A statistical evaluation of Edwards Personal Preference Schedule.

What is the nature of the factor structure of the Edwards Personal Preference Schedule (EPPS)? Edwards supplied 360 sets of responses from 1509 cases used in original normative sample. 15 factor analyses were made. "The results of these analyses revealed an unexpectedly large descrepancy between what the PPS is designed to measure and the actual item factorial content… . In the opinion of the authors the failue of the EPPS to give the expected factor results stems from: (a) using the same item statement in several different items, (b) scoring the same item on two scales, and (c) using the forced-choice item form with equated social desirability of the item statements." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A reexamination of the factor structure of the Center for Epidemiologic Studies Depression Scale: Is a one-factor model plausible?

The Center for Epidemiologic Studies Depression Scale (CES–D) is one of the most widely used measures of depressive symptoms in research today. The original psychometric work in support of the CES–D (Radloff, 1977) described a 4-factor model underlying the 20 items on the scale. Despite a long history of evidence supporting this structure, researchers routinely report single-number summaries from the CES–D. The research described in this article examines the plausibility of 1-factor model using an initial sample of 595 subjects and a cross-validation sample of 661. After comparing a series of models found in the literature or suggested by analyses, we determined that the good fit of the 4-factor model is mostly due to its ability to model excess covariance associated with the 4 reverse-scored items. A 2-factor model that included a general depression factor and a positive wording method factor loading only on those 4 items had fit that was nearly as good as the original 4-factor model. We conclude that although a 1-factor model may not be the best model for the full 20-item CES–D, it is at least plausible. If a unidimensional set of items is required (e.g., for a unidimensional item response theory analysis), by dropping 5 items, we were able to find a 1-factor model that had very similar fit to the 4-factor model with the original 20 items. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Do neuropsychological tests have the same meaning in Spanish speakers as they do in English speakers?

Objective: The purpose of this study was to examine whether neuropsychological tests translated into Spanish measure the same cognitive constructs as the original English versions. Method: Older adult participants (N = 2,664), who did not exhibit dementia from the Washington Heights Inwood Columbia Aging Project (WHICAP), a community-based cohort from northern Manhattan, were evaluated with a comprehensive neuropsychological battery. The study cohort includes both English (n = 1,800) and Spanish speakers (n = 864) evaluated in their language of preference. Invariance analyses were conducted across language groups on a structural equation model comprising four neuropsychological factors (memory, language, visual-spatial ability, and processing speed). Results: The results of the analyses indicated that the four-factor model exhibited partial measurement invariance, demonstrated by invariant factor structure and factor loadings but nonequivalent observed score intercepts. Conclusion: The finding of invariant factor structure and factor loadings provides empirical evidence to support the implicit assumption that scores on neuropsychological tests are measuring equivalent psychological traits across these two language groups. At the structural level, the model exhibited invariant factor variances and covariances. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Is It Prudent to Administer All Items for Each Child Behavior Checklist Cross-Informant Syndrome? Evaluating the Psychometric Properties of the Youth Self-Report Dimensions With Confirmatory Factor Analysis and Item Response Theory.

Through surveying of children in 10 nations with parent, teacher, and Youth Self-Report (YSR) forms of the Child Behavior Checklist (CBCL), cross-informant syndromes (CISs) were derived and cross-validated by sample-dependent methodology. Generalizing CBCL syndromes and norms to nations excluded from its normative sample is problematic. This study used confirmatory factor analyses (CFAs) to test factor model fit for CISs on the YSR responses of 625 Jamaican children ages 11 to 18 years. Item response theory (IRT), a sample independent methodology, was used to estimate the psychometric properties of individual items on each dimension. CFAs indicated poor to moderate model-to-data fit. Across all syndromes, IRT analyses revealed that more than 3/4 of the cross-informant items yielded little information. Eliminating such items could be cost effective in terms of administration time yet improve the measure's discrimination across syndrome severity levels. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Using the Megargee MMPI-based classification system with MMPI-2s of male prison inmates.

