There is a paper out, the abstract of which says:

...Second, respondents significantly underestimated the proportion of [group X] among their colleagues. Third, [members of group X] fear negative consequences of revealing their ... beliefs to their colleagues. Finally, they are right to do so: In decisions ranging from paper reviews to hiring, many ... said that they would discriminate against openly [group X] colleagues. The more [group anti-X] respondents were, the more they said they would discriminate.

Before you go look at the link, any guesses as to what the [group X] is? X-/