Are we #unhappy?
“We’re at our lowest point now in four years as far as our measure of happiness through Twitter goes,”says Peter Dodds lead researcher on what is the earliest of what will likely be many studies of its kind on social media and the state of well-being. Dodds and a team of researchers at the University of Vermont examined more than 4.6 billion tweets from 63 million users over 33 months to determine that happiness levels are declining.
The research, which has been published in PLoS One, measured the level of happiness and unhappiness that appeared in the 4 billion-plus tweets using a 10-point scale to rank words like laughter (8.5), greed (3.06), and terrorist (1.3). The analysis revealed that happiness levels have been steadily declining since 2009 but are declining at an escalating level since the start of 2012.
Certain events triggered exceptionally low points: the death of Osama Bin Laden, the TARP bailout program (which created a cascading, multiple week slide), and September 29, 2008 (the day the DOW dropped almost 1,000 points). Other moments in history had similar impact on happiness levels, such as the 2009 H1N1 pandemic, Michael Jackson’s death, and the death of Patrick Swayze.
Natural disasters also played a role in reducing levels of happiness. The study, in addition to looking at long-term happiness levels, also revealed a number of trends: according to the Twitter data, happiness is at its highest levels on Saturdays, with Fridays and Sundays following closely behind. The unhappiest day of the week is Tuesday. The study also tracked happiness by hour. The happiest hour of the day is 5am to 6am, then happiness levels steadily decline throughout the day with the lowest levels occurring between 10pm and 11pm.
Just as certain evens tipped the scales toward declining levels of happiness, certain events were likely to increase happiness levels as well. Peaks in happiness occurred on Christmas Eve, Christmas Day, New Year’s Eve, New Year’s Day, Valentine’s Day, Easter, Independence Day, and Thanksgiving Day, as well as on Mother’s Day and Father’s Day.
While the authors tout the value of using social media to identify and interpret social science data, they admit that their analysis involved tweets from a “non-representative subpopulation.” The study’s lead researcher, Peter Dodds, explains, “As we have seen in both the work of others and ours, Twitter and similar large-scale, online social networks have thus far provided good evidence that scientifically interesting and meaningful patterns can be extracted from these massive data sources of human behavior…The era of big data social sciences has undoubtedly begun. Rather than being transformed or revolutionized we feel the correct view is that the social sciences are expanding beyond a stable core to become data-abundant fields.”
While the reliability of the data may be called into question, a similar study by Facebook corroborated the data, revealing that people were happiest on weekends and holidays.