Application of Regression Discontinuity to Investigate the Effectiveness of Lidcombe Program on Decreasing of Severity of Stuttering in Primary School Children: A Single-Subject Research

In single-subject research, there is autocorrelation between observations, that violates the assumption of observational independence. The discontinuity regression considers the correlation between observations over time and in different treatment stages (baseline, intervention, etc.). The aim of the current study was to compare the visual and discontinuous regression methods in evaluating the effectiveness of the Lidcombe program on decreasing of severity of stuttering in primary school children.

This study with single subject design conducted on four children (7-9 years) who stutter. These children participated in a Lidcombe program, was scheduled in 22 weeks, with a ABA model. Stuttering severity was measured with a rating scale. In order to analyze the data of the study, visual analysis, effect size and regression discontinuity were used.

Data analysis based on visual analysis as well as discontinuity regression indicated that most participants recorded a significant change, as a decline in severity of syllables stuttered, following the intervention. Regression coefficients on phase change from treatment to follow-up were significant, meaning that the Lidcombe program was effective over time.

The results of this study based on visual analysis as well as discontinuity regression indicated Lidcombe program eventuated in decreases of severity of stuttering in children participated in this study, and provide the preliminary evidence for the use of this treatment method for the children with stuttering.