No abstract

Motives for achievement, affiliation, power as fear of weakness, and power as hope of power were assessed using thematic apperception in two representative national surveys of adults, one in 1957 and the other in 1976. Comparison of the results allows us to mark changes and stable features of national character over the 19-year period. Controls for education and age differences were introduced into the year comparisons for men and women separately. Among men, the achievement motive has remained stable, the affiliative motive has decreased, and both power motives have increased. Among women, both the motives for achievement and power as fear of weakness have increased, but there has been no change in the other two motives. These results are interpreted in terms of large-scale social changes (e.g., the women's movement, the shrinking opportunity for jobs) that have dominated our recent social history.

This investigation of conjugal social support was based on 412 married respondents from a national sample of adults 50 years of age or older. A review of the literature yielded hypotheses concerning the exchange of three forms of support: emotional support, respect, and health-related support. Results showed that older respondents were least likely to provide each form of social support. These findings were reviewed in light of evidence for increased need for support in later life. The perceptual bases of social support were discussed in conjunction with the finding that women perceive less social support within marriage than men.

We adopted a multidimensional approach to the study of the social support convoys of older adults. We distinguished between age and gender differences in four specific dimensions of the social support convoy: (a) existence versus functioning of relationships in the convoy, (b) kinds of relationships (i.e., those with children, siblings, and friends), (c) types of social support (i.e., emotional support, respect, and health support), and (d) receipt versus provision of support. Using a national survey of 718 adults, multivariate and univariate analyses of variance were performed to determine age and gender differences in these four dimensions of social support. The multidimensional approach was useful in pinpointing those aspects of the social support convoy affected by aging. We anticipated that the social support convoy would be devastated by aging. Instead, we found that older people received less support (i.e., emotional and health support) in the absence of sibling relationships. Otherwise, the effects of aging had more to do with what the older person contributed to the convoy than with what he or she received. Women had better social support resources than men, particularly within their friendships. We found no evidence, however, that women's social support advantage counterbalanced the effects of aging on the convoy.

Despite plentiful evidence of a link between social support and adaptation to stress, we still know virtually nothing about why this connection exists. This paper is designed to help the researcher plan a study that offers the optimal test of any given explanation. We assume that the research is founded on a conceptual model of the association between stress, support, and strain. The specificity of such a model will depend upon what has been established from existing knowledge. The paper focuses on key issues in measurement, design, and analysis. A fundamental problem in measurement of social support is how to select from a host of possible strategies. The measurement section introduces important considerations that should govern that choice. The design section alerts the reader to sources of bias that commonly constrain the inferences that can be drawn from social support research. It also addresses the thorny problem of timing measurement to capture a process that transpires over time. The analysis section illustrates how structural equations can be used to model specific social support effects and discusses the circumstances that would make inclusion of additional variables advisable. Finally, we consider the limitations of relying exclusively on quantitative data and illustrate ways in which qualitative research can be used to develop unique information about the role of social support in circumventing the effects of stress.