Using Social Network Analysis to Better Understand Compulsive Exercise Behavior Among a Sample of Sorority Members

in Journal of Physical Activity and Health
Restricted access

Purchase article

USD  $24.95

Student 1 year subscription

USD  $115.00

1 year subscription

USD  $153.00

Student 2 year subscription

USD  $218.00

2 year subscription

USD  $285.00

Background:

Compulsive exercise, a form of unhealthy exercise often associated with prioritizing exercise and feeling guilty when exercise is missed, is a common precursor to and symptom of eating disorders. College-aged women are at high risk of exercising compulsively compared with other groups. Social network analysis (SNA) is a theoretical perspective and methodology allowing researchers to observe the effects of relational dynamics on the behaviors of people.

Methods:

SNA was used to assess the relationship between compulsive exercise and body dissatisfaction, physical activity, and network variables. Descriptive statistics were conducted using SPSS, and quadratic assignment procedure (QAP) analyses were conducted using UCINET.

Results:

QAP regression analysis revealed a statistically significant model (R2 = .375, P < .0001) predicting compulsive exercise behavior. Physical activity, body dissatisfaction, and network variables were statistically significant predictor variables in the QAP regression model.

Discussion:

In our sample, women who are connected to “important” or “powerful” people in their network are likely to have higher compulsive exercise scores. This result provides healthcare practitioners key target points for intervention within similar groups of women. For scholars researching eating disorders and associated behaviors, this study supports looking into group dynamics and network structure in conjunction with body dissatisfaction and exercise frequency.

Patterson is with the Dept of Wellness, Baylor University, Waco, TX. Goodson is with the Dept of Health & Kinesiology, Texas A&M University, College Station, TX.

Patterson (meg_patterson@baylor.edu) is corresponding author.
All Time Past Year Past 30 Days
Abstract Views 66 66 11
Full Text Views 7 7 2
PDF Downloads 6 6 1