The Validity of
Michael D. Franzen
Department of Psychiatry
and Allegheny Neuropsychiatric Institute
Allegheny General Hospital, Pittsburgh
Washington State University
Validity is a term often applied to tests, but as frequently noted, validity is best applied to the inferences drawn from tests rather than applied to the tests themselves. We can speak of the validity of neuropsychological assessment procedures if we expand our conception to include the interpretation process. That still leaves us with the question of whether the uses of neuropsychological test scores are valid. The question of validity cannot be answered outside this context of the application of the information derived from the scores.
The problem is whether the neuropsychological test scores are useful in answering the question at hand. Part of that problem involves the qualitative issue of whether the test data can efficiently and accurately answer the question. Another part of the problem is the extent to which the utility of the test scores outweighs their cost. That is, can the prediction be made in some alternate way that involves other data that are more cheaply acquired? As a corollary of that question, we must then evaluate the extent to which the unique predictive power of the test data increases the overall predictive power of other data including behavioral observations, demographic information, and even other neuropsychological test data. This last question is sometimes referred to as incremental validity. In this chapter, we consider traditional forms of validity and discuss two newer forms -- ecological validity and descriptive validity. The implications for validity concerns of different periods in the development of individuals is also discussed.