The work patterns of women with some work experience over the 1976 to 1989 period were examined as they approach retirement, using data from the Mature Women's cohort of the National Longitudinal Surveys. The data provided information on a sample of women who were between the ages of 30 and 45 in 1967 and who have been interviewed regularly since then. The analysis focused on the years 1967-89, during which the women changed from ages 39-54 to 52-67. The study found that married women reduce their annual weeks worked and hours worked per week at a slower rate than single women in their working years, and at a much slower rate than married men. Although there is a positive correlation in the changes in hours worked per week between spouses as they near retirement, the relationship is not strong. The differences between married and single women and between married women and men may be a result of the probability that both single women and men are more likely to have full-time career jobs and thus build up more retirement benefits than do married women, who are more likely to have part-time, noncareer jobs. (KC)