An Empirical Bayes Cluster-Mean Approach to Correct for Sampling Error in Between-Cluster Effects (Poster 18)

• In Event: AERA Poster Session Eleven
  In Poster Session: Propensity Score, Bayesian Techniques, and Applications in Multilevel Modeling

Abstract

With clustered data, it is often of interest to estimate and compare associations among variables separately for each level. While researchers routinely estimate between-cluster effects using the sample cluster means of a predictor, previous research has shown that such practice leads to biased estimates of coefficients at the between level, and recent research has recommended the use of latent cluster means in the multilevel structural equation modeling framework. However, the latent cluster mean approach requires large sample and specialized software. In this paper, we show how using empirical Bayes estimates of the cluster means (EBM) can also lead to consistent estimates of between-level coefficients. We show in a simulation that EBM outperforms the latent cluster-mean approach in relatively small samples.

Authors

• Hok Chio Mark Lai, University of Southern California
  Presenting Author

• Yichi Zhang, University of Southern California
  Presenting Author

• Feng Ji, University of California - Berkeley
  Non-Presenting Author