Tweeting happiness: 10 happiest and 10 saddest states, based on Twitter rankings

Is it possible to measure the overall average happiness of states by using the words of residents, as posted on Twitter? A team of researchers at the University of Vermont says yes.

Using the Mechanical Turk Language Assessment word list, which scores words on a scale of 1 (sad) to 9 (happy), a recent study examined geotagged tweets from 50 U.S. states and Washington, D.C., coding each tweet for happiness.

"Happy" words were defined as things like rainbow (8.1), love, hope and wonderful, while negative words were defined as things like earthquake (1.9), boo, ugly and lied. Vulgarity also ranked negatively, bumping down scores through what researchers called, "geoprofanity."

In addition to ranking states by happiness, researchers also calculated happiness for nearly 400 cities, tied their data with other existing measures of happiness, such as Gallup surveys, and examined how their data correlates with income and obesity in an area.

Although researchers don't take the context of words into account when looking at tweets, they argue that large amounts of text can still give reliable results.

"An analogy is that of temperature," the research report said. "While the motion of a small number of particles cannot be expected to accurately characterize the temperature of a room, an average over a sufficiently large collection of such particles defines a durable quantity."

Read the full report at www.uvm.edu.