The reliability of score differences on the WAIS,WISC-R,and WAIS-R

Analyzed data from the WAIS, WISC-R, and WAIS-R manuals to determine the reliabilities of the differences between Verbal and Performance IQs, of subtest comparisons, and of differences between subtest and IQ scores. The differences between the Verbal and Performance IQs were moderately reliable: 0.76 (WAIS and WISC-R) and 0.81 (WAIS-R). The means of the reliability coefficients of the differences between the 55 pairs of subtests were appreciably lower: 0.58 (WAIS), 0.62 (WISC-R), and 0.65 (WAIS-R). The most consistently reliable subtest differences were between Block Design and Vocabulary and the least reliable differences between Block Design and Object Assembly. The means of the reliability coefficients of the differences between subtest scores and the Full Scale IQs were 0.53 (WAIS), 0.58 (WISC-R), and 0.62 (WAIS-R). It was concluded that caution should be exercised when interpreting configurations on Wechsler protocols.     

Neurodiagnostic implications of abnormal verbal-performance IQ discrepancies on the WAIS-R: a comparison with the standardization sample.

The neurodiagnostic value of Wechsler Adult Intelligence Scale-Revised (WAIS-R) Verbal-Performance IQ (VIQ-PIQ) discrepancies was evaluated in a brain-damaged sample. The VIQ-PIQ differences of 260 neurologically impaired patients was compared to the WAIS-R standardization sample. The average VIQ-PIQ discrepancy of the patients was significantly larger than the mean IQ difference for normals. Comparison of the number of subjects whose VIQ-PIQ discrepancy equaled or exceeded 95% of the controls revealed no differences for the total sample and for individuals with low average or average intelligence. Conversely, the frequency of abnormal VIQ-PIQ differences was significantly greater among patients with Full Scale IQs 相似文献   

Verbal IQ vs. performance IQ difference scores in males and females from the WAIS-R standardization sample

Clinical research that has used the Wechsler-Bellevue and Wechsler Adult Intelligence Scales with patients who are suffering lateralized cerebral pathology had indicated distinct patterns of VIQ-PIQ discrepancies related to laterality of hemispheric involvement. Research data also have been interpreted to suggest that VIQ-PIQ discrepancy patterns differ as a function of the sex of the patient. However, this latter hypothesis finds less support in recent studies that have employed the newer Wechsler Adult Intelligence Scale-Revised (WAIS-R). This raises the possibility that, other than sex, manifest differences in VIQ-PIQ discrepancies may be more a consequence of normative characteristics in successive editions of the Wechsler scales or of other factors that surround neurologic impairment. Moreover, to date, clinical and research evidence for biologic sex as a mediating factor in VIQ-PIQ discrepancies has been interpreted in the absence of base rates for the distribution of such discrepancies, which occur in the normal population. Thus, the present study reports base rates for the magnitude, frequency, and direction of VIQ-PIQ discrepancies found for the 940 males and 940 females who comprise the WAIS-R standardization sample. No differences of consequence were noted in male vs. female VIQ-PIQ discrepancies. Implications are discussed in the light of earlier and emergent research with the Wechsler scales.     

Empirically observed and statistically derived verbal-performance IQ discrepancies for the old age WAIS-R standardization sample

A formula for the estimation of abnormal VIQ-PIQ discrepancies was evaluated against the actual distribution of Verbal-Performance difference scores in the WAIS-R old age standardization sample. Results showed that the formula is sufficiently accurate for persons 75 years and older.     

Comparison of the WAIS and the WAIS-R with Ss of high intelligence

Compared the performance of highly intelligent Mensa Ss on the WAIS and WAIS-R. Tests were administered in counterbalanced order with approximately 3 weeks between testings. No significant differences was found between initial WAIS and WAIS-R scores, in contrast to the 7-8 point differences in favor of the WAIS IQ scores, reported in the WAIS-R manual. Order of administration had the following effect: WAIS/WAIS-R Ss gained a mean of FS points upon retesting, in contrast to an average 9-point gain for WAIS-R/WAIS testing, Combined order of administration data appeared to distort comparisons of WAIS and WAIS-R with this population.     

Estimating WAIS-R IQ from the Shipley Institute of Living Scale: a replication

Zachary, Crumpton, and Spiegel (1985) introduced a linear regression and continuous norming procedure for estimating Full Scale WAIS-R IQ from the Shipley Institute of Living Scale. The present study replicated their method with 55 adult psychiatric inpatients and day hospital patients. A high correlation (r = .85), an extremely small mean difference in IQ (.8 points), and an acceptable average absolute difference (7.6 points) were found between estimated and obtained WAIS-R. Sines and Simmons tables (1959) for Shipley estimates of WAIS IQ produced a high correlation (r = .86), but large mean and average absolute differences (13.1 and 13.6 points, respectively). The study supports use of the Zachary et al. procedure for estimating WAIS-R IQ from Shipley scores in a psychiatric population.     

