Validation of WAIS-III Four-Subtest Short Forms in Patients with Traumatic Brain Injury

Abstract

In an effort to identify four-subtest Wechsler Adult Intelligence Scale-Third Edition (WAIS-III) short forms valid for estimating Full-Scale IQ (FSIQ) among individuals with traumatic brain injury (TBI), seven tetrad versions of the WAIS-III were evaluated in a convenience sample of patients referred for neuropsychological assessment (n = 176). Estimated FSIQ scores were compared to actual FSIQ scores via correlation analyses, repeated-measures analyses of variance (ANOVAs), and frequency analyses. All short form-estimated FSIQ scores correlated highly with actual scores (all rs > .91, ps < .001). Repeated-measures ANOVAs identified no significant differences between actual and short form-estimated FSIQ scores for two of the seven short forms. These same two short forms had the highest percentage of scores within ±5 points of actual FSIQ scores (75.6% and 71.6%). Thus, two tetrad versions were consistently superior to others in accuracy of estimating FSIQ; these may be helpful when time constraints or other issues necessitate use of an abbreviated battery for estimating FSIQ among individuals with TBI.

Key words:

• assessment
• brain injury
• intelligence
• neuropsychology

Notes

n = 176.

^a Short form validated for use with elderly clinical sample (Donnell et al., 2007).

^b Identified as useful for rapid screening (Sattler & Ryan, 1999).

^c Identified as having the highest validity of 19 tetrad combinations (Sattler & Ryan, 1999).

*All correlations between WAIS-III Actual FSIQ and WAIS-III Short Forms FSIQ were significant (all ps < .001).

**Repeated-measures ANOVAs compared WAIS-III Actual FSIQ with each WAIS-III Short Form.

^a Short form validated for use with elderly clinical sample (Donnell et al., 2007).

^b Identified as useful for rapid screening (Sattler & Ryan, 1999).

^a Short form validated for use with elderly clinical sample (Donnell et al., 2007). In the present study, this form did not differ from WAIS-III Actual FSIQ in ANOVA analyses.

^b Identified as useful for rapid screening (Sattler & Ryan, 1999). In the present study, this form did not differ from WAIS-III Actual FSIQ in ANOVA analyses.