The factor structure and factorial invariance of the 12-item General Health Questionnaire (GHQ-12) across time: Evidence from two community-based samples.

This study provides new knowledge about the factor structure of the 12-item General Health Questionnaire (GHQ-12; D. Goldberg, 1972) through the application of confirmatory factor analysis to longitudinal data, thereby enabling investigation of the factor structure, its invariance across time, and the rank-order stability of the factors. Two community-based longitudinal adult samples with 1-year (n = 640) and 6-year (n = 330) follow-up times were studied. As a result, the correlated 3-factor model (i.e., Anxiety/Depression, Social Dysfunction, and Loss of Confidence) showed a better fit with both samples than the alternative models. The correlated 3-factor structure was also relatively invariant across time in both samples, indicating that the scale has good construct validity. The rank-order stabilities of the factors were low across time, which suggests that the GHQ-12 measures temporal mental state. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Validation of the General Health Questionnaire-12: Using a sample of employees from England's health care services.

Psychiatric interviews were carried out to validate the General Health Questionnaire-12 (GHQ-12; Goldberg, 1972) for use with staff of England's National Health Service (NHS), and to determine the appropriate threshold score to identify probable cases. In a sample of 551 NHS staff, the correlation between the GHQ-12 and the Clinical Interview Schedule-Revised (CIS-R) was found to be .70. The receiver operating characteristic showed that a 3/4 threshold, higher than used in all but one previous study, gave the best conservative estimate of minor psychiatric morbidity. This threshold gave an estimated sensitivity of the GHQ-12 of .69 and specificity of .88. It reduced case rates by between 8% and 17% as compared with lower possible thresholds. Convergent and discriminant validation of the GHQ-12 was demonstrated through comparison with other measures of mental and physical health. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Revisiting the Factor Structure of the Questionnaire on Smoking Urges.

This study assessed the factor structure of the Questionnaire on Smoking Urges (QSU), a commonly used assessment of cravings for cigarettes, with a sample of smokers presenting for treatment in a smoking cessation trial. On the basis of previous research, three confirmatory factor analytic models were tested. Model 1 hypothesized a 26-item, 2-factor model using the items reported in the original QSU analysis by S. T. Tiffany and D. J. Drobes (1991). Model 2 hypothesized a 12-item, 2-factor model comprised of the 6 most robust items found in each of the 2 factors of the original factor analysis. Using the 12 items from Model 2, Model 3 hypothesized a 12-item, 1-factor model. The 2nd model was found to fit the data best. Reliability was also tested using values obtained in this 2nd model, and these estimates were found to be reasonably good. Future research directions for the QSU are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Topical topic. Use of chemotherapy for retinoblastoma

We studied the significance of debt as a risk factor for suicidal ideation and suicide attempts in a nation-wide sample (n = 4868) of the Finnish general population. Those experiencing difficulties in repaying their debts more often than others (37 vs. 16%, P<0.001) had a probable minor mental disorder (GHQ-12 score > or =3). Nevertheless, difficulties in repaying debts were found to be a factor independently associated with suicidal ideation (OR, 2.8; 95% CI, 1.9-4.2). The clinical implication of these findings is that individuals experiencing difficulties in repaying their debts may require psychiatric evaluation as well as socio-economic counselling.     

Confirmatory Factor Analysis of the Psychopathy Checklist: Screening Version in Offenders With Axis I Disorders.

One hundred forty-nine inpatients within a maximum security psychiatric facility were assessed with the Psychopathy Checklist: Screening Version (PCL:SV; S. D. Hart, D. N. Cox, & R. D. Hare, 1995). Within the total sample, 68% had a psychotic disorder and 30% met criteria for psychopathy. Using confirmatory factor analysis, the authors tested the 2-factor PCL:SV model of psychopathy and recent 3- and 4-factor models. Results indicated good fit for each model, with the 4-factor model showing best overall fit. Structural equation modeling was used to determine which psychopathy factors predicted 6-month follow-up of inpatient aggression. The 2-, 3-, and 4-factor models, respectively, accounted for 16%.27%. and 3l% of the variance in aggression. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A comparison of models of drinking motives in a university sample.

