Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Comparison of accuracy in predicting emotional instability from MMPI data: fisherian versus contingent probability statistics

Conference · · Trans. Am. Nucl. Soc.; (United States)
OSTI ID:5653637

The security plans of nuclear power plants generally require that all personnel who are to have access to protected areas or vital islands be screened for emotional stability. In virtually all instances, the screening involves the administration of one or more psychological tests, usually including the Minnesota Multiphasic Personality Inventory (MMPI). At some plants, all employees receive a structured clinical interview after they have taken the MMPI and results have been obtained. At other plants, only those employees with dirty MMPI are interviewed. This latter protocol is referred to as interviews by exception. Behaviordyne Psychological Corp. has succeeded in removing some of the uncertainty associated with interview-by-exception protocols by developing an empirically based, predictive equation. This equation permits utility companies to make informed choices regarding the risks they are assuming. A conceptual problem exists with the predictive equation, however. Like most predictive equations currently in use, it is based on Fisherian statistics, involving least-squares analyses. Consequently, Behaviordyne Psychological Corp., in conjunction with T.W. Mathews and Associates, has just developed a second predictive equation, one based on contingent probability statistics. The particular technique used in the multi-contingent analysis of probability systems (MAPS) approach. The present paper presents a comparison of predictive accuracy of the two equations: the one derived using Fisherian techniques versus the one thing contingent probability techniques.

OSTI ID:
5653637
Report Number(s):
CONF-870601-Summs.
Journal Information:
Trans. Am. Nucl. Soc.; (United States), Journal Name: Trans. Am. Nucl. Soc.; (United States) Vol. 54; ISSN TANSA
Country of Publication:
United States
Language:
English