The question of the comparability of the WISC and WISC-R: Review of the research and implications for school psychologists

The paper reviews WISC/WISC-R comparison studies which have been conducted with a wide variety of samples. Significant WISC/WISC-R IQ score differences have been reported with the WISC-R yielding consistently lower scores of approximately 5-8 IQ points for the three major scales. Several studies do report variable WISC/WISC-R differences for various ages, races, and ability levels. These results have implications for practicing school psychologists. Caution is advised in the interpretation of a WISC/WISC-R difference, as a discrepancy of one SD may not be meaningful. Many students who scored in the borderline classification range on the WISC and who are currently being readministered the WISC-R are scoring in the mentally impaired classification range. This does not necessarily reflect negatively on the validity of the WISC-R, but does document the need to keep intelligence tests up to date. There is a continuing need to exercise caution in the use of individual intelligence tests and to utilize data in addition to WISC-R scores in order to make special education placement decisions.     

A comparison of WISC and WISC-R results of urban educable mentally retarded students

Twenty urban EMR students' WISC-R scores were compared with their previously administered WISC scores. The average interval between administrations was approximately three years. Significant coefficients of correlation were obtained between five corresponding subtests and between Performance IQs and Full Scale IQs. Results from t tests for correlated data indicated that mean WISC-R Verbal, Performance, and Full Scale IQ scores were not significantly lower than their corresponding WISC scores. The results of this study indicated that WISC-R scores in comparison to WISC scores appeared to be more stable in the lower IQ ranges and less vulnerable to cultural bias.     

WISC-R: The first five years

Since its introduction five years ago (1974), 113 articles or papers have appeared regarding the WISC-R, including empirical investigations of its nature, as well as its comparability with a variety of other measures of intelligence and achievement, including the WISC. While not all this research has been carefully done, two general conclusions can be derived from the review. First, although the WISC-R involves modification in administration, design, and presentation of items, as well as a complete restandardization, the literature substantially suggests that it remains very similar in nature to its predecessor, the WISC. Investigations of factor analytic structures, standard errors, reliability coefficients, and subtest intercorrelations support the conclusion that individuals perform on the WISC-R largely the same as they do on the WISC. The second conclusion points out (with few exceptions): consistently lower scores were obtained on the WISC-R than on several other measures, including the WISC, the WAIS, the Slosson Intelligence Test, the Peabody Picture Vocabulary Test, and the Stanford Binet, which was revised shortly before the WISC. These lower scores on the WISC-R may be due to a variety of influences, including examiner variance, an artifact of design leading to inflated scores on the WISC, and finally and most obviously, the restandardization of the scale. The amount of literature that has appeared over the five-year period suggests that practitioners and researchers are as interested in learning about the WISC-R as they were about the WISC. Despite this fact and the conclusion that the WISC and WISC-R are substantially similar, the present authors encourage caution in the overgeneralization of findings until additional literature develops.     

Matched pair comparison of WISC and WISC-R scores

Both the WISC and the WISC-R were administered to 54 children, with one half taking the WISC first and the other 27 taking the WISC-R first. Differences between mean subtest scaled scores and mean IQs were found to be influenced by the sequence of the tests, although all mean scores were higher on the second test when the WISC-R was given first. Data were presented suggesting the WISC-R given first tends to raise WISC scores, and WISC given first yields scores on the WISC-R which are essentially similar, effectively counsteracting the more “difficult” scoring for the WISC-R.     

Comparability of the WISC and the WISC-R in normal children of divergent socioeconomic backgrounds

