The average “happiness” of English-speaking Twitter users has been on a downward slide over the last two years, according to researchers, as scientists attempt to use social data to gauge public health.
The study by researchers at the University of Vermont indicated the downward spiral of happiness, as measured by examining more than 46 million tweets, began dropping more quickly the second half of 2011.
The scientists assigned a “happiness” score to certain keywords, ranging from “pancakes” to “suicide,” to determine happiness levels. For instance, a word such as “laughter” earned an average “happiness score” of 8.5 out of 9, while the word “terrorist” got 1.30. The Twitter data was compared to the tweets’ date, time and other demographic information, like location, to determine which kind of people were feeling sad and when.
The university isn’t the first group to divine Twitter in an attempt to gauge a population’s mood. The Federal Reserve Bank of New York, recognizing social media’s potential to shape public opinion, and is working on creating tools to better monitor and analyze Facebook and Twitter chatter.
Wall Street high rollers, realizing a correlation between the collective mood expressed in millions of tweets and the Dow Jones Industrial Average, analyze Twitter to inform investment decisions.
The data can also be used to determine how well public health programs are working, serving for example, as a barometer on vaccination rates. Another study validated the use of the social media to gauge a population’s attitudes following the swine flu epidemic of 2009, finding more people got vaccinated in areas where people tweeted positive views on vaccination or gave updates on when and where they were going to get their shots.
Finally, other researchers created a machine to see if they could determine a person’s gender by their tweets alone, guessing correctly 76 percent of the time.
These various uses of social media underscore that the new tools convey more than a single message; they can be used as a mirror to gauge larger trends and collective feelings.
Peter Dodds, an applied mathematician at UVM and the lead author on the study, conceded the results skewed toward the younger set and those with smartphones, adding, “Twitter is a signal, just like looking at the words in the New York Times or Google Books… everything we say or write is a distortion of what goes on inside our head.”