Hierarchical Linear Models for Multivariate Outcomes

In this article, we develop a class of two-stage models to accommodate three common characteristics of behavioral data. First, behavior is invariably multivariate in its conceptualization and communication. Separate univariate analyses of related outcome variables are fraught with potential interpretive blind spots for the researcher. This practice also suffers, from an inferential standpoint, because it fails to take advantage of any redundant information in the outcomes. Second, studies of behavior, especially in experimental research, employ smaller samples. This situation raises issues of robustness of inference with respect to outlying individuals. Third, the outcome variable may have observations missing because of accidents or by design. The model permits the estimation of the full spectrum of plausible measurement error structures while using all the available information. Maximum likelihood estimates are obtained for various members of a multivariate hierarchical linear model (MHLM), and, in the context of several illustrative examples, these estimates match closely the results from a Bayesian approach to the normal-normal MHLM and to the normal-t MHLM.

[1]  A. Bryk,et al.  Early vocabulary growth: Relation to language input and gender. , 1991 .

[2]  Y. Thum.  Analysis of individual variation : A multivariate hierarchical linear model for behavioral data , 1994 .

[3]  B. Muthén.  Latent variable modeling in heterogeneous populations , 1989 .

[4]  Jan de Leeuw,et al.  Random Coefficient Models for Multilevel Analysis , 1986 .

[5]  R. Potthoff,et al.  A generalized multivariate analysis of variance model useful especially for growth curve problems , 1964 .

[6]  John B. Willett,et al.  Understanding correlates of change by modeling individual differences in growth , 1985 .

[7]  R. Darrell Bock,et al.  ADDENDUM: MEASUREMENT OF HUMAN VARIATION: A TWO-STAGE MODEL , 1989 .

[8]  J. Magnus,et al.  The Commutation Matrix: Some Properties and Applications , 1979 .

[9]  H. Y. Huang.  Unified approach to quadratically convergent algorithms for function minimization , 1970 .

[10]  J. Mathews,et al.  Extensions to multivariate normal models for pedigree analysis , 1982, Annals of human genetics.

[11]  R. Darrell Bock,et al.  Multilevel analysis of educational data , 1989 .

[12]  S. Raudenbush,et al.  Empirical Bayes Meta-Analysis , 1985 .

[13]  Harvey Goldstein,et al.  Balanced versus unbalanced designs for linear structural relations in two‐level data , 1989 .

[14]  Jeremy MG Taylor,et al.  Robust Statistical Modeling Using the t Distribution , 1989 .

[15]  R. Jennrich,et al.  Unbalanced repeated-measures models with structured covariance matrices. , 1986, Biometrics.

[16]  William C. Davidon,et al.  Variable Metric Method for Minimization , 1959, SIAM J. Optim..

[17]  E. C. Macrae.  Matrix Derivatives with an Application to an Adaptive Linear Decision Problem , 1974 .

[18]  M. Seltzer,et al.  Sensitivity Analysis for Fixed Effects in the Hierarchical Model: A Gibbs Sampling Approach , 1993 .

[19]  Anthony S. Bryk,et al.  A Hierarchical Model for Studying School Effects , 1986 .

[20]  Donald B. Rubin,et al.  A Note on Bayesian, Likelihood, and Sampling Distribution Inferences , 1978 .

[21]  R. D. Bock,et al.  Multivariate Statistical Methods in Behavioral Research , 1978 .

[22]  N. Longford.  A fast scoring algorithm for maximum likelihood estimation in unbalanced mixed models with nested random effects , 1987 .

[23]  Ea Coornish,et al.  The Multivariate t-Distribution Associated with a Set of Normal Sample Deviates , 1954 .

[24]  Harvey Goldstein,et al.  A general model for the analysis of multilevel data , 1988 .

[25]  R. Bargmann,et al.  MAXIMUM LIKELIHOOD ESTIMATION WITH INCOMPLETE MULTIVARIATE DATA , 1964 .

[26]  Roger Fletcher,et al.  A Rapidly Convergent Descent Method for Minimization , 1963, Comput. J..

[27]  D. Harville.  Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems , 1977 .

[28]  Nicholas T. Longford,et al.  Factor analysis for clustered observations , 1992 .

[29]  P. S. Dwyer.  Some Applications of Matrix Derivatives in Multivariate Analysis , 1967 .

[30]  Georgia B. Reed,et al.  Introduction to numerical analysis , 1968 .

[31]  Donald B. Rubin,et al.  Finding maximum likelihood estimates of patterned covariance matrices by the EM algorithm , 1982 .

[32]  R. Fletcher,et al.  A New Approach to Variable Metric Algorithms , 1970, Comput. J..

[33]  John S. J. Hsu,et al.  Bayesian Marginal Inference , 1989 .

[34]  J. Berger.  Statistical Decision Theory and Bayesian Analysis , 1988 .

[35]  David F. Shanno,et al.  Algorithm 500: Minimization of Unconstrained Multivariate Functions [E4] , 1976, TOMS.

[36]  David F. Shanno,et al.  Remark on “Algorithm 500: Minimization of Unconstrained Multivariate Functions [E4]” , 1980, TOMS.

[37]  D. Bates,et al.  Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data , 1988 .

[38]  M D Schluchter,et al.  Analysis of incomplete multivariate data using linear models with structured covariance matrices. , 1988, Statistics in medicine.

[39]  David Michael Burley,et al.  Studies in optimization , 1974 .

[40]  G. Casella,et al.  Explaining the Gibbs Sampler , 1992 .

[41]  Anthony S. Bryk,et al.  Hierarchical Linear Models: Applications and Data Analysis Methods , 1992 .

[42]  Charles E. McCulloch,et al.  Symmetric Matrix Derivatives with Applications , 1982 .

[43]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[44]  C. Dunnett,et al.  A BIVARIATE GENERALIZATION OF STUDENT'S t-DISTRIBUTION, WITH TABLES FOR CERTAIN SPECIAL CASES , 1954 .

[45]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[46]  Stephen W. Raudenbush,et al.  A Multilevel, Multivariate Model for Studying School Climate With Estimation Via the EM Algorithm and Application to U.S. High-School Data , 1991 .

[47]  S. E. Hills,et al.  Illustration of Bayesian Inference in Normal Data Models Using Gibbs Sampling , 1990 .

[48]  Harvey Goldstein,et al.  Multilevel Models in Education and Social Research , 1987 .

[49]  S. Raudenbush.  Educational Applications of Hierarchical Linear Models: A Review , 1988 .