Previous research has suggested the presence of several alternative factorial models for the assessment of drinking motives. In the present study, confirmatory factor analysis was used to assess the factor structure of the Drinking Motives Questionnaire (M. L. Cooper, 1994) in a college sample. The results indicate that a 4-factor model that includes the dimensions of social rewards, affect enhancement, coping, and conformity motives fits the data significantly better than 2- and 3-factor models. Furthermore, the 4-factor model fits equally well for men and women. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structure of the Mayo Older Americans Normative Sample (MOANS) core battery: Replication in a clinical sample.

G. E. Smith et al (1992) confirmed 5 factors to account for the covariance structure of a "core battery" of the Wechsler Adult Intelligence Scale—Revised (WAIS—R), Wechsler Memory Scale—Revised (WMS—R), and Auditory Verbal Learning Test administered to the MOANS. The authors attempted to replicate this model in a clinical sample. Data from 417 clinical cases were used in LISREL confirmatory factor analyses. Modifications of the original secondary variable assignments were necessary to obtain a 5-factor model with adequate fit. This 5-factor model was superior to alternative models. The results support the use of a 5-factor model in the development of factor-based summary indexes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

WAIS-R Factor Structure in Alzheimer's Disease Patients: A Comparison of Alternative Models and an Assessment of Their Generalizability.

The WAIS-R is often used in neuropsychological evaluations of individuals with probable Alzheimer's disease (AD), but its factor structure in this population is unknown. Moreover, theories and past research findings make competing predictions concerning its structure. Using confirmatory factor analysis, the authors compared 5 alternative WAIS-R factor models among 516 AD patients: 1-factor (Spearman's g) and 2-factor (Verbal IQ and Performance IQ) models; a 3-factor model including Verbal Comprehension (VC), Perceptual Organization (PO), and Freedom From Distractibility (FD) factors; a 3-factor model in which Digit Symbol loads on PO rather than FD; and a 3-factor model in which Digit Symbol loads on both PO and FD. Results favored the 3-factor model in which Digit Symbol loads on PO rather than FD. Moreover, this model fit the data best among subsamples of patients defined by age, dementia severity, years of education, and gender. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structure of the Working Alliance Inventory.

The Working Alliance Inventory (WAI; A. O. Horvath and L. S. Greenberg [see PA, Vol 76:24600]) was completed after the 1st psychotherapy session by 84 university counseling center clients and 15 therapists rating their work with 123 clients. The factor structure of these responses was examined using confirmatory factor analysis. A model with 1 general factor, a model with 3 specific factors, and a bilevel model of the factor structure were examined. The bilevel factor structure, with a General Alliance factor as its primary factor and 3 secondary specific factors, fit the data best. The items most indicative of the 3 specific factors were selected to form a 12-item short form of the WAI. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structure of the Brief Symptom Inventory--18 in adult survivors of childhood cancer: Results from the Childhood Cancer Survivor Study.

The factor structure of the Brief Symptom Inventory-18 (BSI-18; L. R. Derogatis, 2000) was investigated in a sample of adult survivors of childhood cancer enrolled in the Childhood Cancer Survivor Study (CCSS; N = 8,945). An exploratory factor analysis with a randomly chosen subsample supported a 3-factor structure closely corresponding to the 3 BSI-18 subscales: Depression, Anxiety, and Somatization. Confirmatory factor analysis with structural equation modeling validated this 3-dimensional structure in a separate subsample, though an alternative 4-factor model also fit the data. Analysis of the 3-factor model showed consistent fit in male and female participants. Compared with available community-based norms, survivors reported fewer symptoms of psychological distress. Together, results support the hypothesized 3-dimensional structure of the BSI-18 and indicate the measure may be useful in assessing psychological distress in this growing population of cancer survivors. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor analysis using delta-rule wake-sleep learning