WISC and WISC-R IQs of two groups of normal ten-year-old children from divergent socioeconomic backgrounds (N = 36) were compared in a counterbalanced research design. Generally, all WISC mean IQs were higher than the WISC-R mean IQs for both groups of children. Significantly higher WISC IQs were obtained on the Verbal and Full Scales of the low socioeconomic group and on the Performance and Full Scales of the high socioeconomic group. Correlations between tests for Verbal, Performance, and Full Scale IQs were.95,.79, and.91 for the low socioeconomic group, and.84,.57, and.87 for the high socioeconomic group. Although practice effects did not appear to significantly affect IQs of the low socioeconomic group, various practice effects were observed in the high socioeconomic group. Negative effects were observed for both WISC and WISC-R on the Verbal Scale, whereas positive effects were observed on both Performance and Full Scales in this group. Analysis of simple effects reveals that WISC and WISC-R IQs did not differ significantly upon first administration, but were significantly different upon second administration. Conclusions were drawn that while WISC-R yields a lower IQ estimate than WISC for children of lower ability, results were not so clear for the children of higher ability, due to practice effects which disproportionately favor increases on WISC IQs upon second administration. The greater practice effects found on the WISC have implications for comparisons of the two tests in counterbalanced research designs. Simple randomized or treatment-by-levels design may be more useful in comparisons of the WISC and the WISC-R because of the bias introduced by counterbalanced designs.     

A longitudinal comparison of the WISC and WISC-R with special education pupils

The purpose of this study was to determine the degree of comparability between the WISC and the WISC-R over time. Hypotheses were tested using a sample of 276 mildly mentally handicapped children enrolled in special education classes who had received a WISC (Trial 1) and a WISC or WISC-R (Trial 2) after an interval of approximately three years. The sample was grouped on the basis of test administered in Trial 2. Group I contained those children who received the WISC on Trial 2 and included 183 children, of whom 121 were males and 62 were females. Mean age at time of initial testing was 8–10; mean age at second testing was 11–8. Group II was composed of 93 children who had been posttested on the WISC-R and included 72 males and 21 females. Mean age at initial testing was 9–5; mean age at posttest was 13–0. The study was conducted in five school districts and one county department of education. Verbal IQ, Performance IQ, and Full Scale IQ for both instruments administered were transcribed from the special education folders of the children. The design used to test hypotheses was repeated measures analysis of variance. Data analysis resulted in rejection of the hypotheses that the WISC-R yields mean Verbal, Performance, and Full Scale IQ scores greater than or equal to the corresponding mean IQ scores of the WISC. It was concluded that the WISC-R may unfairly penalize special education children who are reevaluated with this instrument. Fluctuations in IQ scores due to the instrument of measurement must be recognized, and appropriate action should be taken to insure that children are evaluated for special class placement on comparable bases.     

A comparison between the WISC and WISC-R among educable mentally retarded students

The study compared differences and similarities between WISC and WISC-R scores for 48 ten- and thirteen-year-old EMR students, who were matched according to sex and race. The WISC-R yielded significantly lower VS, PS, and FS IQ scores than the WISC for this predominantly rural Georgia population. The FS IQ scores from the WISC-R averaged 7.5 points lower than the WISC FS IQs, with similar differences noted for VS and PS scores. There was no significant difference in FS IQs between the two age groups. Results suggest that many children classified as “Borderline” or slightly above by the WISC will be classified as “Mentally Deficient” by the WISC-R.     

Suitability of the California Abbreviated WISC for WISC-R subtests with mentally retarded and slow learner children

This study was undertaken to determine the suitability of the California Abbreviated WISC—Form 1 (CAW-1) for use with WISC-R subtests. Subjects were drawn from the files of previously evaluated students from two southern mountain communities. Examination of WISC records within the age and IQ ranges of the CAW-1 (CA 8-0 to 13-6, IQ 50 to 80) indicated that our sample was not significantly different from the CAW-1 noim group (N = 148, r = .85 between CAW-1 & WISC Full Scale IQ). WISC-R recods (N = 178) for the same ranges were examined and were found to be more highly predictive of Full Scale IQs (r = .91, FSIQ = .98 CAWIQ + 2.2). Extensions of the age range from 6-0 to 16-11 and of the IQ range from 40 to 100 were attempted. Final results (N = 284) indicated the CAW-1 could be used with the WISC-R over the full age range and for IQs ⩽ 100 with minor alteration (r = .96, FSIQ = 1.09 CAWIQ - 5.4). The second sample served for cross validation of these results. Using the alteration, results were consistent with original findings (N = 202, r = .93, FSIQ = .98 CAWIQ + 2.2). It was concluded that the CAW-1 was suitable for use with WISC-R subtests with minor alterations.     

