Variance Component Testing in Multilevel Models
- 1 June 2001
- journal article
- Published by American Educational Research Association (AERA) in Journal of Educational and Behavioral Statistics
- Vol. 26 (2), 133-152
- https://doi.org/10.3102/10769986026002133
Abstract
Available variance component tests are reviewed and three new score tests are presented. In the first score test, the asymptotic normal distribution of the test statistic is used as a reference distribution. In the other two score tests, a Satterthwaite approximation is used for the null distribution of the test statistic. We evaluate the performance of the score tests and other available tests by means of a Monte Carlo study. The new tests are computationally relatively cheap and have good power properties.Keywords
This publication has 12 references indexed in Scilit:
- Effects of Schools, Teaching Staff and Classes on Achievement and Well-Being in Secondary Education: Similarities and Differences Between School OutcomesSchool Effectiveness and School Improvement, 2000
- Exact Tests for Variance ComponentsBiometrics, 1996
- Exact Tests for Variance Components in Unbalanced Mixed Linear ModelsBiometrics, 1993
- Lack-of-Fit Tests Based on Near or Exact ReplicatesThe Annals of Statistics, 1989
- A fast scoring algorithm for maximum likelihood estimation in unbalanced mixed models with nested random effectsBiometrika, 1987
- Multilevel mixed linear model analysis using iterative generalized least squaresBiometrika, 1986
- Applying Wald's Variance Component TestThe Annals of Statistics, 1983
- The Use of Control Variates in Monte Carlo Estimation of PowerJournal of the Royal Statistical Society Series C: Applied Statistics, 1982
- Asymptotic Properties of Maximum Likelihood Estimates in the Mixed Model of the Analysis of VarianceThe Annals of Statistics, 1977
- Maximum Likelihood Approaches to Variance Component Estimation and to Related ProblemsJournal of the American Statistical Association, 1977