Elsevier

Schizophrenia Research

Volume 46, Issues 2–3, 15 December 2000, Pages 209-215
Schizophrenia Research

Short form of the WAIS-III for use with patients with schizophrenia

https://doi.org/10.1016/S0920-9964(00)00017-7Get rights and content

Abstract

The recent publication of the Wechsler Adult Intelligence Scale (WAIS-III), the most widely used standard test of intelligence, requires the development of a new short form for use with patients with schizophrenia for many clinical and research purposes. We used regression analyses of complete WAIS-III data on 41 outpatients with schizophrenia and 41 education-, and age-matched healthy subjects to determine the best combination of subtests to use as a short form. Excluding three subtests that are time-consuming to administer, and requiring that the solution includes one subtest from each of the four WAIS index scores, the combination that most fully accounted for the variance in full-scale IQ (FSIQ) for both participants with schizophrenia (R2=0.90) and healthy controls (R2=0.86) included the information, block design, arithmetic, and digit symbol subtests. When the restrictions regarding which subtests could enter were relaxed, the best four-subtest solution included information, block design, comprehension, and similarities. Although the latter explained 95% of the variance in FSIQ for schizophrenia participants and 90% of the variance for healthy controls, it consistently overestimated FSIQ for the schizophrenia group. We recommend the four-factor short form for use in future research and clinical practice in which a quick, accurate IQ estimate is desired.

Introduction

Intelligence tests provide an overall measure of cognitive competence and are predictive of performance on many other neuropsychological measures (Leckliter and Matarazzo, 1989). Thus, in both clinical and research settings, IQ results are useful as a single descriptive measure of overall intellectual ability and also provide an interpretive context for other test results. Many clinicians and researchers utilize abbreviated versions, or short forms, of the Wechsler Adult Intelligence Scale (WAIS)-R (Wechsler, 1981) to estimate full-scale IQ (FSIQ) scores, saving administration time and patient effort.

Short forms developed for the WAIS-R (Kaufman, 1990) may not be optimal for the WAIS-III (Wechsler, 1997) in light of the differences between the tests. Beyond item updating and re-norming, the WAIS-III differs from the WAIS-R in the inclusion of the matrix reasoning test in place of object assembly in the performance IQ, and in the addition of two new optional subtests, symbol search and letter–number sequencing. Although these two tests are not used in the calculation of the FSIQ score, their inclusion expands and clarifies the factor structure of the WAIS-III relative to prior versions of the WAIS. The WAIS-III provides four index, or factor scores, (verbal comprehension, perceptual organization, working memory, processing speed) as well as the familiar full-scale verbal and performance IQs. Thus the WAIS-III potentially provides more information about performance across discrete cognitive domains at the cost of administering additional subtests and an increase in total test administration time relative to the WAIS-R. The complete test takes 80 min, on average, for healthy subjects (Wechsler, 1997), and close to 100 min in a mixed clinical sample (the majority had a diagnosis of substance abuse disorder), with a significant minority requiring even longer (Ryan et al., 1998). Thus, administration of the complete test involves a substantial commitment of patient and examiner time and effort, both of which are frequently in short supply.

Ideally, a short form of the WAIS-III should provide a reliable estimate of FSIQ at a substantial time saving. In addition, it would be desirable to estimate FSIQ using subtests drawn from each of the four index scores, thereby providing a screening level of assessment of these domains. One potential psychometric problem of including measures from each of the factors is that the 11 subtests used to calculate FSIQ are drawn primarily from the verbal and perceptual organization factors. Thus, the inclusion of an equal number of subtests from each factor may systematically underestimate IQ in populations with relative deficits in working memory and processing speed, as is the case in schizophrenia (Psychological Corporation, 1997).

Our approach to short form development was guided by three considerations: (1) length of subtest administration and scoring time; (2) objectivity of scoring rules; (3) representation of scores from each of the four factors. With these criteria in mind, we eliminated several subtests from consideration for inclusion prior to conducting any statistical analyses. We dropped two relatively time-consuming subtests that are not used in the calculation of index scores (comprehension and picture arrangement). We also eliminated the vocabulary and matrix reasoning subtests. The vocabulary subtest involves considerable administration time and the use of detailed scoring rules. The selection of a different subtest from the verbal comprehension factor that involves minimal scoring judgement should enhance reliability and decrease scoring time. The 26-item matrix reasoning subtest (a new subtest from the perceptual organization factor) uses an untimed administration. Therefore, we eliminated this subtest in an effort to constrain total test administration time predictably. We conducted our analyses with the aim of discovering a WAIS-III short form with one subtest from each of the four factors that was highly predictive of FSIQ for both schizophrenia patients and healthy controls. We compared the validity of the four-factor short form with that of the best possible unrestricted four-subtest short form.

Section snippets

Participants

A total of 41 outpatients with either a DSM III-R (APA, 1987) or DSM IV (APA, 1994) diagnosis of schizophrenia or schizoaffective disorder from the Maryland Psychiatric Research Center participated in the study. All patients were diagnosed using a best estimate approach combining information from structured interviews, medical records, and collateral informants when available. The patients received the following subtype diagnoses: undifferentiated, 26; paranoid, 11; disorganized, 1;

Results

Table 2 shows the group means and t-test results for the comparisons of FSIQ and the four index scores. The groups differed on three of the four Index scores and on FSIQ. Similar findings, based on a partially overlapping set of patients, were reported in the WAIS-III technical manual (Psychological Corporation, 1997; pp. 166–168). Because the performance of schizophrenia patients varies across the WAIS-III indices, a short form that provides scores from each of the factors should provide more

Acknowledgements

We would like to thank David Tulsky and Psychological Corporation for providing access to the standardization sample data.

References (11)

  • Diagnostic and Statistical Manual of Mental Disorders (3rd ed. — Revised): DSM-III-R

    (1987)
  • Diagnostic and Statistical Manual of Mental Disorders (4th ed.): DSM-IV

    (1994)
  • J.L. Hintze

    NCSS 6.0.21 Statistical System for Windows

    (1996)
  • A.S. Kaufman

    Assessing Adolescent and Adult Intelligence

    (1990)
  • I.N. Leckliter et al.

    The influence of age, education, IQ, gender, and alcohol abuse on Halstead Reitan Neuropsychological Battery performance

    J. Clinical Psychol.

    (1989)
There are more references available in the full text version of this article.

Cited by (172)

View all citing articles on Scopus

Data included in this paper were presented at the International Congress on Schizophrenia Research in Santa Fe, NM, April 1999.

View full text