Clinical and Statistical Applications of Contingency Space Analysis
|Monday, May 26, 2014|
|11:00 AM–11:50 AM |
|W183a (McCormick Place Convention Center)|
|Area: PRA; Domain: Applied Research|
|Instruction Level: Intermediate|
|CE Instructor: Brian K. Martens, Ph.D.|
|Chair: Jennifer R. Zarcone (Kennedy Krieger Institute)|
|BRIAN K. MARTENS (Syracuse University)|
|Brian K. Martens, Ph.D., is a professor of psychology at Syracuse University. He received an M.S. degree in combined school/experimental psychology from Colorado State University (behavior analysis focus) and a Ph.D. in school psychology from the University of Nebraska-Lincoln. Dr. Martens served as director of training for the Syracuse University Psychology Program from 1998-2007 and as associate chair and chair of the Psychology Department from 2007-2009. He was editor-in-chief of the Journal of Behavioral Education from 2009-2012 and is a past associate editor for the Journal of Applied Behavior Analysis. Dr. Martens is a fellow in Division 16 of American Psychological Association, a member of the Society for the Study of School Psychology, and previously served on the board of directors of the Society for the Experimental Analysis of Behavior. He has published more than 110 articles, books, and chapters concerned with translating findings from basic operant research into effective school-based interventions, functional assessment and treatment of children's classroom behavior problems, and the instructional hierarchy as a sequenced approach to skill training.|
Sequential recording of behavior and its consequences is a common strategy for identifying potential maintaining variables in the natural environment. Disagreement remains over a standard approach to detecting contingent relations in the resulting data as well as a suitable association metric. One approach reported in the literature involves comparing the conditional probability of a consequence given the occurrence of problem behavior to its conditional probability given the absence of problem behavior. This approach, known as contingency space analysis (CSA) can be used to identify the direction and magnitude of potential reinforcement effects from descriptive assessment data. Moreover, joint occurrences of behavior and its consequences can be summarized in a 2 by 2 contingency table for which an operant contingency value (OCV) can be computed. In this presentation, procedures for conducting and interpreting a CSA are described, and data are presented showing various applications of CSA to clinical decision making. The presentation concludes by comparing the OCV to other measures of association using simulated and empirical data. These analyses suggest that CSA as a general analytical approach and the OCV as an index of contingency are useful tools for helping behavior analysts identify contingent relations during a functional behavior assessment.
|Target Audience: |
ABA practitioners and applied researchers.
|Learning Objectives: At the conclusion of this event, participants will be able to; (a) conduct observations of problem behavior and its consequences using modified partial-interval recording, (b) graph and interpret behavior-consequence data in a contingency space analysis (CSA), (c) describe the relationship between CSA, functional analysis, and treatment outcome data, and (d) describe why the operant contingency value (OCV) is a more robust measure than either the phi coefficient or Yule's Q as a measure of association for 2 by 2 contingency tables.|