Random Coefficient Models for Multilevel Analysis
- 1 March 1986
- journal article
- Published by American Educational Research Association (AERA) in Journal of Educational Statistics
- Vol. 11 (1) , 57-85
- https://doi.org/10.3102/10769986011001057
Abstract
We propose a possible statistical model for both contextual analysis and slopes as outcomes analysis. These techniques have been used in multilevel analysis for quite some time, but a precise specification of the regression models has not been given before. We formalize them by proposing a random coefficient regression model, and we investigate its statistical properties in some detail. Various estimation methods are reviewed and applied to Dutch school-career example.Keywords
This publication has 33 references indexed in Scilit:
- Random Versus Nonrandom Coefficient Models for Multilevel AnalysisJournal of Educational Statistics, 1983
- Random versus Nonrandom Coefficient Models for Multilevel AnalysisJournal of Educational Statistics, 1983
- Finding maximum likelihood estimates of patterned covariance matrices by the EM algorithmBiometrika, 1982
- Estimation in Covariance Components ModelsJournal of the American Statistical Association, 1981
- Analyzing Multilevel Data in the Presence of Heterogeneous Within-Class RegressionsJournal of Educational Statistics, 1978
- A Note on a Random Coefficients ModelInternational Economic Review, 1978
- Maximum Likelihood Approaches to Variance Component Estimation and to Related ProblemsJournal of the American Statistical Association, 1977
- Maximum-likelihood estimation for the mixed analysis of variance modelBiometrika, 1967
- Underlining random variables*Statistica Neerlandica, 1966
- Ecological Correlations and the Behavior of IndividualsAmerican Sociological Review, 1950