''Conceptual clarification'' or confusion in ''The study of sensation seeking''

Jackson and Maraun claim that the Sensation Seeking Scales (forms II, IV, and V) do not measure sensation seeking because the scales were developed using factor analyses of items rather than rationally using the original construct. The constructs were changed after the factor analyses revealed a different structure than anticipated and the new constructs were validated using Cronbach and Meehl's 'construct validity'. One cannot assume that the meanings attributed to test items on a rational basis is the same as that given by test respondents. If there are subfactors one cannot assume they are organized in the meaning dimensions hypothesized by the test constructors. Meaning resides in the construct validity of the test as well as the content validity and factor analyses are an important step in establishing content validity. Copyright (C) 1996 Elsevier Science Ltd.