Previous work on the "happiness gap" between liberals and conservatives took a relatively simple route: Just asking. Study subjects were asked to self-report their own happiness levels. In several academic studies (and one by Pew) conservatives repeatedly came out as generally cheerier than their left-wing countrymen.
The new results, published Thursday in the journal Science, took a different approach. Led by Sean Wojcik, a doctoral student in psychology and social behavior at the University of California at Irvine, the experiment analyzed photos and language analysis from the LinkedIn and Twitter profiles of those identified as either liberal or conservative.
"Common sense would dictate that if you want to know how happy someone is, you can ask them," said Peter Ditto, UCI professor of psychology & social behavior and co-author of the paper. "But what do you do if someone says they're happy, but doesn't act that way?"
Indeed, Ditto and Wojcik found more genuine smiles (as measured by standard facial analysis) and more positive language in the Web trail of liberals, even though other members of that group self-reported as less happy in the very same study.
The reason, they say, is that political conservatives have a tendency to self-aggrandize. When they compared happiness self-reports with tests that measured a tendency to enhance one's better qualities, they found that the happiness gap could be explained by a self-enhancement gap. In other words, liberals were being more honest about their personal pitfalls.
"There are two interpretations of this you could make: Either people are happier because this self-enhancement has a positive effect on their lives, or they're just appearing to be happier because of that tendency to self-enhance," Wojcik explained. He believes that a search for positive language and genuine smiles helps suggest that conservatives are in the latter camp.
Of course, this study isn't really any more definitive than the self-reported ones, and Wojcik understands that. If his new method showed the same results as self-reported surveys, it would be another story. "But this does raise more questions than answers," he said.
In future research, he hopes behavioral scientists can figure out how best to weigh data analysis like his versus self-reporting techniques.
"Happiness is really important to almost everybody, it’s the ultimate human experience in many ways," he said. "There are a lot of people interested in expanding happiness into policy making and having a national happiness index that we care about in addition to GDP and things like that. But it's important that we understand who is happier than whom, and that we learn how to measure that happiness comprehensively."