the comparability, equivalence, and parallelism of the wechsler adult intelligence scale - revised (wais-r) and the wechsler intelligence scale for children - third edition (wisciii) was compared using 16 year old youth. a total of 46 subjects, including 16 males and 30 females, were administered the wais-r and the wisc-iii in a counterbalanced order. the wais-r was administered first to 23 subjects, while 23 subjects were administered the wisc-iii first, with an average retest interval of 40 days. repeated measures multivariate analysis of variance (manova) and univariate repeated measures analyses (anova) were used to examine significant test, administration order, and practice effects. the means of the corresponding summary iqs were significantly higher for the wais-r than the wisc-iii. a clear pattern of practice effects was displayed when the wais-r followed the wisc-iii. with the exception of two subtests, the means on the wisc-iii were significantly higher than the corresponding means on the wais-r. the pattern of practice effects for the subtests was ambiguous. finally, ttests examined the differences in correlated variance, and votaw's (1948) test of compound symmetry was employed to compare covariances. the results of these tests showed minor differences. clinical and theoretical implications of these results are discussed.