Multilevel Analysis of Educational Data focuses on the principles and procedures used in the evaluation of educational progress.
The selection first offers information on some applications of multilevel models to educational data, empirical Bayes methods, and a hierarchical item-response model for educational testing. Discussions focus on the interface between levels, group-level model for content elements, an application of empirical Bayes, validity generalization, improving law school validity studies, and summarizing evidence in randomized experiments on coaching. The text then takes a look at difficulties with Bayesian inference for random effects and multilevel aspects of varying parameters in structural models.
The book elaborates on models for multilevel response variables with an application to growth curves and the issues and problems emerging from the application of multilevel models in British studies of school effectiveness, including enduring questions, two-level models, estimation and prediction, and econometric random coefficient modeling. The selection is a dependable reference for educators and researchers interested in the evaluation of educational progress.
- Bayesian methods
- Empirical Bayes
- Generalized least squares
- Profile likelihoods
- E-M algorithm
- Fisher scoring procedures
- Both educational and social science applications
D.B. Rubin, Some Applications of Multilevel Models of Educational Data.
H.I. Braun, Empirical Bayes Methods: A Tool for Exploratory Analysis.
R.J. Mislevy and R.D. Bock, A Hierarchical Item-response Model for Educational Testing.
C. Lewis, Difficulties with Bayesian Inference for Random Effects.
B. Muthéen and A. Satorra, Multilevel Aspects of Varying Parameters in Structural Models.
R.K. Tsutakawa--Discussant, Discussions of Papers by Mislevy and Bock, Muthéen and Satorra, and Lewis.
H. Goldstein, Models for Multilevel Response Variables with an Application to Growth Curves.
J. Gray, Multilevel Models: Issues and Problems Emerging from their Recent Application in British Students of School Effectiveness.
P. McCullagh, Discussant, What Can Go Wrong with Iteratively Re-weighted Least Squares?
A.S. Bryk and S.W. Raudenbush, Toward a More Appropriate Conceptualization of Research on School Effects: A Three-Level Hierarchical Linear Model.
S.W. Raudenbush and A.S. Bryk, Quantitative Models for Estimating Teacher and School Effectiveness.
L. Burstein, K.-S. Kim, and G. Delandshere, Multilevel Investigations of Systematically Varying Slopes: Issues, Alternatives, and Consequences.
H. Swaminathan, -Discussant, Multilevel Data Analysis: A Discussion.
M. Aitkin, Profile Predictive Likelihood for Random Effects in the Two-Level Model.
N.T. Longford, Fisher Scoring Algorithm for Variance Component Analysis of Data with Multilevel Structure.
P.W. Holland, Discussant, Discussion of Aitkin's and Longford's Paper.
R.D. Bock, Addendum-Measurement of Human Variation: A Two-Stage Model.
- No. of pages:
- © Academic Press 1989
- 28th February 1989
- Academic Press
- eBook ISBN:
University of Chicago, Illinois, U.S.A.