We've all heard the stereo­type: Women like to talk. We bounce ideas off each other about every­thing from career moves to dinner plans. We hash out big deci­sions through our con­ver­sa­tions with one another and work through our emo­tions with discussion.

At least, that's what "they" say. But is any of it actu­ally true? Can we really make such sweeping gen­er­al­iza­tions about the com­mu­ni­ca­tion pat­terns of women versus those of men? The research is sur­pris­ingly thin con­sid­ering the strength of the stereo­type: Some studies say yes, women are more talk­a­tive than men. Others say there's no pat­tern at all. Still others say men are even bigger chatterboxes.

Per­haps all this con­tra­dic­tion comes from the dif­fi­culty of studying such a phe­nom­enon. Most of these studies rely on either self-​​reported data, in which researchers gather infor­ma­tion by asking sub­jects about their past con­ver­sa­tional exploits, or obser­va­tional data, in which researchers watch the inter­ac­tions directly. But both of these approaches bring with them some hefty lim­i­ta­tions. For one thing, our mem­o­ries are not nearly as good as we like to think they are. Sec­ondly, researchers can only observe so many people at once, meaning large data sets, which offer the most sta­tis­tical power to detect dif­fer­ences, are hard to come by. Another chal­lenge with direct obser­va­tion is that sub­jects may act in a more affil­ia­tive manner in front of a researcher.

But a new study from North­eastern pro­fessor David Lazer, who researches social net­works and holds joint appoint­ments in the Depart­ment of Polit­ical Sci­ence and the Col­lege of Com­puter and Infor­ma­tion Sci­ences, takes a dif­ferent approach. Using so-​​called "sociometers" - wearable devices roughly the size of smart­phones that col­lect real-​​time data about the user's social interactions - Lazer's team was able to tease out a more accu­rate pic­ture of the talkative-​​woman stereo­type we're so familiar with - and they found that con­text plays a large role.

The research was pub­lished in the journal Sci­en­tific Reports and rep­re­sents one of the first aca­d­emic papers to use sociome­ters to address this kind of question. The research team includes Jukka-​​Pekka Onnela, who pre­vi­ously worked in Lazer's lab and is now at the Har­vard School of Public Health, as well as researchers at the MIT Media Lab­o­ra­tory and the Har­vard Kennedy School.

For their study, the research team pro­vided a group of men and women with sociome­ters and split them in two dif­ferent social set­tings for a total of 12 hours. In the first set­ting, master's degree can­di­dates were asked to com­plete an indi­vidual project, about which they were free to con­verse with one another for the dura­tion of a 12-​​hour day. In the second set­ting, employees at a call-​​center in a major U.S. banking firm wore the sociome­ters during 12 one-​​hour lunch breaks with no des­ig­nated task.

They found that women were only slightly more likely than men to engage in con­ver­sa­tions in the lunch-​​break set­ting, both in terms of long- and short-​​duration talks. In the aca­d­emic set­ting, in which con­ver­sa­tions likely indi­cated col­lab­o­ra­tion around the task, women were much more likely to engage in long con­ver­sa­tions than men. That effect was true for shorter con­ver­sa­tions, too, but to a lesser degree. These find­ings were lim­ited to small groups of talkers. When the groups con­sisted of six or more par­tic­i­pants, it was men who did the most talking.

"In the one set­ting that is more col­lab­o­ra­tive we see the women choosing to work together, and when you work together you tend to talk more," said Lazer, who is also co-​​director of the NULab for Texts, Maps, and Net­works, Northeastern's research-​​based center for dig­ital human­i­ties and com­pu­ta­tional social sci­ence. "So it's a very par­tic­ular sce­nario that leads to more inter­ac­tions. The real story here is there's an inter­play between the set­ting and gender which cre­ated this difference."