False positive low score rates on Wechsler Adult Intelligence Scale-IV México and U.S. version core subtests from neurologically intact, balanced bilingual, Mexican Americans.

Abstract

Objective: To compare false positive rates on Wechsler Adult Intelligence Scale-IV (WAIS-IV) core subtests from the México and U.S. versions among neurologically intact, balanced bilingual, Mexican Americans with varying cut scores (≤ 7; ≤ 6; ≤ 5 scale scores: i.e., ≥ 1.0, ≥ 1.33, and ≥ 1.67 SDs) while estimating preexisting skill level at the normative 50th percentile versus a performance-based estimate.

Methods: Retrospective analysis was conducted on 60 participants (M = 10.18 education years) administered the two WAIS-IV versions (intertest interval = 5.68 days) in a counterbalanced repeated measures design. Sample language balance was established by subtraction of the Spanish from the English Woodcock-Muñoz Language Survey-Revised Picture Vocabulary (WMLS-R PV) subtest scores for each participant (M = -0.18).

Results: Estimating preexisting skill level at the 50th percentile of WAIS-IV norms produced false positive rates of 0.32, 0.17, and 0.06 for the México version and 0.53, 0.37, and 0.24 for the U.S. at cut scores of ≤ 7, ≤ 6, and ≤ 5, respectively. Using the WMLS-R PV to estimate preexisting skill level significantly reduced the false positive rate to between ≤ 3% and zero across both versions and all cut scores.

Conclusions: WAIS-IV false positive rates, likely influenced by racial/ethnically correlated variables of years of education, bilingualism, and qualitative aspects of education, were significantly lower for the México than the U.S. versions across all cut-scores. They were significantly reduced for both versions and all cut scores by using a performance-based estimate of preexisting skill level versus the normative 50th percentile.