hlmLab
Provides functions for visualization and decomposition in hierarchical linear models (HLM) for applications in education, psychology, and the social sciences. Includes variance decomposition for two-level and three-level data structures following Snijders and Bosker (2012, ISBN:9781849202015), intraclass correlation (ICC) estimation and design effect computation as described in Shrout and Fleiss (1979) <doi:10.1037/0033-2909.86.2.420>, and contextual effect decomposition via the Mundlak (1978) <doi:10.2307/1913646> specification distinguishing within- and between-cluster components, with the uncertainty of the contextual contrast obtained from the full fixed-effect covariance matrix. Teaching displays cover simulated intraclass correlations, partial pooling of cluster means, random-slope heterogeneity, and cross-level interaction with an observed Level-2 moderator, following Hofmann and Gavin (1998) <doi:10.1177/014920639802400504> and Hamaker and Muthen (2020) <doi:10.1037/met0000239>. Multilevel models are estimated using 'lme4' (Bates et al., 2015 <doi:10.18637/jss.v067.i01>). An optional 'Shiny' application enables interactive exploration of model components and parameter variation. The implementation follows the multilevel modeling framework of Raudenbush and Bryk (2002, ISBN:9780761919049).
- Centering Decisions in Hierarchical Linear Models: Implications for Research in Organizations
- Fitting Linear Mixed-Effects Models Usinglme4
- Intraclass correlations: Uses in assessing rater reliability.
- On the Pooling of Time Series and Cross Section Data
- The fixed versus random effects debate and how it relates to centering in multilevel modeling.
This page was built for software: hlmLab
