Factor analysis and scale revision.

This article reviews methodological issues that arise in the application of exploratory factor analysis (EFA) to scale revision and refinement. The authors begin by discussing how the appropriate use of EFA in scale revision is influenced by both the hierarchical nature of psychological constructs and the motivations underlying the revision. Then they specifically address (a) important issues that arise prior to data collection (e.g., selecting an appropriate sample), (b) technical aspects of factor analysis (e.g., determining the number of factors to retain), and (c) procedures used to evaluate the outcome of the scale revision (e.g., determining whether the new measure functions equivalently for different populations).

Using IRT to separate measurement bias from true group differences on homogeneous and heterogeneous scales: an illustration with the MMPI.

The authors present a didactic illustration of how item response theory (IRT) can be used to separate measurement bias from true group differences on homogeneous and heterogeneous scales. Several bias detection methods are illustrated with 12 unidimensional Minnesota Multiphasic Personality Inventory (MMPI) factor scales (Waller, 1999) and the 13 multidimensional MMPI validity and clinical scales. The article begins with a brief review of MMPI bias research and nontechnical reviews of the 2-parameter logistic model (2-PLM) and several IRT-based methods for bias detection. A goal of this article is to demonstrate that homogeneous and heterogeneous scales that are composed of biased items do not necessarily yield biased test scores. To that end, the authors perform differential item- and test-functioning analyses on the MMPI factor, validity, and clinical scales using data from 511 Blacks and 1,277 Whites from the California Youth Authority.

An item response theory analysis of self-report measures of adult attachment.

Self-report measures of adult attachment are typically scored in ways (e.g., averaging or summing items) that can lead to erroneous inferences about important theoretical issues, such as the degree of continuity in attachment security and the differential stability of insecure attachment patterns. To determine whether existing attachment scales suffer from scaling problems, the authors conducted an item response theory (IRT) analysis of 4 commonly used self-report inventories: Experiences in Close Relationships scales (K. A. Brennan, C. L. Clark, & P. R. Shaver, 1998), Adult Attachment Scales (N. L. Collins & S. J. Read, 1990), Relationship Styles Questionnaire (D. W. Griffin & K. Bartholomew, 1994) and J. Simpson's (1990) attachment scales. Data from 1,085 individuals were analyzed using F. Samejima's (1969) graded response model. The authors' findings indicate that commonly used attachment scales can be improved in a number of important ways. Accordingly, the authors show how IRT techniques can be used to develop new attachment scales with desirable psychometric properties.

A Method for Generating Simulated Plasmodes and Artificial Test Clusters with User-Defined Shape, Size, and Orientation.

We present a simple method for generating simulated plasmodes and artificial test clusters with user-defined shape, size, and orientation. Our method differs from other cluster generation techniques in that it focuses on the validity of the cluster indicators. For J clusters, indicator validity is defined as the squared correlation ratio between the cluster indicator (i.e., the observed variable) and J-1 dummy variables. The within-cluster correlation structure and the univariate distributions of the cluster indicators are specified with procedures outlined by Fleishman (1978) and Vale and Maurelli (1983). Simulation results illustrate the utility of the method for cluster analysis evaluation research.

The prevalence and biometric structure of pathological dissociation in the general population: taxometric and behavior genetic findings.

Taxometric and biometric analyses were conducted on 2 North American samples to investigate the prevalence and biometric structure of pathological dissociation. Results indicated that approximately 3.3% of the general population belongs to a pathological dissociative taxon. A brief 8-item self-report scale called the DES-T can be used to calculate taxon membership probabilities in clinical and nonclinical samples of adults (a SAS scoring program is provided for this purpose). The genetic and environmental architecture of pathological dissociative symptoms was explored by conducting a biometric analysis on DES-T ratings from 280 identical and 148 fraternal twins. The findings suggest that approximately 45% of the observed variance on the DES-T can be attributed to shared environmental influences. The remaining variance is due to nonshared environmental influences.

A DeFries and Fulker regression model for genetic nonadditivity.

Parameter estimates from the DeFries and Fulker [(DF) Behav. Genet. 15:462-473, 1985] regression method can be greater than unity or less than zero. This occurs when the monozygotic correlation is greater than twice the dizygotic correlation. Sensible values can be obtained in these cases by fitting a constrained DF model that estimates genetic and nonshared environmental variance components only. In this article I demonstrate that the original Df model yields positively biased heritability estimates and negatively biased estimates of shared environmentality when data are significantly influenced by genetic nonadditivity. The magnitude of the bias is algebraically expressed. I then describe a simple regression equation that provides unbiased estimates of the standardized additive and dominance genetic variance components. Results of a study of 6 million twin pairs from the Monte Carlo Twin Registry demonstrate that the DF additive and dominance genetic parameter estimates are virtually identical to those obtained by maximum-likelihood procedures. Finally, I derive the expectations for the constrained DF model and show that the genetic parameter estimates from this model are negatively biased estimates of broad-sense heritability.

Genetic Tobit factor analysis: quantitative genetic modeling with censored data.

Parameters of quantitative genetic models have traditionally been estimated by either algebraic manipulation of familial correlations (or familial mean squares), biometric model fitting, or multiple-group covariance structure analysis. With few exceptions, researchers who have used these methods for the analysis of twin data have assumed that their data were multinormal and, consequently, have used normal-theory estimation methods. It is shown that normal-theory methods produce biased genetic and environmental parameter estimates when data are censored. Specifically, with censored data, (1) normal-theory estimates of narrow-sense heritability are either positively or negatively biased, whereas (2) estimates of shared-familial environmental variance are always biased downward. An alternative method for estimating genetic and environmental parameters from censored twin data is proposed. The method is called genetic Tobit factor analysis (GTFA) and is an extension of the Tobit factor analysis model developed by Muthén (Br. J. Math. Stat. Psychol. 42, 241-250, 1989). Using a Monte Carlo design, the performance of GTFA is compared to traditional quantitative genetic methods in both large and small data sets. The results of this study suggest that GTFA is the preferred method for the genetic modeling of censored data obtained from twins.

