Return to search

Predicting weight loss in blogs using computerized text analysis

An increasing number of people are turning to online blogging communities
devoted to self-change for smoking, shopping, and other behaviors. To understand
processes underlying effective self-change, the current project tracked the language and
social dynamics of a dieting blog community using computerized text analysis. Three
research questions were asked: What predicts weight loss in blogs? What changes in
blogging predict weight loss? Can we predict dropping out or successful weight loss
based on the first two entries? A community of blogs devoted to weight loss was
examined (n = 2530). Most bloggers were female, and on average, around 30 years old,
approximately 200 pounds, with a goal weight of about 140 pounds. A sample of blogs
by females that had blogged at least 15 entries within the first 15 weeks of blogging
resulted in a total of 186 blogs, representing over 9,200 entries for analysis.
Computerized text analysis was used to examine language for rates of self-focus,
emotionality, cognitive processing, keeping food diaries, and social support. Rates of blogging were assessed by word counts, number of active weeks, and mean entries per
week. Social support was assessed through the use of social words, the size of the social
network, along with the positivity and negativity of the comments. The discrepancy
between start and goal weight was also assessed. The results suggested that having larger
weight loss goals and blogging about personal events was a more effective weight loss
strategy than keeping an online food intake diary. The degree to which bloggers were
socially integrated with the blog community was found to be a potent predictor of weight
loss. Online components of behavioral treatment programs could encourage dieters to
browse and comment on other dieters’ progress, and to share personal narratives rather
than simply focusing on the benefits of food intake diaries, nutrition, and exercise. The
current project points to the power of computerized text analytic tools to address
important theoretical and practical social psychological issues that are evolving on the
internet. Specifically, language analysis methods can identify which dimensions of blogging communities can help or hinder self-change processes. / text

Identiferoai:union.ndltd.org:UTEXAS/oai:repositories.lib.utexas.edu:2152/6541
Date16 October 2009
CreatorsChung, Cindy Kyuah
Source SetsUniversity of Texas
LanguageEnglish
Detected LanguageEnglish
Formatelectronic
RightsCopyright is held by the author. Presentation of this material on the Libraries' web site by University Libraries, The University of Texas at Austin was made possible under a limited license grant from the author who has retained all copyrights in the works.

Page generated in 0.0017 seconds