Motherboard

There are lots of things you can infer from Twitter. But while we’re learning what we’re eating or where we’re flying, we haven’t been able to use Twitter to determine how we’re feeling.

This animation, by researchers at Harvard and Northwestern, illustrates the mood in the U.S., as inferred using over 300 million tweets, over the course of the day.

The results? The early morning and late evening appear to provide the highest levels of happiness (first graph). Geographically, the data points to a significantly happier west coast, which is consistently three hours behind the east coast (second graph).

Weekends tend to be happier (see, Twitter does teach us new things!). The peak in mood is observed on Sunday mornings, and the trough occurs on Thursday evenings. See below:

The moods of each tweet were surmised using the University of Florida’s ANEW word list. The maps are represented using density-preserving cartograms, in which the area is scaled in order to be proportional to the number of tweets that originate in that region.

It’s unclear if the mood takes into account emoticons, which are still probably the most telling, if imprecise, measure of mood online – until we kick start that Emotion Markup Language, of course. And it’s also unclear to what extent actually using Twitter impacts one’s mood (think anxiety, confusion, the glee of ego-boosting).

In the meantime, finishing this blog post makes me feel kinda happy, though if Twitter is any indication, that feeling won’t last long.

See a full size poster (pdf) here.