To determine the impact of the Minnesota Multiphasic Personality Inventory—2 (MMPI-2; J. N. Butcher, W. G. Dahlstrom, J. R. Graham, A. Tellegen, & B. Kaemmer, 1989) on the Megargee offender classification system, 1,213 male offenders' responses on the original Minnesota Multiphasic Personality Inventory (MMPI) were rescored and reprofiled as MMPI-2s, and the MMPI-2s of 422 male prisoners were used to estimate their original MMPIs. When classifications based on the original MMPIs were compared with those from MMPI-2s, less than two thirds were classified identically. Therefore, the original Megargee rules should not be used to classify MMPI-2s. A new set of classifactory rules was devised for the MMPI-2 which, on cross-validation, agreed with the original MMPI classifications in 82% of the cases. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Positive and negative global self-esteem: A substantively meaningful distinction or artifactors?

Global self-esteem based on M. Rosenberg's ( 1965) scale is typically treated as a unidimensional scale. However, factor analyses suggest separate factors associated with positively and negatively worded items, and there is an ongoing debate about the substantive meaningfulness of this distinction. Confirmatory factor analysis ( CFA ) was used to evaluate alternative 1- and 2-factor models and to test hypotheses about how the factors vary with reading ability and age. Responses based on the National Longitudinal Study of 1988 (S. J. Ingles et al, 1992) reflected a relatively unidimensional factor and method effects associated with negatively worded items. Such effects are common in rating scale responses, and this CFA approach may be useful in evaluating whether factors associated with positively and negatively worded items are substantively meaningful or artifactors. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Two forms of minority-group test bias as psychometric artifacts with an animal model (Rattus norvegicus).

Controversy abounds over attributing group differences on tests to nature, nurture, or test bias. Limitations of correlational sampling from natural populations necessitate experimental methods to resolve underlying issues. In classicial psychometrics test items are selected from a larger item pool through analysis of item responses in a sample of subjects. Rats of six inbred strains (n?=?366) were tested in multiple mazes to provide a large item pool. Six populations were created, each with differing proportions of each strain. Items selected through independent item analyses within each population yielded six tests. An independent cross-validation sample (n?=?146) provided scores on all six items. This sample was also tested in another set of maze problems defined as the criterion to be predicted. Strain means and intrastrain predictive validities for the six tests varied with strain representation in the population used for item selection (p?     

Reading Motivation: Multidimensional and Indeterminate.

The Motivations for Reading Questionnaire (MRQ) contains 54 items that are posited to tap 11 dimensions of reading motivation. The structural validity of the MRQ was investigated with 2 samples: (a) 328 students in Grades 3-5 from 2 suburban mid-Atlantic elementary schools and (b) 735 students in Grades 3-5 in 2 suburban southwestern elementary schools. With confirmatory factor analyses (CFAs), the theoretical 11-factor structure did not adequately fit the data in either sample. Subsequently, exploratory factor analyses found 8 factors for each sample with 6 factors defined by only 3 or 4 common items. However, a double CFA cross-validation found an inadequate fit for both samples. Given these results, the authors suggest that the MRQ be revised. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The composition and structure of depressive symptomatology in renal disease.

Renal disease is an increasingly common illness among middle-aged and older adults, and is often associated with depression. The Center for Epidemiologic Studies Depression Scale (CES–D) is a widely used self-report screening measure on which responses generally conform to a 4-factor structure, with each factor loading onto a higher-order Depression factor. The current study examines whether this structure is supported among individuals with renal disease (both predialysis and posttransplant kidney disease patients). Persons with renal disease (n = 225) and healthy control participants (n = 230) were recruited from Vancouver General Hospital and the community. Participants completed the CES–D as part of an extended assessment. Confirmatory factor analytic models were computed for the renal disease and healthy control groups. Results support a 4-factor structure for CES–D responses in persons with renal disease and healthy controls. The hierarchical structure of CES–D responses also appears invariant between groups. Factor structure was similar between groups; only for Depressive Affect was the strength of association with the second-order factor greater among the renal disease than healthy control participants. Findings support similarity of CES–D factor structures between patients with renal disease and healthy individuals. Results suggest that the CES–D's 4 factors contribute to measurement of a higher-order Depression factor in both groups; furthermore, 3 of 4 factors appear invariant between groups. As such, this measure can be used confidently to quantify depressive symptoms in individuals with renal disease. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Structure of the Weinberger Adjustment Inventory Self-Restraint Scale and its relation to problem behaviors in adolescence.

