Fragile Findings, Complex Conclusions, and Meta-Analysis in Special Education
- 1 November 1983
- journal article
- research article
- Published by SAGE Publications in Exceptional Education Quarterly
- Vol. 4 (3) , 97-106
- https://doi.org/10.1177/074193258300400312
Abstract
An introduction to meta-analysis as a method of research synthesis is presented. Research in special education is marked by many sources of variation that lead to fragile and complex findings. Although such outcome variation is valuable, controversy surrounds the best method for providing a comprehensive summary. Traditional review methods found in narrative and box-score techniques possess both theoretical and pragmatic difficulties that limit their usefulness. Quantitative methods like meta-analysis are suggested to achieve the scope and form necessary for accumulating evidence. The methods of meta-analysis are discussed, including problem formulation, study location, classifying and coding studies, data analysis, effective size interpretation, data aggregation, and conclusions. The advantages and problems of the meta-analysis approach are then discussed.Keywords
This publication has 25 references indexed in Scilit:
- Distribution Theory for Glass's Estimator of Effect size and Related EstimatorsJournal of Educational Statistics, 1981
- The Efficacy of Special Versus Regular Class Placement for Exceptional Children: a Meta-AnalysisThe Journal of Special Education, 1980
- Vote-counting methods in research synthesis.Psychological Bulletin, 1980
- Methods for Integrative ReviewsReview of Educational Research, 1980
- Statistical versus traditional procedures for summarizing research findings.Psychological Bulletin, 1980
- The Case Against Statistical Significance TestingHarvard Educational Review, 1978
- Primary, Secondary, and Meta-Analysis of ResearchEducational Researcher, 1976
- The test of significance in psychological research.Psychological Bulletin, 1966
- The Difference between Statistical Hypotheses and Scientific HypothesesPsychological Reports, 1962
- Models for testing the significance of combined results.Psychological Bulletin, 1953