We describe a linear network that models correlations between real-valued visible variables using one or more real-valued hidden variables-a factor analysis model. This model can be seen as a linear version of the Helmholtz machine, and its parameters can be learned using the wake-sleep method, in which learning of the primary generative model is assisted by a recognition model, whose role is to fill in the values of hidden variables based on the values of visible variables. The generative and recognition models are jointly learned in wake and sleep phases, using just the delta rule. This learning procedure is comparable in simplicity to Hebbian learning, which produces a somewhat different representation of correlations in terms of principal components. We argue that the simplicity of wake-sleep learning makes factor analysis a plausible alternative to Hebbian learning as a model of activity-dependent cortical plasticity.     

Latent structure of the Postconcussion Syndrome Questionnaire.

The underlying structure of the Postconcussion Syndrome Questionnaire (PCS) was evaluated in a large sample of medical and psychiatric patients. Three potentially viable models were generated using exploratory factor analysis with half of the sample. The other half evaluated the 3-, 4-, and 5-factor models using confirmatory factor analytic procedures. The factor analyses generated compelling data for a 5-factor model for the PCS questionnaire. However, internal consistency for each of the factors argued in favor of the 3-factor model. Balancing internal consistency, confirmatory factor analyses, and parsimony resulted in endorsement of a 4-factor solution for the PCS questionnaire for this sample. The factors are best described as clusters of psychological, somatic, cognitive, and infrequent complaints. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Patterns of stability in adult attachment: An empirical test of two models of continuity and change.

One of the core assumptions of attachment theory is that attachment representations are stable over time. Unfortunately, the data on attachment stability have been ambiguous, and as a result, alternative theoretical perspectives have evolved to explain them. The objective of the present research was to evaluate alternative models of stability by studying adults in 2 intensive longitudinal investigations. Specifically, we assessed attachment representations in 1 sample (N = 203) daily over a 30-day period and in the other sample (N = 388) weekly over a year. Analyses show that the patterns of stability that exist in adult attachment are most consistent with a prototype model—a model assuming that there is a stable factor underlying temporary variations in attachment. Moreover, although the Big Five personality traits exhibited a pattern of stability that was similar to that of attachment, they did not account for the stability observed in attachment. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Confirmatory Factor Analysis of the California Verbal Learning Test in Patients With Epilepsy: Relationship to Clinical and Neuropathological Markers of Temporal Lobe Epilepsy.

Latent constructs involved in California Verbal Learning Test (D. C. Delis, J. H. Kramer, E. Kaplan, & B. A. Ober, 1987) performance were examined using confirmatory factor analysis in 388 epilepsy surgery candidates. Eight factor models were compared. A single-factor model was examined, along with 7 models accommodating constructs of auditory attention, inaccurate recall, and delayed recall in different combinations. The retained model consisted of 3 correlated factors: Auditory Attention. Verbal Learning, and Inaccurate Recall. Validity of this factor structure was examined in a subsample of patients with left and right temporal lobe epilepsy. All 3 factors were related to seizure focus and magnetic resonance imaging hippocampal volume. Only Verbal Learning was related to hippocampal neuropathology, supporting the distinction between learning and attention in the factor structure. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Detecting psychiatric disorders in medical practice using the General Health Questionnaire. Why do cut-off scores vary?

In this study we assessed the accuracy of the General Health Questionnaire in detecting psychiatric disorders in general medical out-patients. A total of 290 newly referred patients were interviewed with the Present State Examination. Prior to the interview, 112 patients completed the full GHQ-60, 100 completed the GHQ-30 and 78 completed the GHQ-12. Data from the first group were used to study the full GHQ-60, together with the GHQ-30 and and GHQ-12, when disembedded from the full questionnaire. In a comparison between the disembedded and the separate versions of the GHQ-30 and GHQ-12 we observed considerable variation in the cut-off scores where a certain sensitivity and specificity was attained. In ROC-analysis, the versions were not materially different in their discriminatory capacity (area under the curve). The use of different criteria to define a 'case' demonstrated that case severity was another source of increasing cut-off scores. Our data demonstrate that the use of disembedded or separate versions of the questionnaire, together with variation in the case criteria can be a major explanation for variation in cut-off scores that was observed in previous studies.     