The authors examined the structure of the Weinberger Adjustment Inventory (WAI) Self-Restraint scale in derivation (n?=?1,286) and cross-validation (n?=?1,154) samples of mostly African American 6th-graders in 3 urban schools. Four models were compared: (1) a 1-factor model; (2) a hierarchical model in which factors representing Impulse Control, Suppression of Aggression, Responsibility, and Consideration of Others were subsumed by a higher order factor; (3) a model that represented these 4 factors as correlated but distinct constructs; and (4) a model that excluded Consideration of Others from the higher order factor. Consistent support was found for the last model based on confirmatory factor analyses and latent-variable analyses examining the relations among self-restraint scales, drug use, delinquency, and aggression. These findings have implications for using the WAI, particularly in studies of adolescent problem behaviors. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structures of the original and brief Fear of Negative Evaluation (FNE and BFNE) scales: Correction to an erroneous footnote.

In recent factor analyses of the original and brief Fear of Negative Evaluation (FNE and BFNE) scales, a 2-factor model was supported, with straightforwardly worded and reverse-worded items loading on separate factors (see record 2004-15398-007). However, a theoretically meaningful alternative model was not fitted previously because the authors mistakenly assumed that unique variances among categorical manifest variables could not be correlated using the weighted least squares mean and variance adjusted estimation procedure implemented in the Mplus program (Muthen & Muthen, 1998-2004). The present article corrects the erroneous footnote in our previous report and briefly mentions results of fitting 1-factor models with correlated unique variances to the same data analyzed previously. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Systematically derived work dimensions: Factor analyses of the Occupation Analysis Inventory.

Developed the Occupation Analysis Inventory (OAI) for use as a taxonomic tool, derived a broad set of human work dimensions (factors) based on that questionnaire, and established some degree of construct validity for the resultant dimensions. 602 OAI work elements (items) were subjected to several factor analyses based on the ratings of 1,414 jobs on the elements and ratings of the elements on their requirements for 102 defined human attributes. The resultant factors were significantly related to the tested abilities of relevant job holders. It is concluded that job-rating factors should be (a) fairly reflective of the various types of work activities and conditions extant in the world of work and (b) unique in their coverage of information relevant to occupational education and career decision making. (79 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Evaluation of the original Ontario Child Health Study scales

This article presents evaluative information on the use of the original Ontario Child Health Study scales to serve as original-level measures of conduct disorder, hyperactivity and emotional disorder among children in the general (non clinic) population. Problem checklist assessments were obtained from parents and teachers of children aged six to 16 and youth aged 12 to 16 drawn from a general population (n = 1,751); and a mental health clinic sample (n = 1,027) in the same industrialized, urban setting. The results showed that the original OCHS scales possess adequate psychometric properties to be used as original-level measures of disorder. Correlations between individual items and their hypothesized scales were very strong, indicating convergent validity, while correlations between the same items and other (non hypothesized) scales were lower, indicating discriminant validity. Item analyses indicated that individual scale items possess both convergent and discriminant validity. Although the scales were skewed to the positive end of the continuum, they demonstrated good internal consistency (all estimates > or = 0.74) and test-retest (all estimates > or = 0.65) reliability. Finally, three different validity analyses confirmed hypotheses about how the original OCHS scales should perform if they provide useful measures of disorder.     

Measurement invariance of the Gifted Rating Scales—School Form across five cultural groups.

This study examined measurement invariance of the Gifted Rating Scales—School Form (GRS–S) across the United States, Puerto Rico, China, South Korea, and Turkey, using multigroup confirmatory factor analysis. A total of 1,817 students were rated by 287 teachers using either translated versions of GRS–S or the original English GRS–S. Results indicate a similar factor structure for the GRS–S across the five locations; six factors with each of the 72 items equivalently loaded to the same latent variable across groups. The metric invariance test and the factor variance and covariance invariance tests reveal that the patterns of factor loadings and the factor variances and covariances are invariant across the five groups. Moreover, the scalar invariance test indicates that item means are equivalent across the groups. These results suggest that the GRS–S has intercultural utility and can be similarly interpreted. Implications and limitations of the present research for gifted identification are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A factor analysis of experimental social desirability and response set scales.

