A multitrait-multimethod analysis of the MCMI-II clinical syndrome scales

Multitrait-multimethod procedures and factor analyses were utilized to assess convergent/discriminant validity of the revised version of the Millon Clinical Multiaxial Inventory (MCMI-II) clinical syndrome scales and effects of item overlap on scale performance. Except for the Delusional Disorder Scale (PP), convergent validity was supported, but poor discriminant validity was found for the Alcohol Dependence (B) scale, Drug Dependence (T) scale, and Scale PP. Item overlap did not influence performance for many scales. In fact, results indicate that some scales, such as Bipolar:Manic (N), may perform better in overlapping form, whereas others, such as Somatoform (H) and Scales B and T, are most affected by item overlap. Factor structure was generally stable, regardless of whether overlapping or nonoverlapping scales were utilized. However, substance abuse/dependence emerged as an independent factor when common item artifact was removed. Findings are discussed in terms of their implications for clinical practice and further research.     

Masculinity,femininity, and self-esteem: A multitrait-multimethod analysis

A multitrait-multimethod matrix analysis was conducted to determine (1) the degree to which self-report trait and behavior measures of sex role orientation converge, (2) the degree to which measures of masculinity and self-esteem can be discriminated, and (3) the relative degree of relationship of trait and behavior measures of sex role orientation to self-esteem. The results indicated that trait and behavior measures of masculinity and femininity showed little convergence, that trait masculinity did not meet the criteria for discrimination from self-esteem, and that trait masculinity was more strongly correlated with self-esteem than was behavior masculinity. These results were stronger for women than for men.I would like to express my appreciation to John Millar and Shauna Gatten for their assistance in data collection and coding, and to Paul Biner and an anonymous reviewer for their comments on earlier drafts of this article.     

The analysis of multitrait-multimethod matrices via constrained components analysis

Multitrait-Multimethod (MTMM) matrices are often analyzed by means of confirmatory factor analysis (CFA). However, fitting MTMM models often leads to improper solutions, or non-convergence. In an attempt to overcome these problems, various alternative CFA models have been proposed, but with none of these the problem of finding improper solutions was solved completely. In the present paper, an approach is proposed where improper solutions are ruled out altogether and convergence is guaranteed. The approach is based on constrained variants of components analysis (CA). Besides the fact that these methods do not give improper solutions, they have the advantage that they provide component scores which can later on be used to relate the components to external variables. The new methods are illustrated by means of simulated data, as well as empirical data sets.This research has been made possible by a fellowship from the Royal Netherlands Academy of Arts and Sciences to the first author. The authors are obliged to three anonymous reviewers and an associate editor for constructive suggestions on the first version of this paper.     

A typology of behavior problems in preschool children

The purpose of this study was to produce a typology of behavior problems in preschool children. Distinct subtypes were identified through the use of cluster analytic techniques on data from the Behavior Assessment System for Children (BASC)-Parent Rating Scales. Analyses were based on archival data collected on a sample of 268 children, aged 2 to 5 years, who were referred to determine their suitability for a day treatment program. Five distinct and reliable subtypes of preschool children's behavior problems emerged. Ratings of adaptive and social skills based on the BASC Adaptive Scales and measures from the Parenting Stress Index served as external variables to assess the distinctiveness of the derived subtypes. The findings provide evidence for the reliability and validity of the subtypes identified in this study that are consistent with the findings of previous subtyping investigations focusing on school-aged children.     

Convergent and discriminative validity of interview and questionnaire measures of personality disorder in mentally disordered offenders: a multitrait-multimethod analysis using confirmatory factor analysis

Measures of personality disorder from the International Personality Disorder Examination, Personality Diagnostic Questionnaire, and Millon Clinical Multiaxial Inventory (MCMI-II) were obtained from detained male mentally disordered offenders (N = 156), and convergent and discriminant validity were examined by confirmatory factor analysis of the multitrait-multimethod matrix. Hierarchical comparisons of models varying in their specification of trait and method variance established the appropriateness of a model supporting both convergence and discrimination across methods, but these were variable across constructs and measures. Convergence was good for avoidant, schizoid, and antisocial disorders, but poor for histrionic, narcissistic, and obsessive-compulsive disorders. Avoidant, schizoid, and schizotypal disorders were not clearly distinguishable from each other. Measurement error attributable to method variance was substantial for all instruments and for most disorders. The commonly alleged superiority of interview over questionnaire methods was not supported, and the MCMI-II demonstrated proportionately more "true" variance. However, assessment methods may be differentially sensitive to different kinds of personality disorder problems.     

