Meta-Analytic Pooling of Intraclass Correlation Coefficient Estimates (Poster 35)

• In Event: Thursday Poster Session 10:50 am
  In Poster Session: Research Syntheses and Meta-Analyses in Education: Exemplar Studies and Methodological Guidance

Abstract

Intraclass correlation coefficient (ICC) estimates are necessary for several statistical techniques. Researchers need accurate ICC estimates when conducting prospective power analyses for clustered data scenarios. In addition, meta-analysts require reasonable ICC values when adjusting effect sizes to account for clustered data. The accuracy of both analyses hinge on the value of the ICC estimate that accurately reflects the primary study.

The study evaluates how well meta-analytically pooled ICC estimates recover the population ICC parameter value when using different ICC variance formulae for the inverse variance weights. We found that the variance formula that uses a normalizing transformation (Fisher, 1970) performs well across most conditions. Additionally, we caution against using the formula derived by Smith (1957) to estimate the population ICC.

Authors

• Bethany Hamilton Bhat, University of Texas at Austin

• Tasha Beretvas, University of Texas at Austin