International Meeting for Autism Research: Systematic Review and Multi-Metric Meta-Analysis of Social Stories™ Research

Systematic Review and Multi-Metric Meta-Analysis of Social Stories™ Research

Friday, May 21, 2010
Franklin Hall B Level 4 (Philadelphia Marriott Downtown)
3:00 PM
C. R. Peterson , School of Education, University of Wisconsin–Stout, Menomonie, WI
D. B. McAdam , Division of Neurodevelopmental and Behavioral Pediatrics, University of Rochester School of Medicine, Rochester, NY
D. A. Napolitano , Division of Neurodevelopmental and Behavioral Pediatrics, University of Rochester School of Medicine, Rochester, NY
J. Breidbord , Autism Research Centre, University of Cambridge, Cambridge, United Kingdom
Background: Used to provide information about social situations to people with a clinical autism diagnosis, Social Stories™ may be a convenient strategy for improving specific functional skills and reducing specific problem behaviors. As story-based materials have become common in classroom treatment programs, intervention procedures have been subject to continual revision leading to a large corpus of highly varied research. Meta-analyses give some evidence of questionable intervention efficacy but provide limited information about the quality of intervention studies and the suitability of research reports for determination of Social Stories™ empirical support.

Objectives: This review of current Social Stories™ research aims to synthesize results from studies with any experimental design and to summarize characteristics of research reporting.

Methods: Reports of Social Stories™ interventions published between 1993 and 2009 were identified via comprehensive pearl growing. Characteristics of intervention procedures (e.g., assessment of comprehension), experimental methods (e.g., study design), and research reporting (e.g., description of participants) were coded. Intervention efficacy was summarized using common graphical-data overlap metrics (i.e., PND and PEM) and those that were developed recently for comparison with standard effect sizes (i.e., PAND and NAP).

Results: Empirical investigations included 6 controlled trials and 39 studies that used a single-case experimental design. Among all single-case evaluations, 13 studies reported detailed characteristics of each participant (e.g., standardized assessment information) and 20 studies included three or more demonstrations of experimental effect (e.g., across participants, across behaviors). Although 29 investigations used other techniques (e.g., differential reinforcement) as part of a Social Stories™ strategy, recent reports describe specific evaluation of Social Stories™ or evaluation of specific intervention components. For each metric, relatively high efficacy was found for interventions used to reduce problem behaviors and those used in combination with other treatment techniques; however, overall efficacy differed significantly among several metrics (41.2 PAND, 47.2 NAP, 65.6 PND, 72.5 PEM). The PND results are consistent with prior findings of questionable intervention efficacy.

Conclusions: Methods-oriented review of Social Stories™ research shows increasing interest in specific intervention studies. The divergence of efficacy metrics supports ongoing efforts to improve methods of single-case meta-analysis. Other targets for scientific research include the format and text of intervention materials.

See more of: Treatment
See more of: Mandell
See more of: Clinical & Genetic Studies