Skip to main content
Open Access Publications from the University of California

UC Irvine

UC Irvine Electronic Theses and Dissertations bannerUC Irvine

Network Quality over Quantity: Exploring the Influence of Network Structure and Function on U.S. Older Adult Physical and Mental Health


Does the structure of an individual’s personal network effect their physical and mental health? Or does the structure of the network influence individual health by influencing the functions of the network or the individual’s health behaviors? In my dissertation, I examine these questions using two waves of nationally representative data from the National Social Life, Health, and Aging Project and employing conditional change models.

In study 1, I examine the associations between network structure, network function, and health. I find no direct associations between baseline network structure and later health. However, I do find that baseline network structure is associated with later network function, which, in turn, is associated with health. I also find limited support for a feedback loop in which baseline health has effects on later network structure. Perceived social support may be more important for the health of older adults than network structure characteristics, but network structure is meaningful in that it provides the source from which support can be derived.

In study 2, I examined the effects of spousal loss on depression across different levels of spousal support, including baseline network structure and network function to observe whether they served as a buffer to spousal bereavement. The findings revealed that losing a spouse was positively associated with the number of depressive symptoms, but only for those with highly supportive spouses. None of the network structure and function measures at baseline were associated with later depression. The findings imply that spousal bereavement in late life depends on how supportive the spouse was.

In study 3, I examine the social network factors that influence health risk behaviors—specifically, cigarette use and alcohol consumption. Findings show that personal networks can have protective effects on smoking or drinking; however, for current smokers and drinkers, personal networks are enabling. Having more support and having health discussion partners reduced the adoption of health-compromising behaviors. In conclusion, health-compromising behaviors may not necessarily be adopted as a coping mechanism, but may be a pro-social activity that is increased when there are others in the network who also engage in similar activities.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View