Estimating WAIS IQ from the Shipley Institute of Living Scale: a replication

This study represents a replication of the linear regression techniques and continuous norming used to estimate age-adjusted WAIS IQ from the Shipley Institute of Living Scale. The sample was composed of 125 male, right-handed, psychiatric patients, who represented five age decades between 20 and 70. The overall correlation coefficient between observed and estimated IQs was .79. The data suggest that the prediction formula is robust across all five age groups. The data also show the Shipley estimated WAIS Full Scale IQ to be somewhat better when compared with the continuously normed WAIS IQ than when compared with the WAIS tabled norms. Thus, the present results provide support for the utility of the continuous norming procedure for predicting WAIS IQ from the Shipley and argue for its use.     

Test-retest stability of the WAIS-R in a clinical sample

This study assessed WAIS-R test-retest stability for a sample of 21 psychiatric and neurological patients. The test-retest interval ranged from 2 weeks to 144 weeks, with a mean of 38 weeks. Subtest stability coefficients were highly significant; only Similarities, r(19) = .45, p less than .05, fell below a correlation of .74. The Verbal, Performance, and Full Scale IQs yielded highly significant stability coefficients (.79, .88, and .86). Nevertheless, the range of gain or loss for any single individual was comparatively large. On the Full Scale IQ, 18 of 21 (86%) subjects showed changes of 3 points or more, and 8 of 21 (38%) actually changed IQ classifications. Gain or loss in total score was unrelated to initial level of brightness on the Full Scale, age at initial testing, or days between examinations. Conversely, gain or loss on the Full Scale was associated strongly with years of education, r(19) = .55, p less than .01. When the sample was divided into brain-damaged and psychiatric subgroups, the difference in retest gain on the Full Scale IQ was nonsignificant.     

The relative utility of the shipley-hartford scale: Prediction of WAIS-R IQ

This study examined the utility of the Shipley-Hartford Scale in predicting WAIS-R Full Scale IQs when the test was administered with and without time limits in samples of cases obtained in a general acute care hospital (N = 156). While the correlation between predicted Wetzel IQ Score based on Paulson and Lin's formula and actual Full Scale WAIS-R IQ was nearly the same under both conditions and similar to previously reported values, systematic overestimation of WAIS-R IQ was found. Stepwise multiple regression analysis showed slightly more accurate prediction with the speed than the power administration. However, underestimation of above average Full Scale IQs and overestimation of below average Full Scale IQs also was found. Advanced age was associated with low conceptual quotients, which suggests that normal older persons may be identified incorrectly as cognitively impaired. These findings point to significant limitations in the use of this paper-and-pencil estimate of intellectual functioning.     

Wechsler performance IQ greater than Verbal IQ index in a forensic sample: a reconsideration

Explored the relationships of the Performance IQ (PIQ) greater than Verbal IQ (VIQ) to type of crime, ethnicity, and reading disability in a corrections sample of 70 adult males incarcerated on felony charges. The PIQ greater than VIQ sign was not related to Full Scale IQ or to violent vs. nonviolent crime, per se. The PIQ greater than VIQ sign showed a trend toward association with Ethnicity (black vs. white) and was related significantly to reading disability, with the reading disabled inmates more likely to show the sign, and to type of crime, with perpetrators of sex crimes most likely (87%) to show the sign and those incarcerated for murder or attempted murder least likely (33%) to show it. The difference in the proportion of inmates who showed the sign in these two classes of violent crimes (murder and sex crimes) was significant, and further analysis showed that with murder excluded, PIQ greater than VIQ occurred significantly more frequently in those accused of violent crimes than for nonviolent crimes. The latter findings suggested that differences between studies in the relationship of PIQ greater than VIQ and violence may be the result of differences in the proportion of murderers in the violent samples. Additional analyses indicated that the significant relationships between PIQ greater than VIQ and both type of crime and reading disability were most likely independent of ethnicity and each other.     

WAIS-R verbal-performance IQ discrepancies in persons with lateralized lesions: utility of a seven subtest short form.

