Fiji norms and a cross-cultural clinical validation of the General Health Questionnaire (GHQ 20).

Norms are presented for the 20-item version of the General Health Questionnaire (GHQ 20) that were obtained from a non-clinical group of the Fiji population. The data was required to give meaning to the results that had been obtained from a group of former hostages and the families of hostages that were still in custody. With the omission of two items, the outcome endorsed the robust psychometric properties of the instrument in terms of reliability and factor structure, and supported its use with quite a different cultural population from that for which originally it was designed. A validation study also showed expected differences between the normative group and the hostage group, with members of the hostage group classified after clinical interviews showing the most significant stress and receiving the highest scores on all four subscales as well as the full scale.

The psychometric cross-cultural validation of the Impact of Event Scale.

Data is presented from a sample drawn from the general Fiji population to standardise the Impact of Events Scale (IES) and to prepare percentile tables from which validating comparisons could be made with the scores of two target groups from a previous Fiji study. Detailed statistical analyses gave strong support for a general factor and marginal support for the two specified subscale/factors of the measure. Comparisons of group mean scores, and others using the percentile tables, gave confidence that the full IES rating scale was appropriate for the purpose for which it was used, and indicate that clinicians in Fiji might use the IES confidently to validate the judgments of traumatic stress they make from their interviews--presently they have no such psychometric tool available.

A psychometric evaluation and cross-cultural validation of the 21 item Hopkins Symptom Checklist.

Data relating to the standardisation of a short version of the 21 item Hopkins Symptom Checklist (HSCL 21) is presented. It confirms the presence of three factors, and provides group means and percentile scales that enable comparisons to be made between the scores of a non-clinical sample and those of two separate target groups that had been under stress. These comparisons give confidence that the rating scale is appropriate for the purpose for which it was used, and suggest that the HSCL 21 could be a useful tool for clinicians in Fiji to use as an adjunct to standard interviews in the assessment of the effects of trauma.

Styles of psychological adjustment in diabetes: a focus on key psychometric issues and the ATT39.

While knowledge of the biomedical factors in diabetes has grown in a steady and systematic fashion over the past 60 years, attempts to define, measure, and understand the relevant psychological constructs have only begun recently. In particular, there is need for psychometrically sound tests to tap these dimensions. This study examined the psychometric characteristics of the ATT39, a promising measure of psychological adjustment to diabetes. The results, based on three patient samples and using the FACTOREP factor-matching procedure, suggested that the ATT39 has a large single factor only. This new subscale appears clinically to measure the integration of diabetes and its treatment into the lifestyle and personality of a patient with diabetes.

Attitudes to AIDS: a comparative analysis of a new and negative stereotype.

Two propositions about attitudes, which have previously been supported with respect to the mentally ill, were examined with respect to AIDS patients. The first, that people attach a stigma to the AIDS patient, was strongly supported, and two quite independent components of the stigma were found. One of these components identified as dependence, was closely related to the attributes of typical cancer patients and coronary heart patients, while the other, identified as low moral worth, clearly distinguished the AIDS patient from the other two groups of patients. The second proposition, that attitudes to AIDS are not strongly related to age, sex and occupational background, was largely supported. However there was some evidence that males rated AIDS patients lower on moral worth than did females.

Community attitudes to mental illness in New Zealand twenty-two years on.

Three propositions about attitudes to mental illness derived from Nunnally (Popular Conceptions of Mental Health. Holt, Rinehart & Winston, New York, 1961) were examined with the semantic differential technique as it was used originally by Olmsted and Ordway (Final Report to National Institutes of Mental Health, 1963). Attitudes were compared between several studies using the same measures, that ranged over 22 years from 1962 to 1984. The results were remarkably consistent across all studies, indicating that the community had persistently negative attitudes towards the mentally ill and was no more likely today to want to play a major role in the care of the mentally ill than was the case more than 20 years ago. As a consequence it appears that there will need to be substantial and permanent attitude change, of a kind fleetingly observed over the period of the study, before the professional care of the mentally ill in the community may be expected to have maximum impact.

Comparative perceptions of driver ability--a confirmation and expansion.

