A comparison of latent trait and latent class analyses of Likert-type data

This paper brings together and compares two developments in the analysis of Likert attitude scales. The first is the generalization of latent class models to ordered response categories. The second is the introduction of latent trait models with multiplicative parameter structures for the analysis of rating scales. Key similarities and differences between these two methods are described and illustrated by applying a latent trait model and a latent class model to the analysis of a set of life satisfaction data. The way in which the latent trait model defines a unit of measurement, takes into account the order of the response categories, and scales the latent classes, is discussed. While the latent class model provides better fit to these data, this is achieved at the cost of a logically inconsistent assignment of individuals to latent classes.The author wishes to thank Clifford C. Clogg, Otis Dudley Duncan and Benjamin D. Wright for their helpful comments on an earlier version of this paper.     

Stability of the Millon Clinical Multiaxial Inventory among psychiatric inpatients

This study investigated the stability of the Millon Clinical Multiaxial Inventory (MCMI) in a sample of psychiatric inpatients over a longer time period and from different perspectives than previous reliability studies. Test-retest reliability was assessed from three perspectives: correlation/regression, equivalence of dimensional structures across testings upon different admissions, and stability of personality style across testings. Some stability from a correlational perspective was found, with higher stability estimates indicated for the basic personality scales in comparison with the clinical symptom scales which is consistent with Millon's theory. The internal structure of the MCMI was essentially identical across testings. Although source of reinforcement and instrumental or coping style were stable across testings beyond chance, the MCMI profiles of a considerable number of subjects were inconsistent across testings. Correspondingly, clinicians should be cautious in using the MCMI to make categorical decisions regarding personality style with inpatients who are tested at admission and who have had repeated hospitalizations.     

Conformity of two sociometric relations

The problem of comparing two sociometric relations or measurements (A andB) recorded in distinct sociomatrices was originally discussed by Katz and Powell in the early 1950's and Hubert and Baker in the late 1970's. The problem is considered again using a probabilistic model designed specifically for discrete-valued network measurements. The model allows for the presence of various structural tendencies, such as reciprocity and differential popularity. A parameter that isolates the tendency for actors to choose other actors on both relations simultaneously is introduced, and estimated conditional on the presence of other parameters that reflect additional important network properties. The parameter is presented as a symmetric index but is also generalized to the predictive (A onB orB onA) situation. This approach to the problem is illustrated with the same data used by the earlier solutions, and the unique nature of the two relations in the data set (A = received choices,B = perceived choices), as it affects the modeling, is discussed. Significance tests for the parameter and related parameters are described, as well as an extension to more than two relations.Research support provided by National Science Grant #SES84-08626 to the University of Illinois at Urbana-Champaign. I am grateful to Dawn Iacobucci and Sheila Weaver for assistance with the research reported here, and to Carolyn Anderson, James Green, David Holtgrave, and four anonymous referees for comments on the paper.     

THE EFFECT OF MODELLING ON DRINKING RATE

Three male college seniors were asked to drink beer at their normal rate in a simulated tavern setting. Each was paired with a confederate, also a male college senior, in an ABACA single subject design. In the baseline conditions, the confederate matched the drinking rate of the subject. Baseline and all subsequent conditions were continued in 1-hr sessions until a stable drinking rate was achieved. In Condition B, the confederate drank either one third more or one third less than the subject's baseline rate. In Condition C, the direction was reversed. All three subjects closely matched the confederate's drinking rate, whether high or low. All subjects reported they were unaware of the true purpose of the study.     

Best linear prediction of composite universe scores

The problem of predicting universe scores for samples of examinees based on their responses to samples of items is treated. A general measurement procedure is described in which multiple test forms are developed from a table of specifications and each form is administered to a different sample of examinees. The measurement model categorizes items according to the cells of such a table, and the linear function derived for minimizing error variance in prediction uses responses to these categories. In addition, some distinctions are drawn between aspects of the approach taken here and the familiar regressed score estimates.The author thanks Robert L. Brennan, Michael J. Kolen, and Richard Sawyer for helpful comments and corrections, and anonymous reviewers for suggested improvements.     