A confirmatory factor analysis of the Impact of Event Scale using a sample of World War II and Korean War Veterans.

This study assessed the factor structure of the Impact of Event Scale (IES), a measure of intrusion and avoidance, using a sample of World War II and Korean War veterans who had experienced combat 40–50 years earlier. A series of 3 confirmatory factor analytic models were specified and estimated using LISREL 8.3. Model 1 specified a 1-factor model. Model 2 specified a correlated 2-factor model. Model 3 specfied a 2-factor model with additional cross-factor loadings for Items 2 and 12. Model 3 was found to fit the data. In addition, this model was found to be a better explanation of the data than the other models. Also in addition, the correlations between the Intrusion and Avoidance factors and the 4 subscales of the 28-item General Health Questionnaire were examined to determine the distinctiveness of the two IES factors. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Investigating Population Heterogeneity With Factor Mixture Models.

Sources of population heterogeneity may or may not be observed. If the sources of heterogeneity are observed (e.g., gender), the sample can be split into groups and the data analyzed with methods for multiple groups. If the sources of population heterogeneity are unobserved, the data can be analyzed with latent class models. Factor mixture models are a combination of latent class and common factor models and can be used to explore unobserved population heterogeneity. Observed sources of heterogeneity can be included as covariates. The different ways to incorporate covariates correspond to different conceptual interpretations. These are discussed in detail. Characteristics of factor mixture modeling are described in comparison to other methods designed for data stemming from heterogeneous populations. A step-by-step analysis of a subset of data from the Longitudinal Survey of American Youth illustrates how factor mixture models can be applied in an exploratory fashion to data collected at a single time point. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor structure of the Counselor Rating Form—Short.

The relative goodness of fit of four competing factor models of the Counselor Rating Form—Short was examined across client (N?=?191) and nonclient (N?=?111) samples by confirmatory factor analysis. The four models tested were (a) the single, general, positive-evaluation factor; (b) the three orthogonal factors (expertness, trustworthiness, and attractiveness); (c) the three oblique factors; and (d) a two-step hierarchical-factor model consisting of the three independent first-order factors and an independent second-order general factor. Results of the confirmatory factor analyses supported the validity of the two-step hierarchical-factor model for both the client and nonclient samples. The factor structure of this two-step hierarchical-factor model was found to be invariant across the two samples. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Confirmatory factor analysis of the WAIS–III standardization data.

Confirmatory factor analyses with the standardization data of the Wechsler Adult Intelligence Scale–Third Edition (Wechsler, 1997a) compared 6 models with 1 to 4 factors for 11- and 13-subtest versions of the test. Three factors usually fit the data better than 2 factors, but 2-factor models were more parsimonious. A 2-factor model with a Verbal Comprehension factor (Vocabulary, Similarities, Information, and Comprehension) was as good as and sometimes better than the 2-factor model defined by the traditional separation of Verbal and Performance subtests. For 3-factor models, alternative specifications of processing speed subtests on either the Perceptual Organization or Freedom From Distractibility factor were comparable, and specifying a 4th factor for Digit Symbol and Symbol Search had little advantage in comparison with 3-factor models with correlated errors for the 2 subtests. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Factor models for assessing the relation between creativity and intelligence.

Demonstrated that the usual 1st-order factor model is inappropriate for analyzing the factor structure of creativity and intelligence tests. An alternative model that allows for the estimation of unique covariances between the fluency and originality scores is proposed. Seven data sets were used to compare the fit of this model to the alternative model, whereby the unique covariances were constrained to be zero. Findings support the model with unconstrained unique covariances; when this model is used, creativity and intelligence are more readily recognized as separate dimensions. (25 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)