share with us one idea how you could use this in your professional life AND one area it could help you with a change in your personal life!

• begins with baseline phase, looks at 1 behavior
• intervention begins, criterion is set
• criterion increases or decreases during each subphase
• behavior changes gradually over time during intervention phase - once it levels out, criterion is increased or decreased
• at each criterion subphase there could be a reinforcing consequence
• Criterion increases or decreases to desired level of performance

Limitations:

• can be ambiguous whether or not the data show a trend that follows closely enough to changes in criterion (behavior change in the desired direction that is gradual)
• Remember: a causal relationship be claimed only if behavior follows same pattern as criterion, so data should increase/decrease in a step-like fashion
• The more step-like the data is, the more likely it is a causal relationship
• Sometimes, a participant may make changes in performance that change more quickly than the criteria
• For example - the goal is for someone to increase from running 1 mi to 5 mi... the criteria increase by .25 miles but by the 4th day, the person can already run 4.5 miles... although this is an increase, as desired, we cannot say it was definitively due to the changing-criterion design

** When this is the case, mini-reversal phases may be the answer

Important Features of this Study

Variations of the design

Graph

Sources

Target behaviors it could address & benefit

Morgan, R. L., & Hiebert, R. (1982). The changing criterion design: Assessment of running behavior in a handicapped youth.Psychological Reports, 50, 1287 – 1293.

Kazdin, A. E. (2011). Single-case research designs: methods for clinical and applied settings (2nd ed.). New York: Oxford University Press.

This study was trying to decrease the time it took for an individual to run one lap so that they could participate in physical activities with peers

Criterion was a decrease in time to run a lap, once a participant ran 2 sessions below time the next subphase would start which the new criterion which was a decrease in time by 1 second

When a criterion was met, the individual got points to use to buy items and received praise

If a criterion was not met, the individual had to run another lap

Looking at the graph:

• baseline collected
• subphases
• performance-related behavior
• gradual decrease in mean running time, performance relates to intervention

Mini-reversal Phases:

• occur during the intervention phase
• return to a previous phase before setting next criterion
• rationale is same as ABAB design... to show that the behavior is under the control of the intervention, not extraneous factors

Distributed-changing-criterion design:

• use when there are multiple tasks that are interdependent and cannot all be improved at once
• for example: homework for math, reading, and science
• criterion (let's say amount of time spent studying) is spread amongst all three subjects

performance related behaviors

• decreasing compulsive behaviors like blurting out
• increasing minutes an individual reads or exercises

some benefits would be gradual increase or decrease of a behavior, not having to withdraw intervention, & focusing on one behavior