Related Experiment Videos
WAIS-R short-forms: criterion validity in healthy and clinical samples
J R Crawford1, B Mychalkiw, D A Johnson
1Department of Psychology, King's College, University of Aberdeen, UK.
The British Journal of Clinical Psychology
|November 1, 1996
Summary
This study evaluated regression equations for estimating Wechsler Adult Intelligence Scale-Revised (WAIS-R) IQs using short-forms. The six- and seven-subtest short-forms demonstrated superior validity in neurological samples.
Area of Science:
- Psychometrics
- Neuropsychology
- Cognitive Assessment
Background:
- The Wechsler Adult Intelligence Scale-Revised (WAIS-R) is a widely used measure of adult intelligence.
- Estimating full-length WAIS-R IQs from shorter test batteries (short-forms) can be valuable for efficiency in clinical and research settings.
- Previous regression equations exist to facilitate this estimation, but their validity requires ongoing evaluation, particularly in diverse populations.
Purpose of the Study:
- To cross-validate regression equations for estimating full-length WAIS-R IQs from short-forms.
- To compare the criterion validity of different short-form lengths and their accuracy in estimating IQs.
- To assess the performance of these estimation methods in both healthy and neurological patient samples.
Main Methods:
- Regression equations developed by Crawford, Allan & Jack (1992) were applied to estimate full-length WAIS-R IQs.
- Two cross-validation samples were utilized: a healthy group (N=153) and a neurological group (N=471).
- Correlations between estimated and full-length IQs were analyzed, alongside an examination of IQ over/underestimation.
Main Results:
- In the healthy sample, the correlations between short-form estimates and full-length WAIS-R IQs were comparable to the original findings by Crawford et al. (1992).
- In the neurological sample, six- and seven-subtest short-forms (Crawford et al., 1992; Warrington, James & Maciejewski, 1986) showed significantly higher criterion validity.
- These specific longer short-forms did not systematically under- or overestimate IQs in the neurological group, unlike other shorter forms.
Conclusions:
- The regression equations for estimating WAIS-R IQs from short-forms are generally reliable in healthy individuals.
- For neurological populations, longer short-forms (six or seven subtests) offer superior criterion validity and accuracy in IQ estimation.
- These findings support the use of specific, longer short-forms for more precise WAIS-R IQ estimation in individuals with neurological conditions.