Are you happy? Are you really, genuinely happy in life? Go ahead and take a second to think about it. That can be a really tough question to answer. After all, how do we define happiness? Rank it? Measure it?

Well, if you ask the researchers at the University of Vermont, you go by people’s Tweets. Attempting to gauge humanity’s levels of happiness, the team devised a mood algorithm of sorts by enlisting Amazon‘s Mechanical Turk service, which assigned emotions to words ranging from “pancakes” to “suicide.” Then they hit Twitter — crunching through a global sampling of 46 billion words from 63 million Twitter users over a three-year span — to determine an overall ranking of people’s emotions.

Past happiness studies have often relied on using tools like polls and surveys, but those techniques are pretty flawed. (Does it surprise anyone that people tend to “fudge the truth” when taking such surveys?) Still, the team needed some quantifiable way to measure the intangible, leading them to the 140-character messaging service.

So how happy is mankind? Well, apparently not as much as we were a couple of years ago. In the December 7th issue of the PLoS ONE journal, the researchers explained, “After a gradual upward trend that ran from January to April 2009, the overall time series has shown a gradual downward trend, accelerating somewhat over the first half of 2011.”

Click to enlarge

“It appears that happiness is going down,” says Peter Dodds, an applied mathematician at UVM and the study’s lead author. There could be many reasons for this. According to the study, when people have felt the most to down in the dumps, it coincided with events that took them outside their typical routines. And world events also lined up with mood drops, including news of the spread of swine flu, the U.S. economic bailout, and the tsunami in Japan. (Actor Patrick Swayze’s death even swayed the numbers.)

This is an intriguing study using an innovative approach… but is it science? Really, how reliable can this data be? For example, say Mechanical Turk ranked the word “fail” as a 2 on a happiness scale of 10. Does that mean “epic fail” is always scored as sad — even when it’s a funny joke that makes tons of Twitter followers laugh out loud? And what about older or non tech-savvy folks? Isn’t Twitter sort of self-selecting a particular demographic, making any Twitter sampling inherently skewed? Well, yeah — Dodds acknowledges this to some extent, but his team believes the service “is nearly universal now.”

Flawed or not, the researchers aren’t done chasing down happiness yet. Future studies will take a similar approach using The New York Times and Google Books.

[via PsychCentral]