Long search for a short WAIS: Stop looking.

Abstract

The time-consuming aspect of the Wechsler intelligence scales has prompted their frequent abbreviation in clinical practice. The use of selected items from each subtest has been a particularly attractive method of shortening because it reduces administration time by about S0% and yet gives scores for each subscale. To test the reliability of scores obtained from this method, 200 protocols of the Wechsler Adult Intelligence Scale (WAIS) were rescored according to short-form procedure and reliabilities based on split-half correlations obtained. It was hypothesized that one could predict the reliability of the shortened WAIS on the basis of the Spearman-Brown formula, and that in testing, as in other fields, "you get what you pay for." Results confirmed these hypotheses-a short form is not an adequate substitute for the full WAIS.The Wechsler intelligence scales are among the most widely used and highly regarded tests in clinical practice. They are also among the most time consuming, and many kinds of abbreviations have been used. One extensive study (Holmes, Armstrong, Johnson, & Ries, 1965), for example, showed that 90% of the Wechsler Adult Intelligence Scales (WAISs) were given in abbreviated form. In an attempt to retain the virtues of these tests, but shorten administration time, much research has been directed to establishing the reliability and validity of various methods of abbreviation.A review of the literature suggests a fourfold typology of test abbreviation. Shortening has been achieved through the elimination of 1 The coding and computer work in this study were made possible by a grant from the Veterans Administration Central Office. The data were collected while all the authors were affiliated with the Ann Arbor Veterans Administration Hospital. Appreciation is expressed to the Psychological Corporation for making available information on how reliability coefficients were computed in the original WAIS standardization study.The statements and conclusions of the authors are the result of their own study and do not necessarily reflect the opinion or policy of the Veterans Administration.2 Requests for reprints should be sent to Margaret B, Luszki,

Efficacy of the Satz-Mogel short form WAIS—R for tumor patients with lateralized lesions.

“…Finally, the percentage of patients for whom the categorization of IQ and age-corrected scale scores changed in using the short form was rather disturbing. About one fifth of patients showed one or more category changes in IQ scores, which is similar to the percentage reported by Luszki et al (1970), which they suggested was a problem serious enough to cause concern. One third to one half of the patients showed at least one category change in subtest scores, the only exception being for the Comprehension subtest.…”

“…Alternatively, other researchers determined the percentage of patients correctly classified by the Satz-Mogel short form according to Wechsler IQ standards. Luszki, Schultz, Laywell, and Dawes (1970) reported correct FSIQ classifications for 82% of their patients, whereas Massad et al (1988) found an agreement rate of 89%.…”

A comparison of the intertest accuracy of two short forms of the WAIS-R

“…The findings thus provide a partial confirmation of the conclusion of Luszki et al (1970) that short forms involve a loss of clinically relevant information that pertains to intellectual strengths and weaknesses. However, given the reality of continued use of short forms, it is important to note that the WAIS-RM involves significantly less distortion of WAIS-R intertest scatter than does the S-M.…”

“…A number of investigators have shown that the Satz-Mogel short form (S-M) distorts the pattern of intertest scatter and therefore provides less reliable cognitive profiles than the full WAIS-R. (See Zimmerman & Woo-Sam, 1973.) Luszki, Schultz, Laywell, and Davis (1970) demonstrated that the subtest reliability of the Satz-Mogel approach would be reduced by nearly 50% in comparison with the WAIS. These authors thus concluded that psychologists should "stop looking" for a clinically reliable WAIS short form.…”

Structural Factor Analysis Experiments with Incomplete Data

“…For example, time constraints may require the definition of a variety of short forms, requiring only half of the available measures (Luszki, Schultz, Laywell & Dawes, 1970). Figure 2 includes four different configurations of models using only four of the eight WAIS-R measures.…”