Meta-Analysis of the Effect on Communication Intervention for Students with Severe and Multiple Disabilities : Based on Single-Subject Research Designs

Abstract

Background & Objectives: The purpose of this study is to examine the effect of communication intervention on students with severe and multiple disabilities using single-subject design research. Method: This study analyzed nine single-subject design studies that were published in Korea Citation Index-listed journals(including candidate journals). In order to calculate the effect sizes of the communication intervention, this study derived Tau U scores presented as one of the methods to calculate the effect sizes of single-subject design research. After measuring Tau U scores according to the variables of individual studies, this study used Kruskal-Wallis test and Mann-Whitney U test to compare and analyze the effect sizes according to moderating variables in dependent variables, demographic characteristics (the types of disability, genders, ages, and school levels), and research characteristics (research designs, intervention environments, intervention period, treatment fidelity). Results: According to the results of calculating Tau U as effect sizes in order to examine the effect of communication intervention on desirable behaviors and problem behaviors in students with severe and multiple disabilities, in a total of 9 analysis papers, Tau U as the effect size on desirable behaviors (the number of analyses = 156) to be expected that the effect size of dependent variables would be positive was 0.93, whereas Tau U as the effect size on problem behaviors (the number of analyses = 33) to be expected that the effect size of dependent variables would be negative was -0.87. Conclusion: Thus, it was confirmed that the effects of intervention on desirable behaviors and problem behaviors were very effective overall in the papers that were analyzed. There were mostly high effect sizes according to diverse intervention variables in the studies on communication intervention of students with severe and multiple disabilities. Also, according to the results of the analysis of thediverse moderating variables, there were significant differences in the types of disabilities reported. Lastly, this study discussed the direction for future studies on communication intervention on students with severe and multiple disabilities.

Effects of single‐case reading interventions for students with and at‐risk of emotional and behavioral disorders in grades K–12: A quantitative synthesis

“…We then conducted moderator analyses using the weighted average of Tau‐U for each coded category (Bae & Park, ; Ninci et al, ). We utilized a rank‐based nonparametric test, Kruskal–Wallis H test (Kruskal &Wallis, ), in SPSS for determining the presence of statistically significant differences between two or more categorical variables on Tau‐U using an α level of 0.05 (5%); p values are approximated from χ 2 distributions.…”