Investigation of the WISC-R for use with retarded children: Correlation with the 1972 Stanford-Binet and comparison of WISC and WISC-R profiles

A group of 80 mentally retarded youngsters, aged 6 to 16, was tested on the WISC-R, primarily to assess the continuity of measurement between the old and new WISCs. The WISC-R IQs correlated .65 to .82 with Stanford-Binet IQ for a subsample of 45 children, resembling the coefficients between the 1949 WISC and the Binet for retarded groups. In addition, the WISC-R test profiles for the 80 children corresponded closely to the WISC test profiles for many retarded samples. Thus, there was evidence to support the continuity of the WISC-R with its predecessor for retarded populations.     

A comparison of the WISC and WISC-R

So as to compare the results of the WISC and WISC-R, both instruments were administered to 58 children randomly selected from a school population of 583. All administration and scoring was performed by the same psychologist, with a two-month interval separating the administrations for each child. All IQs were significantly higher (p <.01) on the WISC, with the Performance difference being greater than the verbal difference. Also, 8 of the 10 required subtest scaled scores were significantly greater (p <.05) on the older instrument. Regression equations were obtained to predict WISC-R IQs from WISC scores.     

WISC,WISC-R,S-B L&M,WRAT: Relationships and trends among children ages six to ten referred for psychological evaluation

Thirty children ages 6-10 were selected who had been referred for psychological evaluation from the public schools. WISC-R was significantly lower than WISC FS IQs and similar to S-B L&M IQs (1972). Each scale predicts reading, spelling, and achievement in a significant manner. The use of the WISC-R over the WISC will rosult in greater numbers of children being classified as retarded.     

Deficits in bannatyne's sequencing ability on the WISC and WISC-R for mildly retarded and borderline functioning children over time

The WISC and WISC-R protocols of 94 special education students (51 white males, 5 black males, 30 white females, and 8 black females) who tested in the mildly retarded and borderline range of abilities (IQs of 50–78) were examined. Their test performance was then followed longitudinally for two additional WISC and WISC-R assessments. Mean age of subjects was 8 years at Test 1, 10.75 years at Test 2, and 14.9 years at Test 3. As predicted, subjects had lower IQ equivalent scores on Bannatyne's Sequencing Ability category than their IQ equivalent scores on the Verbal Comprehension or Perceptual Organization factors of the WISC and WISC-R for all three testings. Results were interpreted as supporting the notion that mildly retarded children and learning disabled children have qualitatively similar learning patterns.     

Direct comparisons of the factor structures of the WISC and the WISC-R

The factor structures of the WISC and the WISC-R were compared, using the data for age groups 7.5, 10.5, and 13.5 years in the two standardization samples. The results for three different types of solution—hierarchical, multiple-group, and principal-factor—agreed in demonstrating that the structures of the two scales are as similar as the structure of either scale is for different age groups.     

The seductive nature of WISC-R short forms: An analysis with gifted referrals

The use of a short form of the WISC (California Abbreviated WISC for the Intellectually Gifted) (CAW-IG) developed for screening gifted referrals was investigated with the WISC-R. Two methods were used to estimate IQ. and the results indicated that the short form could be effective in the prediction of Full Scale IQ. This was true even when the prediction was based on data from the original WISC. Discussion centered on the implications of these results for the use of shortened IQ tests. A number of problems were examined, and practitioners are cautioned against indiscriminate use of such procedures.     

Comparison of the WISC-R and PPVT for a group of mentally retarded students

This study compared the WISC-R and PPVT scores for a group of 65 students classified as mentally retarded, ranging in age from 7.5 to 14.5. Pearson product moments (r) and correlated t values were employed in the analysis of data. The evidence from the present study indicated that for this sample of mildly retarded children and youth, the PPVT IQ score is significantly higher than the Full Scale IQ from the WISC-R. The current findings indicate that the same general pattern exists for the WISC-R and PPVT as for the WISC and PPVT scores.     