Genetic and environmental influences on item response pattern scalability.

Numerous studies have examined how genetic and environmental factors determine individual differences on multi-item personality scales. Few studies, however, have examined how genes and the environment influence the route by which individuals obtain their scores on these scales. Specifically, on a multi-item test, dozens of item response patterns result in equivalent total scores, though some response patterns are more likely to be observed than others. For many scales it may be of interest to determine the genetic and environmental influences on the item response patterns, as well as the sum of the item responses. We discuss a latent trait measure of item response pattern scalability, called Zl (Levine and Drasgow, 1982), and investigate the properties of this index from a behavioral genetics perspective. Using a large sample of identical and fraternal twins from the Minnesota Twin Registry (Lykken et al., 1990), item response pattern scalability is shown to be moderately heritable. On the four scales of the Multidimensional Personality Questionnaire (Tellegen, 1982) that were investigated, approximately 20% of the variation in scalability was due to genetic diversity between subjects of our sample. Follow-up analyses, using a factor-analytically based, genotype-environment model of item response behavior, indicated that specific genetic and environmental factors play a substantial role in determining item response pattern variation.

Assessment of coping: an empirical illustration of the problem of inapplicable items.

The prevalence and effects of inapplicable items on the Ways of Coping Questionnaire (WOC) were studied. Three hundred seventy-six college students completed the WOC twice, referring to both heterogeneous (i.e., different types of) and homogeneous (the same) stressors. In addition to the standard procedure for administering this questionnaire, subjects were asked to identify items that were inapplicable to the stressors with which they were coping. Item inapplicability ranged from 2.1% to 83.9%, with a mean of 21.5% inapplicability across all WOC items. Application of a correction for item inapplicability demonstrated that situational effects are confounded by this phenomenon and may not hold up when the correction is applied. Implications of these findings include the possibility that previous research has over- or misidentified situational effects on coping.

The Tridimensional Personality Questionnaire: Structural Validity and Comparison with the Multidimensional Personality Questionnaire.

Cloninger's Tridimensional Personality Questionnaire (TPQ) is a self-report inventory designed to assess Novelty Seeking, Harm Avoidance, and Reward Dependence, the three primary dimensions of his Biosocial Learning Model of normal and abnormal personality. We examined the structural validity of the TPQ and the relations among the TPQ lower- and higher-order scales to those of the Multidimensional Personality Questionnaire (MPQ; Tellegen, 1982) in a sample of 1,236 adults. The factor structure of the TPQ was congruent with Cloninger's predicted three-factor genotypic structure with one notable exception: the component scales of the Reward Dependence dimension share essentially no variance, and thus load on different factors. Both bivariate and multivariate analyses indicate that the TPQ and the MPQ share considerable variance, but that each inventory contains variance unpredicted by the other. In addition, the TPQ Harm Avoidance dimension appears to tap primarily a Negative Emotionality or neuroticism factor, rather than a disposition toward behavioral inhibition. These results support a number of Cloninger's predictions concerning the structural and external validity of the TPQ, but also suggest that the TPQ may fail to adequately operationalize several components of his model.

Computerized adaptive personality assessment: an illustration with the Absorption scale.

This article introduces the theory behind and applications of adaptive personality assessment based on the item response theory. Two adaptive testing strategies were compared: (a) fixed test length and (b) clinical decision. Real-data simulations, based on the item responses from 1,000 subjects who had previously taken the 34-item Absorption scale (Tellegen, 1982) by means of paper-and-pencil format, were used to illustrate these strategies. Results suggest that computerized adaptive personality assessment works impressively well. With the fixed-test-length strategy, a 50% savings in administered items was achieved with little loss of measurement precision. In the clinical-decision testing strategy, individuals who were extreme on the Absorption trait were identified with perfect accuracy using, on average, 25% of the available items. The implications of these results for personality research and assessment are discussed.

The Effect of Inapplicable Item Responses on the Structure of Behavioral Checklist Data: A Cautionary Note.

In recent years, researchers from varied disciplines have come to rely on behavioral checklists for the assessment of behavior and personality processes. Although checklist items of specific behavioral content are often irrelevant or inapplicable to subgroups of respondents, response categories designed specifically to measure this factor are often conspicuously absent from inventory protocols. A Monte Carlo simulation was conducted to investigate the effect of inapplicable item responses on the structure of behavioral checklist data. Results indicate that the presence of a small to moderate percentage of inapplicable item responses can seriously obfuscate data analyses.

Deficient motor synchrony in schizophrenic disorders: clinical correlates.

A laboratory measure of synchronization was used to assess the differential ability of schizophrenics, affectives, and normal controls to take advantage of stimulus predictability. It was hypothesized that (1) schizophrenics will perform on this task in a way that distinguishes them from other groups, (2) clinically observed motor anomalies will be associated with deficient motor synchrony, and (3) deficient motor synchrony will be associated with more severe clinical ratings of thought disorder. Twenty-one schizophrenic, 8 affective, and 16 normal controls were studied. The results were consistent with the hypotheses; schizophrenic subjects had distinctive performance patterns, especially at 40 bpm, which was associated with both motor and thinking disturbance. The authors discuss additional analyses that suggest that deficient motor synchrony is associated with negative symptoms, certain ward behaviors, and not with demographic variables, and that among unmedicated schizophrenic subjects, the performance patterns are worse.