Despite the sensitivity of neuropsychological tests to educational level, improved diagnostic accuracy for demographically corrected scores has yet to be established. Diagnostic efficiency statistics of Wechsler Adult Intelligence Scale-III (WAIS-III) and Wechsler Memory Scale-III (WMS-III) indices that were corrected for education, sex, and age (demographically corrected) were compared with age corrected indices in individuals aged 16 to 75 years with moderate to severe traumatic brain injury (TBI) and 12 years or less education. TBI participants (n = 100) were consecutive referrals to an outpatient rehabilitation service and met careful selection criteria. Controls (n = 100) were obtained from the WAIS-III/WMS-III standardization sample. Demographically corrected indices did not provide higher diagnostic efficiency than age corrected indices and this result was supported by reanalysis of the TBI group against a larger and unmatched control group. Processing Speed Index provided comparable diagnostic accuracy to that of combined indices. Demographically corrected indices were associated with higher cut-scores to maximize overall classification, reflecting the upward adjustment of those scores in a lower education sample. This suggests that, in clinical practice, the test results of individuals with limited education may be more accurately interpreted with the application of demographic corrections. Diagnostic efficiency statistics are presented, and future research directions are discussed.