WAIS-R Verbal-Performance IQ difference scores for Ward's (1990) seven subtest short form and the complete WAIS-R were examined in patients with lateralized and diffuse lesions. For both versions, the expected Performance > Verbal pattern was observed in the right hemisphere lesion group, while no summary score differences were seen in the left hemisphere group. Verbal-Performance IQ discrepancies for the short form fell within +/- 5 points of the WAIS-R discrepancy scores in about 75%of the cases, regardless of lesion location. Statistically reliable IQ differences between the complete and abbreviated WAIS-R attained 66%, 91%, and 89% agreement for the left, right, and diffuse groups, respectively. The results support the clinical utility of the seven subtest short form.     

Predicting WAIS-R IQ premorbid ability: combining subtest performance and demographic variable predictors.

The WAIS-R standardization sample was used to develop regression formulas designed to predict WAIS-R IQ scores. In contrast to previous approaches that focused on either "hold" measures or demographic variables as predictors of premorbid functioning, these two domains were used conjointly in the present study. WAIS-R subtests served as measures of present ability, whereas age, race, sex, education, and occupation served as the demographic predictors. Because any WAIS-R subtest may be impaired following brain injury, no subtest(s) were determined a priori to be "hold" measure(s). Instead, 33 regression formulas were developed each combining 1 of the 11 WAIS-R subtests with demographic variables, to predict IQ scores. The current equations generally doubled the amount of variance accounted for by demographic variables alone and accounted for more variance in WAIS-R IQs than any other currently available method of predicting premorbid cognitive functioning. Regression to the mean was not a significant problem. The regression equations are reported to aid in the estimation of premorbid WAIS-R IQ. Tables of predictive accuracy are also presented.     

Application of the four-subtest WAIS-R short form with an older clinical sample

Investigated the validity of the Arithmetic + Vocabulary + Picture Arrangement + Block Design subtest combination as an estimate of the WAIS-R Full Scale IQ in a sample of 55 elderly patients. The correlation between the short-form IQ and the WAIS-R Full Scale IQ was highly significant, r(53) = .95, p less than .001. A pairwise t-test computed between the mean IQs for the two forms was nonsignificant, t(54) less than 1. Approximately one-fourth of the Ss showed changes in intelligence classification levels when the short form was used. The short form does not provide a precise estimate of IQ; however, it appears to have clinical utility as a screening device to identify patients with atypical cognitive functioning. The short form correctly categorized 84% of the current sample with respect to the presence or absence of abnormalities on the full WAIS-R.     

Cross-validation of six short forms of the WAIS-R in a healthy geriatric sample

A number of methods have been developed in recent years to estimate full-scale IQ scores from abbreviated versions of the revised Wechsler Adult Intelligence Scale (WAIS-R), including Silverstein's (1982) two- and four-subtest short forms, Reynolds, Willson, and Clark's (1983) four-subtest short form, and Brooker and Cyr's (1986) two-, three- and four-subtest short forms. Short forms are of particular interest to the clinician and researcher who need a brief screening scale of cognitive abilities for the elderly, but their effectiveness with a normal aging sample has not been established. The present study examines the accuracy of six methods of estimating full-scale IQs from abbreviated versions of the WAIS-R in a middle-aged to elderly nonpsychiatric sample of 28 healthy males. Although all six methods were found to yield scores that were correlated highly with full scale scores, the method of Reynolds et al. (1983) resulted in short-form IQ estimates that were significantly different statistically from full-scale IQ scores. Although most of the six short-form methods appear to be useful screening measures for cognitive functioning and longitudinal changes in intellectual functions in the elderly, their use for classification purposes is not supported in the present study.     

Personality differences between eating-disordered women and a nonclinical comparison sample: a discriminant classification analysis

The present study evaluated the extent to which eating-disordered and nonclinical comparison samples could be differentiated on self-reported personality measures of autonomy and sociotropy and a projective measure of dependency. Sixty-two women meeting diagnostic criteria for eating disorder and a nonpsychiatric comparison group of 62 women completed the autonomy and sociotropy subscales of the Personal Style Inventory-II and the Rorschach Oral Dependency Scale (ROD). A discriminant classification analysis indicated 85.5% of eating-disordered subjects and 88.7% of control subjects were identified accurately from their scores on autonomy, sociotropy. and ROD. Results suggest that the experience of eating disorders is associated with a mixed clinical presentation characterized by issues related to sociotropy-dependency, and especially, achievement-related vulnerabilities. Potentially fruitful areas for future study include assessing the utility of autonomy and sociotropy as predictors of eating disturbances in prospective research and evaluating their utility in predicting treatment prognosis among patients with eating disorders.