There were tem­poral trends too. “The sad­dest was Hunter Col­lege High School,” says Bar-​​Yam. “It turns out that we col­lected data just after they came back from vacation.”

Alan Mis­love of North­eastern Uni­ver­sity in Boston says that using Twitter for such analysis has pos­i­tives and negatives.

On one hand, it gives researchers access to a large number of users and a diver­sity of topics. On the other, the analysis can be con­founded because users may be biased, or have mul­tiple accounts, or be using cor­po­rate accounts. “There is an on-​​going debate about how to use sites like Twitter to mea­sure cer­tain trends,” says Mislove.

Read the article at New Scientist →