This research will investigate linkages between life course social inequalities, psychosocial factors, and adult physical health. Specifically, differences in several dimensions of socioeconomic status (SES) and race-ethnicity will be examined as potential social determinants of differences in psychological factors (e.g., affect, personality style, sense of control, self-evaluation, perceived discrimination, religiosity/spirituality) and social relationship factors (e.g., marital history and quality; intergenerational relationships and quality; other kin and nonkin relationships and quality). These psychological and social relationship factors, in turn, will be examined in terms of their influence on health status (health perceptions, functional status, symptom and illness reports, mortality). The hypothesis that differences in psychological factors and social relationship factors help mediate and moderate the influence of SES and race-ethnicity on health will be explored. Further, two hypotheses about the pathways linking psychological and social factors to physical health will be evaluated: 1) the """"""""health behavior mediation"""""""" hypothesis, which posits that differences in psychological and social relationship factors lead to differences in health behaviors, which, in turn lead to differences in physical health, and 2) the """"""""psychosocial stress moderation"""""""" hypothesis, which posits that differences in psychological and social factors lead to differences in the way individuals adapt to life histories of cumulative disadvantage, which, in turn, lead to differences in physical health. All analyses also will explore differences across gender. This work will use expansive new data from three large longitudinal population sample surveys: 1) the Wisconsin Longitudinal Study (WLS), a study of 10,317 Wisconsin high school graduates in 1957, which includes information from main respondents in 1957 (age 18), 1964 (age 25), 1975 (age 36), 1992-93 (age 54), and 2003- 04 (age 65); information from siblings in 1977,1992-93, 2004; and information from graduate spouses and widows in 2004; 2) the National Survey of Families and Households (NSFH) 1987-88,1992-93, 2002-03, a study of 13,007 U.S. adults aged 19-95 in 1987-88; and 3) the National Survey of Midlife in the United States (MIDUS), 1995, 2004, which includes information from 3,485 primary respondents aged 25-74 in 1995 and 1,000 siblings of primary respondents. Multivariate regression models, event-history models, hierarchical linear models, and structural equation models will be estimated as appropriate.
Showing the most recent 10 out of 11 publications
