Publication bias in meta-analysis of single-case research designs based on hierarchical linear models.

Research Project

Project/Area Number	24730530
Research Category	Grant-in-Aid for Young Scientists (B)
Allocation Type	Multi-year Fund
Research Field	Educational psychology
Research Institution	Joetsu University of Education
Principal Investigator	OKUMURA Taichi 上越教育大学, 学校教育研究科(研究院), 准教授 (90547035)
Project Period (FY)	2012-04-01 – 2015-03-31
Project Status	Completed (Fiscal Year 2014)
Budget Amount *help	¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000) Fiscal Year 2014: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000) Fiscal Year 2013: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000) Fiscal Year 2012: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Keywords	階層的線形モデル / 単一事例実験 / メタ分析 / マルチレベル分析
Outline of Final Research Achievements	This study examined the factors and degrees of publication bias when single-case research design data are synthesized by the hierarchical linear models (HLM). Results can be summarized into the following three points: (1) Distributions of effect sizes of single-case designs that were reported in Japan were distorted, implying that the publication biases existed in these studies. (2) When subjects are screened by their pretest scores, treatment effects tend to be overestimated by the regression to the mean effect, and even non-effective treatments can be evaluated as largely effective. (3) When publications of the single-case studies depend on the statistical significance of the treatments and only published cases are used for meta-analyses by the HLM, originally non-effective treatments can be evaluated as if they were considerably effective ones.