A note on WISC-WISC-R differences with mentally retarded and borderline functioning adolescents

Records of 94 randomly selected students from the Department for the Mentally Retarded in the Special School District of St. Louis County (mean age 14.8 years) were reviewed. Forty-nine subjects had been given the WISC (25 males and 24 females), and 45 subjects had been given the WISC-R (31 males and 14 females). Although no statistically significant differences were found between the two groups, the WISC tested lower than the WISC-R by an average of 1.8 points in Verbal IQ, 2.6 points in Performance IQ, and 2 points in Full Scale IQ. This finding was consistent with other studies that have indicated WISC-WISC-R differences of about 2 points for children over I I years of age.     

A comparison between the WISC and WISC-R among a clinical referred population

This study compared the three scales of the WISC and the WISC-R for degree of intercorrelation. The differences among scores and among variances were also computed for the two tests. Fifty-eight subjects from a clinical referred population were studied, including children classified as EMR, Learning Disabled, Emotionally Disturbed, and children in regular classrooms. Results included significant correlations in all cases, no systematic changes in variance, and significantly lower IQs on the WISC-R for all children in this sample, except the EMRs.     

Factor structure of the WISC‐V in four standardization age groups: Exploratory and hierarchical factor analyses with the 16 primary and secondary subtests

This study examined the factor structure of the Wechsler Intelligence Scale for Children‐Fifth Edition (WISC‐V) with four standardization sample age groups (6–8, 9–11, 12–14, 15–16 years) using exploratory factor analysis (EFA), multiple factor extraction criteria, and hierarchical EFA not included in the WISC‐V Technical and Interpretation Manual. Factor extraction criteria suggested that one to four factors might be sufficient despite the publisher‐promoted, five‐factor solution. Forced extraction of five factors resulted in only one WISC‐V subtest obtaining a salient pattern coefficient on the fifth factor in all four groups, rendering it inadequate. Evidence did not support the publisher's desire to split Perceptual Reasoning into separate Visual Spatial and Fluid Reasoning dimensions. Results indicated that most WISC‐V subtests were properly associated with the four theoretically oriented first‐order factors resembling the WISC‐IV, the g factor accounted for large portions of total and common variance, and the four first‐order group factors accounted for small portions of total and common variance. Results were consistent with EFA of the WISC‐V total standardization sample.     

The impact of the Flynn effect on LD diagnoses in special education

Because of the Flynn effect, IQ scores rise as a test norm ages but drop on the introduction of a newly revised test norm. The purpose of the current study was to determine the impact of the Flynn effect on learning disability (LD) diagnoses, the most prevalent special education diagnosis in the United States. Using a longitudinal sample of 875 school children who were initially diagnosed with LD on the Wechsler Intelligence Scale for Children-Revised (WISC-R), children experienced a significant decline in IQ when retested on the third edition of the WISC (WISC-III) compared to peers who were tested on the WISC-R twice. Furthermore, results from logistic regression analyses revealed that the probability of a rediagnosis of LD on reevaluation significantly decreased, in part, because of this decline on the WISC-III. These results are discussed in terms of their implications for both basic research and educational policy.     

WISC-R factor structures among White,Mexican American,and African American children: A research note

The factor structure for the Wechsler Intelligence Scale for Children-Revised (WISC-R) was investigated for White (n = 183), Mexican American (n = 129), and African American (n = 139) nonreferred school-age children (6.8 to 14.6 years) of low-to middle-class socioeconomic background. Exploratory and confirmatory factor analyses were performed. The WISC-R factor structure model tested was Kaufman's (1975) three-factor solution that consisted of Verbal Comprehension (VC, Factor 1), Perceptual Organization (PO, Factor 2), and Freedom from Distractibility (FD, Factor 3). The results of the exploratory analyses showed Kaufman's three-factor solution for the three ethnic groups, but the order of Factor 2 and Factor 3 was reversed for the Mexican American and African American samples—thus raising questions about the comparability of the WISC-R factor structure across groups. The confirmatory analyses, which used the orthogonal nested factors approach by Gustafsson and Balke (1993), resulted in a model (for each of the ethnic groups) in which a G factor explained most of the variance (followed by the nested first-order factors—VC, PO, and FD). The results of the confirmatory analyses provide some support for David Wechsler's (1974) original intentions that the WISC-R structure is best described as having a general, or global, factor. © 1997 John Wiley & Sons, Inc.