The role of reinforcement and nonreinforcement in an operant frustration effect

Male albino rats were run in a discrete-trial two-bar operant analog of the double alley. Completion of a FR 4 response chain on the first bar was rewarded 50% of the time for the 12 experimental Ss, but was never rewarded for the 12 control Ss. Both groups received consistent reward at the end of a FR 4 chain on the second bar. Eighty-four trials were given at a rate of four trials per day. A significantly faster rate of responding on the second bar was found following nonrewarded first-bar ratios than following rewarded first-bar ratios. This frustration effect was not attributable to response depression, since the nonrewarded performance of the experimental group exceeded that of the control group.     

The construct validity of Institutional Commitment: A confirmatory factor analysis

The present study examined the underlying structure of the variable Institutional Commitment by testing for the convergence, or lack thereof, among different indicators of the construct as represented by three theoretical frameworks (Tinto, 1975, 1987; Bean, 1985; Huselid and Day, 1991). Confirmatory factor analyses revealed that Institutional Commitment could be decomposed into two multiple indicators of the same latent construct: a general factor that groups items related to institutional quality, practical value of an education, utility of an education, fit between student and institution, and loyalty to the institution and another factor represented by items indicating similarity of values (Affinity of Values). Moreover, the study established the predictive validity of each subcomponent on different outcomes related to student persistence. While Institutional Commitment was found to have a significant direct effect on both students' intents to persist and actual persistence behavior, Affinity of Values was not as equally predictive of measures of student retention.Paper presented before the 1991 ASHE Annual Meeting. Boston, Massachusetts.