Further crossvalidation of regression-based neuropsychological norms with an update for the Boston Naming Test.

Regression-based norms for the Trail Making Test, Boston Naming Test, and Wisconsin Card Sorting Test, which we published in 1991 and 1993, have been criticized by Fastenau (1998) as having overcorrected for demographic influences in a sample of 63 older adults. We present data from new, independent participant samples that are consistent with expectations from the regression-based norms. We propose that Fastenau's findings in this instance resulted from the nonrepresentative nature of his relatively small sample, rather than from statistical deficiencies of regression based norms. Our currently published norms on one of the four tests considered here, the Boston Naming Test, are based upon a participant sample that was small and had inadequate representation of young adults. We address this by providing updated norms based on a much larger and more representative sample (N = 531).