Forty-nine nursing home residents were randomly assigned to a visitation condition focusing on conversational interaction, a visitation condition in which the playing of cognitively challenging games supplemented conversation, or a no-treatment control condition. Each subject in a visitation condition was visited by an undergraduate student twice per week for 8 weeks. Before and after the visitation period, all subjects were given four tests of cognitive functioning (vocabulary, matrices, memory, problem-solving), three tests of morale (Life Satisfaction Index A; Philadelphia Geriatric Center Morale Scale, self-perceived health), and were rated by nursing home activity directors on morale, program participation, alertness, sociability, and physical condition. A multivariate analysis of covariance, in which age, education, and length of nursing home residency were covariates, revealed a reliable overall effect for the treatment (p = .001). Subjects in both visitation conditions generally demonstrated improved performance relative to control subjects, and subjects in the conversation-plus-games condition demonstrated the greatest improvement. The univariate effects for memory, self-perceived health, and ratings of sociability were reliable.

Sixty women, aged 30, 35, 40, and 45, participated in a retrospective interview concerning psychosocial changes in their adult lives. Subjects' responses provided self-report data concerning specific psychosocial changes, and judges who read the interview protocols provided independent ratings of major psychosocial transitions. The distributions of self-reported changes and rated transitions across both chronological age and family cycle phases were examined statistically. Rated transitions were found to be reliably related to chronological age but not to family cycle phase. Specifically, 78% of the subjects manifested a major transition commencing between ages 27 and 30. That transition was characterized initially by personal disruption, followed by reassessment and finally by increased psychological well-being. Both chronological age and family cycle phase were reliably related to a variety of self-reported changes. The implications for conceptions of adult developmental changes are discussed.

This symposium focuses on understanding developmental experiences in the adult lives of women, including the nature of psychosocial changes and their relation to chronological age, family-cycle phase, and historical time. Harris, Reinke, and Ellicott report on a research program oriented to hypothesis testing. They interviewed 124 middle-class women randomly selected from seven age-groups (30, 35, 40, 45, 50, 55, and 60) about their life experiences retrospectively. Noteworthy findings included major transitions associated with ages 27–30 and with the preschool, launching, and postparental phases of the family cycle. In particular, the family cycle, which implies a context of relationships and social roles, illuminated many psychosocial regularities across women’s lives. In her discussion, Hancock describes some theory-generating research which suggests that women’s developmental crises, precipitated by ruptures in relationships, challenge assumptions about relationships and impel women toward active self-emergence.

ConclusionIn summary, Denney's failure to obtain the original results with her reanalyses can be attributed to the fact that she (a) scored some variables differently, (b) included an inappropriate subjects who was not entered into the original analyses, (c) excluded appropriate subjects who were used in the original analyses, (d) included an additional invalid dependent measure previously dicarded, and (e) may have used a different residualization algorithm. In view of these differences, it is not surprising that the multivariate and unvariate results that she obtained were different from those originally reported.We hope that now the procedural differences that account for the inconsistencies alleged by Denney are clear.

No abstract