Three Papers on Social Participation over the Life Course

Social participation encapsulates the involvement and participation of individuals in social
activities (e.g., informal social gatherings, affiliations to community organizations), and has long
been a subject of interest for sociologists. Recent scholarship has since established a strong
positive association between social participation and health, suggesting that social participation
can buffer the negative health effects of stress and promote healthy behavior through social
influence, among other pathways. There are however three key limitations of prior research.
First, perhaps driven by anxieties around rising health costs of an aging population, studies on
social participation and health overwhelmingly focus on older populations. Second, many are
interested in examining societal change in social connectedness over time, but given the use of
repeated cross-section data, are at risk of conflating age and cohort effects. Third, research often
treats social participation solely as a characteristic of the individual, even though the social
participation of proximate others may also affect one’s own outcomes – e.g., spouses may be
influenced by their friends’ health behavior, and in turn influence their partners. We know very
little about how social participation operates in the context of interdependent individuals such as
spousal dyads. This dissertation addresses existing gaps in the literature by applying the life
course perspective to the study of social participation and health. I do this through a series of
papers that (1) examine how social participation varies over age and cohort; (2) establish how the
association between social participation and health changes with age; and (3) investigate how
social participation and health is associated in the context of marital dyads. The first paper uses
data from the Americans’ Changing Lives (ACL) study, a longitudinal dataset collected from the
same individuals over 25 years (1986-2011). I employ a Multivariate Bayesian generalized
additive mixed model to estimate age-cohort trajectories of formal and informal social
participation. I find that changes in social participation by age and cohort are less drastic than
commonly assumed; older adults seem to compensate for age-graded declines in informal social
participation by increasing their formal social participation. Any anxiety around societal declines
in social connectedness precipitated by past studies is overblown – later-born cohorts seem to
have similar (or greater) levels of social participation compared to older cohorts. In the second
paper, I use data from ACL once again, employing growth curve models to estimate how the
association between social participation and health changes with age. I find that formal social
participation (e.g., attendance and engagement in community groups and organizations) becomes
more important for males as they age – the negative association between formal social
participation and depressive symptoms becomes stronger in old age. Using data from the Health
and Retirement Study, the final paper utilizes actor-partner interdependence models to examine
social participation and mental health among married couples. I find evidence supporting the
hypothesis that spousal social participation is positively associated with one’s own mental health
(i.e., partner effects), even after accounting for interdependencies in mental health between
spouses. Overall, the dissertation applies life course principles to provide a more comprehensive
view of social participation and its associations with health outcomes. Findings suggest that
social participation in late life should be considered alongside social participation earlier in the
life course, and that the social participation of proximate others can also influence our own
health outcomes.