Simplifying continuous monitoring of multiple-response/multiple-subject classroom interactions

In order to facilitate the field monitoring of three subjects interacting according to one or more of 18 response categories, a modified version of several available, but oftentimes mechanically incompatible, observational procedures was designed. Its continuous recording strategy, sectioned into one-minute observational units, enabled researchers to derive highly representative behavior samples, and when accompanied by the specially tailored coding form and recording apparatus, observers achieved over 90% agreement across all reliability sessions. This procedure provides applied researchers with a simple, highly reliable, and adaptable observation tool for continuously and simultaneously monitoring the behaviors of one or more subjects.     

Treatment of overweight children: The relationship between dependent measures and a clinically significant outcome

Behavioral treatments for overweight children have not fared well in achieving or maintaining clinically significant weight losses. The use of more appropriate dependent measures which also take height, sex, and age into consideration is suggested. A multiple-baseline analysis of the behavioral treatment of seven overweight children was conducted. Results support the contention that a more clinically significant outcome emerges when appropriate measures are used.     

Component models for three-way data: An alternating least squares algorithm with optimal scaling features

A review of the existing techniques for the analysis of three-way data revealed that none were appropriate to the wide variety of data usually encountered in psychological research, and few were capable of both isolating common information and systematically describing individual differences. An alternating least squares algorithm was proposed to fit both an individual difference model and a replications component model to three-way data which may be defined at the nominal, ordinal, interval, ratio, or mixed measurement level; which may be discrete or continuous; and which may be unconditional, matrix conditional, or row conditional. This algorithm was evaluated by a Monte Carlo study. Recovery of the original information was excellent when the correct measurement characteristics were assumed. Furthermore, the algorithm was robust to the presence of random error. In addition, the algorithm was used to fit the individual difference model to a real, binary, subject conditional data set. The findings from this application were consistent with previous research in the area of implicit personality theory and uncovered interesting systematic individual differences in the perception of political figures and roles.This paper is part of a Thesis performed by Richard Sands under the direction of Forrest Young at the L. L. Thurstone Psychometric Laboratory, University of North Carolina at Chapel Hill. Thanks are extended to Drs. Charles Schmidt and Andrea Sedlak for the use of their political role data set.     

Structural analysis of subjective categorical data

A general approach to the analysis of subjective categorical data is considered, in which agreement matrices of two or more raters are directly expressed in terms of error and agreement parameters. The method provides focused analyses of ratings from several raters for whom ratings have measurement error distributions that may induce bias in the evaluation of substantive questions of interest. Each rater's judgment process is modeled as a mixture of two components: an error variable that is unique for the rater in question as well as an agreement variable that operationalizes the true values of the units of observation. The statistical problems of identification, estimation, and testing of such measurement models are discussed.The general model is applied in several special cases. The most simple situation is that underlying Cohen's Kappa, where two raters place units into unordered categories. The model provides a generalization and systematization of the Kappa-idea to correct for agreement by chance. In applications with typical research designs, including a between-subjects design and a mixed within-subjects, between-subjects design, the model is shown to disentangle structural and measurement components of the observations, thereby controlling for possible confounding effects of systematic rater bias. Situations considered include the case of more than two raters as well as the case of ordered categories. The different analyses are illustrated by means of real data sets.The authors wish to thank Lawrence Hubert and Ivo Molenaar for helpful and detailed comments on a previous draft of this paper. Thanks are also due to Jens Möller und Bernd Strauß for the data from the 1992 Olympic Games. We thank the editor and three anonymous reviewers for valuable comments on an earlier draft.     

The development of an autonomy scale

The purpose of this study was to develop a valid and relaible instrument to measure four areas of autonomy. The instrument is composed of four subscales that measure the following constructs of autonomy: (a) family loyalty autonomy, (b) value autonomy, (c) emotional autonomy, and (d) behavioral autonomy. A detailed four-step procedure was used to provide the scale with construct, content, and predictive validity. Presented here is the four-stage process used to develop the Autonomy Scale, data supporting the validity and reliability of the scale, and the final version of the instrument.This article is based on dissertation research in partial fulfillment of degree requirements at Texas Woman's University.