A Twitter map of happiness, diet and fitness in New York, San Francisco, and Salt Lake

549
twitter map

Social media websites like Twitter and Facebook are more and more important in our daily life. People like to exchange information about health, happiness, and fitness on the websites.

One important feature of Twitter is that many users use mobile phones to tweet and share their locations (e.g., geotagged tweets). This provides a new data source to see people’s daily life.

In a study newly published in Applied Geography, researchers used publicly available geotagged tweets to examine neighborhood happiness, food, and physical activity in San Francisco, New York and Salt Lake.

Researchers explored 2.8 million tweets during February – August 2015. They linked 99.8% of tweets with GPS coordinates.

After that, researchers grouped tweets based on sentiment (1-9 points, sad to happy), food words (e.g., fish, chicken, and taco), and physical activity words (e.g., scuba diving, baseball, gardening, and soccer).

The result showed that New York had most happy tweets, followed by San Francisco and Salt Lake.

In New York, happiest tweets were near Central Park, Greenwich Village, SOHO, and Meatpacking District.

In San Francisco, happiest tweets were near areas boarding the bay or ocean and near Golden Gate Park as well as Mission District.

In Salt Lake, tweets in most areas were quite happy (6.0-6.5 on the 9-point scale), but tweets in the west and south sides had lower sentiment.

About 3.1-6.6 % of tweets were about food, and 16-17% of them mentioned healthy food and 10% of tweets mentioned fast food restaurants.

In all three places, coffee, beer, Starbucks, pizza, ice cream, tea, wine, burger, sushi, chocolate, cake, chicken and salad were on the top 25 popular foods list.

About 1.7% of tweets were about physical activity. Most of the tweets were about non-gym activity. New York had most tweets about gym.

In all three places, walking, dancing, running, yoga, swimming, golf, workout were on the top 10 popular physical activity lists.

Interestingly, the mean caloric density of food references was 250-261 calories/100g. The mean caloric cost of physical activity references was 217 – 219 calories/30 min workout.

In addition, tweets about non-fast food were happier than fast food tweets, and tweets about physical activity in New York were happier than tweets in Salt Lake.

Researchers suggest that Twitter data can provide better understanding of the well-being and health behaviors of communities. Governors can use the data to design better programs and polices to improve people’s health.


Citation: Nguyen QC, Kath S, Meng HW, Li D, Smith KR, VanDerslice JA, Wen M, Li F. (2016). Leveraging geotagged Twitter data to examine neighborhood happiness, diet, and physical activity. Applied Geography, 73: 77-88. doi: http://dx.doi.org/10.1016/j.apgeog.2016.06.003.
Figure legend: This Knowridge.com image is for illustrative purposes only.