WAIS-R Reliability and Standard Errors for Persons 75 to 79, 80 to 84, and 85 and Older

Abstract
Reliability of the WAIS-R for a sample of 226 normal elderly volunteers was determined. Subjects were divided into three age groups: 75 to 79, 80 to 84, and 85 and older. For each group, internal consistency estimates and standard errors of measurement (Sem1), estimation (Sem2), and prediction (Sem3) were calculated for the IQs and subtests. Vocabulary (.96 to .92) was consistently the most reliable subtest, while Picture Arrangement (.71 to .54) and Object Assembly (.74 to .56) were the least reliable. The summary scores were highly reliable, with coefficients for the Verbal Scale that ranged from .97 to .95, for the Performance Scale from .92 to .89, and for the Full Scale from .97 to. 96. It was concluded that the WAIS-R is a reliable instrument for evaluating elderly subjects.