Anticipated Retirement : generalized locus of control, interest in the future, and morale

A sample of 87 persons, expecting to retire within from 20 years to 1/4 year and predominantly white, married, professional women with moderate incomes and high educational achievement, was recruited by mail after being identified by professional, work place, and religious organization associates. Levenson locus of control scale (1974), Marlowe-Crowne Social Desirability Scale (Crowne & Marlowe, 1960), PGC Morale Scale (Lawton, 1975). and 2 interest in retirement scores as well as time to retirement and demographic data were obtained for most subjects.

A previously reported positive relationship between generalized internal locus of control and interest in the future (Rotter, 1966) was examined by correlational analysis as well as by multitrait-multimethod analysis. Pre-retirement morale was explored with multiple regression statistical techniques.

Although the obtained associations between locus of control and retirement interest were not statistically significant, the pattern of association was consistently as expected: retirement interest was more positively associated with internal locus of control than with powerful other or chance locus of control. The pattern of association between locus of control scores and retirement interest was similar to previously observed associations between locus of control and other cognitive variables (Levenson, 1974): internal and powerful other locus of control were associated weakly and similarly with retirement interest and chance locus of control was associated negatively and more powerfully with retirement interest.

Pre-retirement morale was significantly and consistently associated with powerful other and chance locus of control and with time to retirement. Powerful other and chance locus of control were associated similarly and negatively with morale, and greater time to retirement was associated with higher morale. The association between retirement interest and morale was consistently negative though not significant.

Powerful other locus of control appeared to associate differently with the cognitive and affective variables examined, and this finding was taken to suggest that Levenson's tri-dimensional concept of locus of control is a useful differentiation of the construct.

Retirement interest, evaluated as a psychological trait, demonstrated minimal convergent, discriminant, and construct validity.

In no instance did associations with the independent variables account for as much as 35% of the variance in retirement interest or morale scores, and it was concluded that retirement interest and pre-retirement morale are not yet adequately understood. In addition, it was inferred from the response characteristics of this group of pre-retirees and their associates that social isolation in retirement could develop from miscommunication with associates before retirement.