ATLANTA — The mood of your tweets may foreshadow the outcome of your dieting efforts, a new study suggests.
Don’t believe it? Researchers at the Georgia Institute of Technology say the method they’ve devised successfully predicts the outcome of one’s diet an incredible 77% of the time.
Twitter users who tweet in a positive or optimistic light are generally more future- and people-oriented than their less positive peers, and tend to “have a greater sense of achievement in their social interactions,” Munmun De Choudhury, the study’s lead researcher, says in a university release.
This sense of positivity and accountability is believed to be responsible for one’s ability to stick to a diet regimen. These individuals also often engage with other topics relating to health and fitness on Twitter.
An example of a tweet considered positive would be: “If we never stumble we never fall. If we never fall we never fail, and if we never fail we never grow!”
A more negative tweet could entail one expressing how they feel as if they’ve lost control, whether in general or in a particular dimension of their life. The authors say a user might post “Feel rough as old boots this morning :/ Ankle hurts, shin hurts, chest hurts, head hurts.”
This study was able to go in greater depths than previous related studies in that it looked at 700 users who linked their MyFitnessPal (MFP) and Twitter accounts.
“Considering these data sources together and applying an established causality testing methodology allows us to validate for the very first time the efficacy of social media and quantified self-sensing in revealing risk to diet compliance,” De Choudhury explains.
More than 2 million tweets and 100,000 MFP food diary entries were examined in this groundbreaking study. The researchers compared their diary entries to tweets during the corresponding time period, examining how the tone and “linguistic attributes” of the tweets matched the user’s adherence to their diet plan.
Still, De Choudhury argues that more research should be done into exploring “the dynamics of events around when or how soon an individual’s diet is likely to fail… allow[ing] for proactive measures to be taken to help ensure more positive health outcomes.”
The study is detailed in the paper, Computational Approaches Toward Integrating Quantified Self Sensing and Social Media, with the findings being presented at the 20th ACM Conference on Computer-Supported Cooperative Work and Social Computing.