Main content
Exploration of suboptimal modeling choices - Ordinal modeling as a way to better measure effect size heterogeneity?
Date created: | Last Updated:
: DOI | ARK
Creating DOI. Please wait...
Category: Project
Description: The project investigates the influence of model specification on the heterogeneity of effect sizes. We compared heterogeneity estimates (I2, H2 and CV) between linear regression and ordinal regression models. We fitted the two models on data from 16 multilab direct replication projects which used a two-group treatment / control design and meta-analyzed the results. The project contains the data, the analysis code, the preprint of the publication as well as supplementary material.
Start managing your projects on the OSF today.
Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.