Twitter and Facebook have become clearinghouses for public sentiment, so it makes sense that Facebook has developed a new app aimed at determining “gross national happiness”–the moment-by-moment measure of glee, misery, and indifference for its 300 million daily users.
According to TechCrunch:
Data is collected from “public and semi-public forums” on Facebook, which is all anonymized before its analyzed. To determine if a particular status message is happy or sad (or neither), the app searches for popular phrases and words that the engineers have associated with each sentiment.
You can adjust the graph by sliding the bar at the bottom of the screen. You can also adjust the zoom by dragging the handlebars on the slider, and can actually watch happiness jump hour-to-hour, though it’s a bit difficult to navigate when you’re zoomed in that far. It’s fun to play around with, but you aren’t going to find many surprises: happiness generally hits a low on Mondays, then gradually grows up through the weekend when it drops again as the work-week begins. Peaks are all found around holidays, with Thanksgiving drawing the most happiness. Also worth nothing: this year there was an abrupt drop in happiness in late June, which is likely associated with the tragic death of Michael Jackson.
The idea of using something like “Gross National Happiness” has been mooted before–the word was invented by the monarch of the secluded Buddhist kingdom Bhutan, who suggested an alternate way of measuring national progress, based on well being. (Bhutan, incidentally, is one of the poorest countries in the region, although its populace is legendary for being spiritual.)
Numerous experts have suggested ways of measuring something as amorphous as “happiness,” and Bhutan is keen on creating a reliable index. Meanwhile, the recent economic meltdown even has Nobel-prize winning economist Joseph Stiglitz advocating a metric that emphasizes well-being over growth.
Facebook’s experiment is obviously a far more limited concept–public sentiment isn’t really a proxy for public well-being. You can bitch about football teams or shed a tear for MJ, while still leading a very happy life.
One technical issue strikes me as particularly tricky. I would bet that Facebook profiles, in terms of the “happiness” they express, betray a certain bias: It’s more typical for people to complain about how frustrated they are rather than about how content they are. Economists call that idea “loss aversion“–the idea that we’re wired to weigh losses more heavily than gains. I wonder how does Facebook works around such effects–if they exist. Would that skew the results to display more negativity, and would that in turn send the tone of our conversations spiraling downward?