A Longitudinal Look at the Predictors of Four Types of Retirement

The general purpose of the study was to further the understanding of the retirement process. More specifically, the study investigated several personal and work-related predictors of individual retirement status that have received relatively little past examination. By gaining a better understanding of the retirement process, policy makers and organizational leaders can use this knowledge to make informed decisions affecting older workers, their organizations, and society in general. Public release data of the Health and Retirement Study (HRS)--a recent, large, representative, national panel study that focuses on the health, retirement, and aging of persons born between the years of 1931 and 1941--were analyzed. In-home, face-to-face interviews were conducted in 1992, with follow-up interviews conducted every two years by phone. From survey years 1992 through 2000, samples of working individuals over the age of 50 were used to examine nine hypotheses and four different operationalizations of retirement. The hypotheses involved the following predictor variables: gender, partner status, characteristics of one's partner (i.e., retirement status and health), number of dependents, obsolescence of job skills, "demanding" job characteristics, perceived age-related discrimination, work and retirement attitudes, and flexibility of organizational policies regarding older workers. Event history analysis using logistic regression was used to predict the dichotomous criterion variables, while linear mixed model analysis for repeated measures was used to predict the continuous dependent variable. After statistically controlling variables such as age, health, and finances, results indicated that gender, partner's retirement status, work and retirement attitudes, and organizational flexibility were each predictors of at least one type of retirement. In addition, the statistical interaction of gender with partner status predicted retirement defined in several ways. The study expanded what is known about retirement predictors by examining actual versus intended retirement behavior and increased the generalizability of past research by utilizing national panel data. The study also highlighted the importance of clearly defining retirement measures and explored the complete versus partial retirement typology. It was suggested that flexible work options may improve work and retirement attitudes by making the transition into retirement a gradual process, and may keep older employees in the work force longer.

Gerontology

17590