People who live in the Bible belt stretching across the American South into Texas are less happy than residents of the west or New England.
That's according to an analysis of 10 million tweets (called a "hedonometer") created by a team at the Vermont Complex Systems Center.
The team coded each tweet based by the number of happy-seeming words (rainbow, love, beauty, hope, wonderful, wine, for example) or sad (damn, boo, ugly, smoke, hate, lied).
Curses, deemed negative, made a big difference in the results. For example, tweeters in Beaumont, Texas, deemed an unhappy bunch, tended to use the words "shit, ass, damn, gone, no, bitch, and hell" frequently.
Analyzing the overall emotional tone of a writer or speaker by counting the number of different types of words is a standard technique in psychological research--the newest thing here is that the researchers applied the method to tweets.