A multitrait-multimethod model with minimal assumptions

A new model of confirmatory factor analysis (CFA) for multitrait-multimethod (MTMM) data sets is presented. It is shown that this model can be defined by only three assumptions in the framework of classical psychometric test theory (CTT). All other properties of the model, particularly the uncorrelated-ness of the trait with the method factors are logical consequences of the definition of the model. In the model proposed there are as many trait factors as different traits considered, but the number of method factors is one fewer than the number of methods included in an MTMM study. The covariance structure implied by this model is derived, and it is shown that this model is identified even under conditions under which other CFA-MTMM models are not. The model is illustrated by two empirical applications. Furthermore, its advantages and limitations are discussed with respect to previously developed CFA models for MTMM data.     

An empirical application of confirmatory factor analysis to the multitrait-multimethod matrix

Though most social psychologists are aware of the use of factor analysis as an exploratory method to uncover latent dimensions, factor analysis can also be used in a confirmatory mode to test specific hypotheses. A confirmatory solution is unique and cannot be rotated. One possible use of confirmatory factor analysis is with the multitrait-multimethod matrix. Jaccard, Weber, and Lundmark's (1975) multitrait-multimethod matrix of two traits and four methods is analyzed using confirmatory factor analysis. Given the very small sample size the analysis is primarily illustrative. A simple two factor model with errors of measurement correlated across measures using the same method satisfactorily fits the data. Both discriminant and convergent validity are high and none of the methods has higher reliability or less correlated measurement error. The reliability of measures estimated by maximum likelihood factor analysis is lower than the test-retest reliability since method variance is subtracted out. Confirmatory factor analysis is recommended over the Campbell-Fiske criteria.     

Personality and the four faces of positive affect: A multitrait-multimethod analysis using self- and peer-report

The present study investigates the relationship between personality and positive affects (contentment, love, interest, and joy). Using multitrait-multimethod analysis and structural equation modeling procedures with self- and peer-report, the authors demonstrated the importance of differentiating between the positive affects. A multidimensional model fitted better than a general-factor model. Concerning the personality–affect relationship, the strongest relations were found between extraversion and both joy and love, agreeableness and love, openness to experiences and interest, and neuroticism and contentment. Thus, results suggest that it is insufficient to concentrate solely on extraversion when investigating the personality–positive affect relationship.     

Hyperactive behavior in english schoolchildren: A questionnaire survey

The differences in apparent prevalence of childhood hyperactivity between England and North America are more than 20-fold. This paper seeks to illuminate the reasons for the difference by applying a standard measure [Conners Teacher Rating Scale] to 437 English schoolchildren, aged 6 to 9 years, and comparing them with published surveys from the United States and Australasia. The rating scale had moderate interrater reliability and stability over time; it distinguished between children referred for psychiatric treatment and normal schoolchildren. Component analysis yielded a factor of hyperactivity-inattentiveness, distinct from conduct disorder. The mean hyperactivity scores of English children were higher than in most U.S. surveys but lower than in New Zealand. The national differences in rates of diagnosed hyperactivity are likely to reflect diagnostic inconsistency rather than true differences in children's behavior.     

A multitrait-multimethod analysis of variance of teachers' ratings of aggression,hyperactivity, and inattention

The convergent and discriminant validities of three teacher rating scale measures of the traits of hyperactivity, aggression, and inattention were explored, using the multitrait-multimethod matrix approach of Campbell and Fiske (1959), as well as an analysis of variance procedure (Stanley, 1961). In the present study teachers rated children from their elementary school classrooms on the above traits. The results provided strong evidence for convergent validity. Data also indicated that these traits can be reliably differentiated by teachers, suggesting that research aimed at better understanding the unique contributions of hyperactivity, aggression, and inattention is warranted. The respective benefits of analyzing multitrait-multimethod matrices by employing the ANOVAprocedure or by using the Campbell and Fiske (1959) criteria were discussed.     

