Stereotypes suggest women are warmer and more empathetic than their male counterparts, and now a study has suggested this may be true - at least on social media.
The research was conducted by the World Wellbeing Project at the University of Pennsylvania.
It uses scientific techniques for measuring psychological wellbeing based on analysis of social media.
By analysing 65,000 Facebook users, the team found women used language that was "characteristic of compassion and politeness", compared to the more impersonal and distant language used by men.
But it revealed women's language is just as assertive as men.
"We found women mentioned friends, family and social life more often, whereas men swore more, used angrier and argumentative language and discussed objects more than people," said Margaret Kern, co-author of the study.
"In many ways, the topics most used by women versus men are not surprising as they fit common gender stereotypes. The computational methods let us make visible what the human mind does to automatically categorise people and things that we encounter in our everyday life."
Kern said that the team's algorithms are able to correctly predict the gender of a particular user "over 90 percent of the time".
The results don't necessarily reassert the gender binary though, Kern stressed.
"Our results demonstrate gender is a complex, multi-faceted and fluid concept," she said.
"But this study does show self-reported gender does influence the way people express themselves on Facebook.
The algorithms looked at other facets of language, too; how "affiliative", or socially connected, posts were, and how assertive.
Andrew Schwartz, who is a professor at Stony Brook University and was also involved in the study, said it offers a "fresh perspective" on gendered language.
"Here we were able to use a novel technique to explore psychological dimensions. While some previous work suggests men are more assertive, the language in Facebook didn't reflect this," he said.
The findings are published in the journal PLOS One.
This article was originally published by WIRED UK
