Outcome-Reporting Bias in Special Education Intervention Research Using Experimental and Quasi-Experimental Group Designs: A Conceptual Replication.

This article focuses on a conceptual replication of a study examining outcome-reporting bias in special education intervention research by comparing statistically significant and nonsignificant outcomes reported in unpublished dissertations with those in corresponding published journal articles. Analyzing 40 special education dissertations with matched publications from 2006 to 2015, the study found that statistically significant outcomes were 1.48 times more likely to be published than nonsignificant ones, indicating the presence of outcome-reporting bias, though smaller than previously reported. Significant moderators of this bias included academic intervention outcomes, randomized controlled trial designs, participant race (samples with ≥50% non-White), and high-incidence disability status. The authors highlight the under-publication of dissertations and recommend their inclusion in systematic reviews and meta-analyses to reduce publication bias and better inform special education research and practice.