Hiding valuables: A questionnaire study of mnemonically risky behavior

A survey investigated incidences where individuals intentionally hid objects but later forgot the item's location. Surveys from 111 younger and 99 older adults yielded the following prototypical scenario: the person hid the valuable 1 year ago to prevent theft, and waited 1 month before an initial 30-minute search, which included looking in dresser drawers plus 1.5 other locations. The person made three additional searches (total searches = 4) for the missing valuable over a 2 week period. Three-quarters of the hidden valuables were eventually found (usually by accident) about 1 month after hiding the object. About half of the time, the person had already searched the hiding area. Older adults typically hid jewellery from thieves, while younger adults hid money from friends and relatives. Older adults searched more often for longer periods of time for the hidden object, and took longer to eventually find the object, compared with younger adults.     

A further validation of the bem sex role inventory: A multitrait-multimethod study

The present study used a multitrait-multimethod technique [D.T. Campbell and D. W. Fiske (1959) “Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix,” Psychological Bulletin, 56, 81–105] to examine Bem's [(1974) “The Measurement of Psychological Androgyny,” Journal of Consulting and Clinical Psychology, 42, 152–162] use of an instrumental-expressive dichotomy to conceptualize the Masculinity and Femininity scales of the Bem Sex Role Inventory as statistically orthogonal constructs. Results indicate that while perhaps reliable, masculinity and femininity lack clear convergent and discriminant validity. Therefore, Bem's contention that masculinity and femininity are orthogonal constructs is questioned. Results are discussed in light of trait- vs. method-driven research.     

A comparative analysis of the effects of social skill training and teacher-directed contingencies on social behavior of preschool children with disabilities

This study investigated the effects of an analog social skill training program and a teacher-directed contingency strategy on promoting social responding among four preschool children with mild disabilities and their peers. The social skill training program consisted of five lessons targeting the skill areas: Initiating, responding, and maintaining social interactions. The teacher directed contingency consisted of prompting target children when they were not socially involved with peers, and providing verbal praise when they were socially involved with peers. Subjects were divided into two training pairs and received both interventions in a reversal design. Presentation order was counterbalanced across the two training pairs. Results indicated that the teacher directed contingency was more effective in promoting higher levels of social responding in a free-play setting. Results also suggest that social skill training may not always be a necessary prerequisite to promote increased social interactions among children who display low rates of social interaction with peers, but possess necessary skills.     

The decomposition of multitrait-multimethod matrices

Existing models for multitrait-multimethod matrices are evaluated and a new composite direct product model with a multiplicative property is proposed. The estimation of parameters in the composite direct product model is considered and an example of its application is provided. A cross-validation to compare models is reported.     

Distinguishing personality and biodata items using confirmatory factor analysis of multitrait-multimethod matrices

Hard biodata, soft biodata, and personality items were developed to measure three personal constructs (Dependability, Demeanor, and Ambition). This resulted in nine sets of items forming a three-method (personality, soft biodata, and hard biodata) by three-trait (Dependability, Demeanor, and Ambition) Multitrait-Multimethod (MTMM) matrix. Four models were proposed to represent the nine scales and each was tested on a sample of 477 job applicants and subsequently cross-validated on a sample of 958 job applicants, using confirmatory factor analysis. The model where the personality and soft biodata items represented one factor and the hard biodata items represented another, distinct factor, was both the best fitting and most parsimonious model, thus confirming Asher's (1972) taxonomy.This research was based in part on a doctoral dissertation submitted to Wayne State University. I would like to thank committee members Lois Tetrick (Chair), Alan Bass, Sebastiano Fisicaro, and Thomas Naughton. An earlier version of this paper was presented at the 8th annual conference of the Society for Industrial and Organizational Psychology (SIOP), San Francisco, CA: April, 1993.