Differential Reinforcement of Other Behavior (DRO)
Study: Kennedy & Haring (1993)

Describe how students were selected to participate in the study:

Participants were identified by their special education teachers as students who engaged in problem behavior at a level that interfered with instruction and required intervention. Teachers also completed the Motivation Assessment Scale (MAS; Durand & Crimmins, 1988) for each student and problem behavior. Results were used to identify topographies of problem behavior potentially related to instructional demands (i.e., scores for escape motivation were at least 30% higher than other subscales).

Describe how students were identified as being at risk for academic failure (AI) or as having emotional/behavioral difficulties (BI):

Teachers identified whether students engaged in problem behavior that interfered with instruction, and reported students’ histories of engaging in problem behavior. All participants had severe intellectual disabilities.

ACADEMIC INTERVENTION: What percentage of participants were at risk, as measured by one or more of the following criteria:

• below the 30th percentile on local or national norm, or
• identified disability related to the focus of the intervention?

%

BEHAVIORAL INTERVENTION: What percentage of participants were at risk, as measured by one or more of the following criteria:

• emotional disability label,
• placed in an alternative school/classroom,
• non-responsive to Tiers 1 and 2, or
• designation of severe problem behaviors on a validated scale or through observation?

%

Please describe the study design:

A three-part multielement design was used to evaluate the effects of different DRO procedures across tasks for each participant. An initial multielement design was utilized to demonstrate that problem behavior occurred at comparable levels across tasks during a baseline phase. A second multielement design was used to compare the effects of DRO tasks across three different reinforcers/rewards (i.e., escape, reward, and combined). A final multielement phase was used to demonstrate the efficacy across tasks of the most effective procedure from the comparison phase. A Latin square counterbalancing of tasks and DRO conditions across students was employed so that no single DRO condition was associated with a specific task.

Clarify and provide a detailed description of the treatment in the submitted program/intervention:

Three variations on DRO were evaluated across three tasks including shelving books, straightening chairs, and sorting dishes. Escape DRO: Duration of tasks varied as a function of the omission of problem behavior in escape DRO. Tokens were earned for omitting problem behavior for the duration of an instructional trial (i.e., 15 s). Once 10 tokens were earned, the instructional task was terminated. Sessions lasted a maximum of 10 min (i.e., sessions were terminated when either 10 tokens were earned or 10 min had elapsed). During breaks following token exchange, students sat at a table for 10 min before beginning the next session. Verbal praise was delivered for completion of demands throughout sessions. Reward DRO: Tokens were earned for omitting problem behavior for the duration of an instructional trial (i.e., 15 s). A reward was earned if a student accumulated 10 tokens by the end of the task. Rewards were selected individually for each student by teacher nominations of preferred, age-appropriate activities. All sessions lasted 10 min, regardless of whether 10 tokens were earned. When 10 tokens were not earned, students were asked to sit at a table for 10 min until the next session began. Verbal praise was delivered for completion of demands throughout sessions. Combined DRO: The contingencies in both the escape and reward DRO conditions were combined. Students earned tokens for the omission of problem behavior for the length of an instructional trial (15 s). Once 10 tokens were accumulated, the session was terminated and the student-specific reward from the reward DRO condition was delivered. Sessions were terminated if 10 min elapsed before 10 tokens were earned.

Clarify what procedures occurred during the control/baseline condition (third, competing conditions are not considered; if you have a third, competing condition [e.g., multi-element single subject design with a third comparison condition], in addition to your control condition, identify what the competing condition is [data from this competing condition will not be used]):

Baseline: Each session lasted for 10 min and a 10-min interval occurred between sessions. During instruction, students received praise when they completed a trial. Problem behavior was ignored. Following completion of a task, students were asked to take a “break” and sit in a chair at a table.

Please describe how replication of treatment effect was demonstrated (e.g., reversal or withdrawal of intervention, across participants, across settings)

Effect was demonstrated via rapid alternation of conditions, with task randomly assigned to conditions (multielement design). Design included a baseline comparison across tasks as well as a comparison of the effects of the most effective treatment on problem behavior across tasks to control for task as a rival explanation for differences in level of problem behavior.

Please indicate whether (and how) the design contains at least three demonstrations of experimental control (e.g., ABAB design, multiple baseline across three or more participants).

The design contains at least three demonstrations of experimental control. In the multielement condition comparing DRO procedures to one another, 7-10 comparisons between the combined DRO conditions (most effective for all participants) and the second-most effective alternative condition were present.

If the study is a multiple baseline, is it concurrent or non-concurrent?

N/A

How was the program delivered?

Individually
Small Group
Classroom

Study measures are classified as targeted, broader, or administrative data according to the following definitions:

• Targeted measures
  Assess outcomes, such as competencies or skills, that the program was directly targeted to improve.
  • In the academic domain, targeted measures typically are not the very items taught but rather novel items structured similarly to the content addressed in the program. For example, if a program taught word-attack skills, a targeted measure would be decoding of pseudo words. If a program taught comprehension of cause-effect passages, a targeted measure would be answering questions about cause-effect passages structured similarly to those used during intervention, but not including the very passages used for intervention.
  • In the behavioral domain, targeted measures evaluate aspects of external or internal behavior the program was directly targeted to improve and are operationally defined.
• Broader measures
  Assess outcomes that are related to the competencies or skills targeted by the program but not directly taught in the program.
  • In the academic domain, if a program taught word-level reading skill, a broader measure would be answering questions about passages the student reads. If a program taught calculation skill, a broader measure would be solving word problems that require the same kinds of calculation skill taught in the program.
  • In the behavioral domain, if a program taught a specific skill like on-task behavior in one classroom, a broader measure would be on-task behavior in another setting.
• Administrative data measures apply only to behavioral intervention tools and are measures such as office discipline referrals (ODRs) and graduation rates, which do not have psychometric properties as do other, more traditional targeted or broader measures.

If you have excluded a variable or data that are reported in the study being submitted, explain the rationale for exclusion: