Worrying about what other people think at work is something women have to contend with far more than men, a small new study suggests.
The experiment included only 32 participants, but economists in Germany say their "clear and consistent" findings are thorough evidence of a newly-identified gender discrepancy.
Experiments have found that when people are placed in pairs, the contributions of both male and female workers depend on how they feel about their female partner. Basically, the authors explain, in pretty much every interaction a woman has in the workplace, 'likeability' is either an asset or a hurdle.
"For men, on the other hand, likeability matters only if they interact with the opposite sex," they add.
In all-male groups, being likeable was neither an asset nor a hurdle. In the study, male participants cooperated and coordinated to the same extent, regardless of their feelings towards one another. Only in mixed groups did they appear more sensitive to the perceptions of others.
"As soon as one of them (or both) is a woman, however, the situation changes," the authors describe.
"Then, likeability considerations become relevant, turning low likeability into a disruptive factor - in a sense an exogenous 'hurdle' - that impedes successful co-operation and reduces performance outcomes. Women always face this potential hurdle, men don't."
To figure this out, researchers split participants into pairs or groups to play games in which imaginary financial rewards were given for cooperation and coordination.
Beforehand, each participant was given a photograph of their teammates and asked to rank them based on their likeability at first sight. In a cruel twist, they were then told how they had been ranked by their teammates.
During one of the games, two partners were asked to contribute up to 6 euros to a joint investment. The greedy gamble was essentially to keep more for yourself and win off of your partner's sacrifice. In the end, the authors say men gave about 4.05 euros on average, while women gave about 3.92.
Still, that was for all groups. In single sex groups, the women who didn't like their partner much contributed 30 percent less on average than those who felt some mutual affinity. For men, liking one another made little difference - the average contributions were fairly similar.
Only when the teams were mixed, did men begin to show some vulnerability to this factor. When mutual likeability was low, men gave 50 percent less than if it was high, while women gave only 37 percent less.
During another game, where players try to choose the same number - women in all-female groups were willing to gamble much less when likeability was low. Whereas men in all-male groups went 'all in' no matter what the level of likeability was.
"We expected that there would be a meaningful gender difference in behaviour," economist Leonie Gerhards from the University of Hamburg, Germany, told Newsweek.
"However, we had not expected that this difference would be so stark."
This isn't to say there's something intrinsically different about men and women; they could simply be living up to cultural expectations.
Whatever the cause, the authors maintain it's a factor that has the ability to impede team cooperation, especially in all-female and mixed teams. They even go so far as to use it to explain at least a part of the gender wage gap.
"Aggregated over all rounds of both games and all teams, women earn on average 4.36 [percent] less than men in our experiment," the authors write.
"In same-sex teams, the gender pay gap is even larger and amounts to 7.75 [percent] lower earnings for women on average."
Previous research in economics and social psychology has also noted important gender differences in social interactions that might play into larger injustices. Some studies suggest, for instance, that likeability and success are negatively correlated for women, a conundrum known as the 'likeability trap'.
But while it certainly seems plausible that our concern for what other people think can influence our behaviour and cooperation at work, more research will be needed before we can claim any gender discrepancies with certainty.
Gerhards calls their study a mere 'hint' at that underlying truth.
The study was published in The Economic Journal.