Robustness issues in multilevel regression analysis

Statistica Neerlandica - Tập 58 Số 2 - Trang 127-137 - 2004
Cora J. M. Maas1, Joop J. Hox1
1Department of Methodology and Statistics, Faculty of Social Sciences, Utrecht University, P.O. Box 80140, NL-3508 TC Utrecht, the Netherlands

Tóm tắt

A multilevel problem concerns a population with a hierarchical structure. A sample from such a population can be described as a multistage sample. First, a sample of higher level units is drawn (e.g. schools or organizations), and next a sample of the sub‐units from the available units (e.g. pupils in schools or employees in organizations). In such samples, the individual observations are in general not completely independent. Multilevel analysis software accounts for this dependence and in recent years these programs have been widely accepted. Two problems that occur in the practice of multilevel modeling will be discussed. The first problem is the choice of the sample sizes at the different levels. What are sufficient sample sizes for accurate estimation? The second problem is the normality assumption of the level‐2 error distribution. When one wants to conduct tests of significance, the errors need to be normally distributed. What happens when this is not the case? In this paper, simulation studies are used to answer both questions. With respect to the first question, the results show that a small sample size at level two (meaning a sample of 50 or less) leads to biased estimates of the second‐level standard errors. The answer to the second question is that only the standard errors for the random effects at the second level are highly inaccurate if the distributional assumptions concerning the level‐2 errors are not fulfilled. Robust standard errors turn out to be more reliable than the asymptotic standard errors based on maximum likelihood.

Từ khóa


Tài liệu tham khảo

Afshartous D., 1995, Determination of sample size for multilevel model design

Browne W. J.(1998) Applying MCMC methods to multilevel models Unpublished Ph.D. Thesis University of Bath UK.

10.1007/s001800000041

Bryk A. S., 1992, Hierarchical linear models

Busing F., 1993, Distribution characteristics of variance estimates in two‐level models

Cohen J., 1988, Statistical power analysis for the behavioral sciences

Eliason S. R., 1993, Maximum likelihood estimation, 10.4135/9781412984928

Goldstein H., 1995, Multilevel statistical models

Hox J.J., 1998, Multilevel modeling: when and why, 147

Hox J. J., 2002, Multilevel analysis; techniques and applications, 10.4324/9781410604118

10.1207/S15328007SEM0802_1

Huber P. J., 1967, Proceedings of the Fifth Berkeley symposium on mathematical statistics and probability, 221

10.4135/9781849209366

10.2307/1164848

Longford N. T., 1993, Random coefficient models

Rasbash J., 2000, A user's guide to MlwiN, Multilevel Models Project

Raudenbush S. W., 2002, Hierarchical linear models

Raudenbush S., 2000, HLM 5, Hierarchical linear and nonlinear modeling

Snijders T. A. B., 1999, Multilevel analysis. An introduction to basic and advanced multilevel modeling

Van der Leeden R., 1994, First iteration versus IGLS/RIGLS estimates in two‐level models: a Monte Carlo study with ML3

Van der Leeden R., 1997, Applications of bootstrap methods for two‐level models

Verbeek M., 2000, A guide to modern econometrics

10.2307/1912526

Thông tin
Thông tin xuất bản

Statistica Neerlandica

Tập 58 Số 2

127-137

Thông tin tác giả