The validity of computer-based personality assessment products: Recommendations for the future

摘要

The use of computers in psychological assessment is now well established. A central aspect of this field, computer based interpretations of personality and psychodiagnostic tests, has been the subject of widespread criticism due to the lack of demonstrated validity. The present paper separates methodological issues in validation research from technical issues in the development of objective interpretation rules and in the computer application of these rules, and makes six basic recommendations for future development: (a) attend to the reliability of all variables, especially predictors; (b) focus on specific (narrow-band) predictions; (c) simply restate the empirical findings in their raw and perhaps contradictory form; (d) apply each system only to populations that resemble the derivation population; (e) use the “no interpretation” option; and (f) consider building in a capacity to accomodate varying base rates. Use of these guidelines would result in simpler, more specific systems, which would be viewed as tools for the assessment psychologist and not as finished products in their own right.