Abstract
Differences in subjects who remained with or dropped out of a research project were evaluated using discriminant analyses. Subjects were 79 factory employees who received ten different questionnaires measuring perceptions and attitudes during a 22 month research period. Attrition was not predicted by any of the eight personality variables. Situational variables significantly predicted attrition; subjects who completed all questionnaires were more likely to have been assigned to one set of sister teams, to have been socialized into the organization by group/formal methods, and to have been initially assigned to the night shift. Discriminant analysis is a statistical tool to determine some of the bias of longitudinal data.
Get full access to this article
View all access options for this article.
