Sleep complaints are common among older adults, and poor sleep has been found to predict chronic diseases and mortality. Many studies suggest that social participation benefits healthy aging. We examined the relationships between older adults’ social participation and their sleep using two waves (2005–2006–2010–2011) of data from the National Social Life, Health, and Aging Project (NSHAP). The NSHAP recorded older adults’ social participation (including religious attendance, volunteer work, and attendance at meetings of organized groups) over five years, and included self-reported sleep duration in both waves and, in the second wave, measures of insomnia symptoms and measures of sleep patterns and rhythms using actigraphy for a subsample. Cross-sectional analysis of the second wave indicates that those reporting higher levels of social participation had better actigraphic sleep but not better self-reported sleep. However, longitudinal analysis suggests that change in social participation was not associated with actigraphic or self-reported sleep characteristics in the second wave data. Further analysis using fixed-effects models showed no association between changes in social participation and changes in self-reported sleep duration. Thus, although older adults with greater social participation slept better, we did not find that increasing social participation improved sleep. These findings imply that a self-selection process may at work; or if social participation does affect sleep, the causal effect may be over a much shorter time frame than five years.