Intercorrelations between 19 response set scales, based upon the scores of 110 students, were factor analyzed and the factors rotated orthogonally. Edwards' SD scale and 6 experimental social desirability scales had their highest loadings on the 1st factor. 3 scales containing neutral items in which the probability of a True response to the items varied between scales were found to have their highest loadings on 3 different factors. Scales designed to measure the tendency to give deviant True responses to items with socially undesirable scale values, to items with socially desirable scale values, and to items with neural scale values were found to have their highest loadings on 3 different factors. The tendency to give deviant True responses to items with socially undesirable scale values was found to be related to the tendency to give deviant False responses to items with socially desirable scale values. The tendency to mark items as doubtful and the tendency to answer items marked doubtful as True were identified as 2 factors unrelated to social desirability tendencies. (31 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Are special norms for minorities needed? Development of an MMPI F scale for Blacks.

As part of a large-scale investigation of the need for special MMPI norms for adult Black American test Ss, 882 normal 18–65 yr old Blacks from Alabama, Michigan, and North Carolina were tested. Item analyses of these protocols revealed 33 items that met the 10% or less endorsement criterion used to develop the MMPI F scale. Comparisons were made between White and Black endorsements of the items on this new scale, items of the standard F scale, and additional items in the MMPI pool that met the 10% criterion but were not included in the original F scale. White endorsement patterns agreed with the Blacks' F scale, but Black endorsement patterns agreed with only one third of the standard F-scale items. Further, Blacks showed comparable levels of infrequency on only 6 of the 38 supplementary F items. Although these results do not necessarily indicate that special norms for the clinical scales are necessary, the amount of difference between responses of Blacks and Whites to rarely endorsed items suggests that for Blacks, this new scale may be a more accurate measure of correlates associated with endorsement of deviant items than the standard F scale. (19 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The cardinal features of cognitive and noncognitive dysfunction and the differential efficacy of tacrine in Alzheimer's disease patients

The Alzheimer's Disease Assessment Scale (ADAS), frequently used in clinical trials to assess overall pathology of Alzheimer's disease (AD), comprises two subscales. The cognitive subscale (ADAS-COG) consists of 11 items, and the noncognitive subscale consists of 9 items. Factor analyses were carried out on ADAS-COG and ADAS-NONCOG item scores from the most recent and largest (n = 663) placebo-controlled, multicenter, 30-week study (970-61) of tacrine in patients with AD conducted by the clinical research group at Parke-Davis Pharmaceutical Research. Through factor analyses the primary dimensions of variation in the ADAS-COG and ADAS-NONCOG were defined. Obliquely rotated three principal factors of ADAS-COG and three principal factors of ADAS-NONCOG have been interpreted as three cardinal features of cognitive function corresponding to memory, language, and praxis, and three cardinal features of noncognitive function corresponding to agitation, depression, and lack of concentration. Reliably defined factors of ADAS-COG enabled comparisons of longitudinal changes in cognitive dysfunction. Factor scores at week 30, adjusted to baseline factor scores, were used to compare the effects of tacrine with those of placebo on cognitive cardinal features. Additionally, the effect of concurrent depression on cardinal features of cognitive dysfunction was evaluated by gender.     

Unitary alcoholism factor and its personality correlates.

Administered a 135-item alcohol behavior questionnaire and the short-form MMPI to 160 new admissions to a state hospital. The alcohol questionnaire was constructed with substantial item redundancy so that specific factors could be defined. Factor analysis of the 135 items revealed 1 large alcoholism factor plus 10 small specific factors that were not essential features of alcoholism. A principal components analysis of the 42 items most highly related to the single major alcoholism factor povided scoring weights. Item analysis techniques were used to identify the MMPI items which were most discriminating between Ss with high and low alcohol-abuse scores. 49 MMPI items were found to relate highly to the alcohol-abuse factor scores. The 49 items were found to contribute most to scoring of HS, D, Hy, and Pt clinical scales, while the contribution to Pd was substantially lower. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structure of the MMPI-168 in a psychiatric population.

Administered the 1st 168 items of the MMPI to 505 psychiatric patients. Factor analysis of the resulting data identified factors of Somatization, Depression, Psychotic Distortion, Low Morale, and Acting Out, plus a separate Masculinity-Femininity factor. Results compare favorably with those from the technically best of the item-level analyses of the complete MMPI that have been reported by others. Factor scoring keys that can be used with an abbreviated short-form administration are presented. (PsycINFO Database Record (c) 2010 APA, all rights reserved)