This study aimed to investigate earlier Swedish and United States research which has shown that drivers tend to rate themselves as more skillful and less risky than the average driver, and to extend the range of driver characteristics on which such ratings are made. A total of 178 male and female drivers rated "me as a driver," "an average driver" and "a very good driver" on the following eight relevant bipolar semantic differential scales: foolish-wise, unpredictable-predictable, unreliable-reliable, inconsiderate-considerate, dangerous-safe, tense-relaxed, worthless-valuable, irresponsible-responsible. The results confirmed expectations that a substantial majority of drivers, up to 80%, would rate themselves above average on a number of important characteristics, but also showed that they rated themselves below "a very good driver." The ratings did not vary significantly across demographic categories. Implications for road safety are briefly discussed.

Investigation of 'weight pathology' of 58 mothers of anorexia nervosa patients and 204 mothers of schoolgirls.

'Weight pathology'--defined by Kalucy et al. (1977) as 'deviations in weight, shape, eating behaviour and activity'--has been hypothesized to be unduly common in families with anorexia nervosa. It was investigated in this study by means of questionnaires evaluating both weight history and attitudes towards weight-related matters in 58 mothers of anorexia nervosa patients and 204 mothers of schoolgirls of similar age and socioeconomic status (SES). No support was found for the hypothesis. A family history of aberrant weight and mother's current weight and past weight histories showed no significant differences between the groups. 142 other mothers of schoolchildren and 446 attenders at Weight Watchers also completed the questionnaire on attitudes towards weight-related matters. Mothers of patients showed a lower concern on all scales than did all other groups. The findings are discussed in relation to earlier controlled studies of family factors in anorexia nervosa.

Multiple Replication of Factor Structure: A Logical Solution for a Number of Factors Problem.

A factor replication procedure which provides a solution for the numbers of factors problem for multiple scale questionnaire responses, is described and demonstrated. In the procedure a series of factor analyses are conducted in which a reduced number of factors is used for each successive rotation until a structure is found which is replicated across subject groups. The factor matching s index, which relates loadings as they fall about a series of prescribed levels, is used to compare factors in order to show the level of similarity achieved. The procedure is illustrated using the responses of three groups of subjects to the Hopkins Symptom Checklist. Results show that the five factor structure claimed by the authors of the measure is not supported by the data but that there is clear evidence of three replicable factors. It is suggested that the procedure provides the basis for a significant improvement in decisions related to the number of factors problem.

Simple Versus Complex Factor Analyses of Responses to Multiple Scale Questionnaires.

Some effects of using inappropriate criteria for sufficiency of factors are discussed, and examples from the literature used to show how procedures leading to the rotation of large numbers of factors may result in fragmentation and difficulty in interpretation. Analyses of two psychometrically equivalent matrices containing an imposed scale structure led to dissimilar solutions, neither of them revealing the imposed structure, when a minimum eigenvalue of 1.00 was used as a criterion for sufficiency of factors. More conservative (two and three-factor) solutions revealed the imposed scale structure in both matrices. Comparisons with higher-order solutions showed considerable similarity between two-factor solutions and between three-factor solutions at different levels. These contrasted with substantial dissimilarities found in equivalent four-factor solutions. Though the main purpose of the paper is to outline procedures which might reasonably reduce factor fragmentation with its consequent problems of replication and interpretability, the implications of the alternative problem of Underfactoring are also briefly discussed and some solutions suggested.

The Structure Of The Eysenck Personality Inventory: A Comparison Between Simple And More Complex Analyses Of A Multiple Scale Questionnaire.

Responses of 392 students to the Eysenck Personality Inventory were analyzed using conventional factor-analytic techniques and a nonmetric multidimensional scaling method. Rotating the first two factors gave a result clearly comparable with an earlier third-order analysis, while a three-factor rotation neatly clustered the original Extraversion, Neuroticism, and Lie scale items. A three-dimensional non-metric analysis appeared to provide no more information for users of the questionnaire than was given by a comparable two-dimensional analysis which had produced a solution closely resembling that of the two-factor rotation. The conclusion reached was that psychometrically useful information may be more readily revealed by simple and rationally restricted analyses than by exhaustive, more complex, and higher order solutions.

Internal control, powerful others, and chance: a confirmation of Levenson's factor structure.

The Internal-External control scales of Rotter and Levenson, and the Social Desirability scale of Marlowe and Crowne were completed by 71 male and 85 female students. Factor analysis clearly confirmed the three-factor structure underlying Levenson's questionnaire, while further analyses revealed close agreement with her findings on scale characteristics and interrelationships. Some new evidence for the multidimensionality of Rotter's scale was also presented.