Random-effects modeling of categorical response data

被引:110
作者
Agresti, A [1 ]
Booth, JG [1 ]
Hobert, JP [1 ]
Caffo, B [1 ]
机构
[1] Univ Florida, Dept Stat, Gainesville, FL 32611 USA
来源
SOCIOLOGICAL METHODOLOGY 2000, VOL 30 | 2000年 / 30卷
关键词
D O I
10.1111/0081-1750.t01-1-00075
中图分类号
C91 [社会学];
学科分类号
030301 ; 1204 ;
摘要
In many applications observations have some type of clustering, with observations within clusters tending to be correlated. A common instance of this occurs when each subject in the sample undergoes repeated measurement, in which case a cluster consists of the set of observations for the subject. One approach to modeling clustered data introduces cluster-level random effects into the model. The use of random effects in linear models for normal responses is well established. By contrast, random effects have only recently seen much use in models for categorical data. This chapter surveys a variety of potential social science applications of random effects modeling of categorical data. Applications discussed include repeated measurement for binary or ordinal responses, shrinkage to improve multiparameter estimation of a set of proportions or rates, multivariate latent variable modeling, hierarchically structured modeling, and cluster sampling. The models discussed belong to the class of generalized linear mixed models (GLMMs), an extension of ordinary linear models that permits nonnormal response variables and both fixed and random effects in the predictor term. The models are GLMMs for either binomial or Poisson response variables, although we also present extensions to multicategory (nominal or ordinal) responses. We also summarize some of the technical issues of model-fitting that complicate the fitting of GLMMs even with existing software.
引用
收藏
页码:27 / 80
页数:54
相关论文
共 117 条
[81]   LOGIT AND MULTILEVEL LOGIT MODELING OF COLLEGE GRADUATION FOR 1984-1985 FRESHMAN STUDENT-ATHLETES [J].
MCARDLE, JJ ;
HAMAGAMI, F .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1994, 89 (427) :1107-1123
[82]  
McCullagh P., 2019, Generalized Linear Models
[83]   MAXIMUM-LIKELIHOOD VARIANCE-COMPONENTS ESTIMATION FOR BINARY DATA [J].
MCCULLOCH, CE .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1994, 89 (425) :330-335
[84]   Maximum likelihood algorithms for generalized linear mixed models [J].
McCulloch, CE .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (437) :162-170
[85]  
MCGILCHRIST CA, 1994, J ROY STAT SOC B MET, V56, P61
[86]   ASYMPTOTIC PROPERTIES OF MAXIMUM LIKELIHOOD ESTIMATES IN MIXED MODEL OF ANALYSIS OF VARIANCE [J].
MILLER, JJ .
ANNALS OF STATISTICS, 1977, 5 (04) :746-762
[87]   CHILD HEALTH, BREAST-FEEDING, AND SURVIVAL IN MALAYSIA - A RANDOM-EFFECTS LOGIT APPROACH [J].
MONTGOMERY, MR ;
RICHARDS, T ;
BRAUN, HI .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1986, 81 (394) :297-309
[88]   Family and sociodemographic in fluences on patterns of leaving home in postwar Britain [J].
Murphy, M ;
Wang, DL .
DEMOGRAPHY, 1998, 35 (03) :293-305
[89]   Design and analysis issues in community-based drug abuse prevention [J].
Murray, DM ;
Moskowitz, JM ;
Dent, CW .
AMERICAN BEHAVIORAL SCIENTIST, 1996, 39 (07) :853-867
[90]   Latent variable modeling of longitudinal and multilevel data [J].
Muthen, B .
SOCIOLOGICAL METHODOLOGY 1997, VOL 27, 1997, 27 :453-480