Meta-Analysis for Single-Subject Research: When Does It Clarify, When Does It Obscure?

Document Type

Article

Journal/Book Title/Conference

Remedial and Special Education

Volume

8

Issue

2

Publication Date

1987

First Page

43

Last Page

48

Abstract

Comments on the percentage of nonoverlapping data (PND) approach proposed by T. E. Scruggs et al (see record 1987-29788-001) for synthesizing single-S research. The present authors take issue with the PND method, which reduces functional analyses to a 1-number summary, for the following reasons: (a) the most relevant and credible evidence from single-S designs lies in the patterns of change across time; (b) summaries based on PND may miss vital idiosyncracies in behavior within and across studies; (c) syntheses of literature based on PND may misrepresent procedural facts and outcomes; and (d) syntheses based on PND attempt, inappropriately, to draw conclusions about the relative merits of broad categories of intervention.

This document is currently not available here.

Share

COinS