Researchers are now using data from Twitter to study our emotions:
Positive-mood tweets peaked twice a day throughout the world, early in the morning and again near midnight. The morning peak came later on weekends, presumably because people slept in. That the cycle was similar on weekdays (when pressures like work deadlines and school exams pile up) and weekends (when most people are more relaxed) showed that sleep schedules and circadian rhythms were important influencers of mood, regardless of day-to-day stresses, the authors reported.
This article was published in Science, which is a very prestigious journal. However, I think it was probably accepted for its novelty and not the quality of the research itself. I scroll through about 3 dozen research articles everyday with findings that are just as interesting.
Overall, I’m a little hesitant about studies that study Twitter users in broad aggregates. My experience with Twitter is that its users fall into three categories:
- People who use it to communicate with friends
- Businesses that use it as a marketing and customer service portal
- Technophiles and self-absorbed people who use it to feed their addiction or vanity
I’m not sure those types of people add up to an homogenous whole. As a research tool I think Twitter has more utility as something that allows researchers to slice and dice users across narrowly defined demographics based on their social interactions. Someone’s twitter posse is probably a more natural study group than strangers that check off the same demographics boxes on a survey.
Health Kismet Blog 