Advisory Committee Chair
Gareth R Dutton
Advisory Committee Members
Sarah-Jeanne Salvy
Marissa Gowey
Olivio Clay
Document Type
Thesis
Date of Award
2019
Degree Name by School
Master of Arts (MA) College of Arts and Sciences
Abstract
Adult behavioral obesity interventions often show improved attendance and weight loss when delivered in a group, as opposed to individual, therapy format. However, research on the exact mechanisms through which group therapy confers additional weight loss is lacking. Pediatric obesity interventions have recently begun to utilize social network analysis (SNA) to identify structural ties that may improve adherence and weight loss; however, this research is limited and has yet to be replicated in an adult weight loss program. The current study is the first to evaluate social networks in an adult behavioral weight loss intervention. Four weight loss groups were recruited (N=65). Whole-network analyses assessed whether in-degree ties (peer nominations) and reciprocal ties (where peers nominate each other) predicted increased weight loss (%) and attendance after 6 months. Additionally, the study computed measures of similarity (homophily) to test whether individuals who formed ties with those similar to them had improved attendance and weight loss. Popularity ties did not significantly predict attendance or weight loss. Similarly, reciprocated support ties did not significantly predict weight loss or attendance (both p>0.05). However, in-degree support ties significantly predicted attendance (β=0.255, p=0.032) but not weight loss (β=0.103, p=0.398). In-degree interaction ties also significantly predicted attendance (β=0.302, p=0.010) and weight loss (β=0.291, p=0.014). Reciprocal interaction ties significantly predicted attendance (β=0.279, p=0.018) and weight loss (β=0.249, p=0.038). Finally, homophily of social ties based on baseline weight, race, or income was not correlated with weight loss (all p>0.05), but sharing ties with those of similar baseline weight (r=0.320, p=0.042) and different incomes (r=-0.463, p=0.002) predicted greater attendance. Together, these findings suggest that social networks can impact attendance and weight loss outcomes in adult weight loss groups. Forming ties with others of similar baseline weight may improve attendance among a population with higher attrition rates. Additionally, individuals with higher income may find themselves with greater social capital in their weight loss groups, but only those who affiliated with others of different income levels appeared to show improved attendance.
Recommended Citation
Borgatti, Alena, "The Weight of Social Relationships: Social Network Analysis in an Adult Weight Loss Program" (2019). All ETDs from UAB. 1217.
https://digitalcommons.library.uab.edu/etd-collection/1217