Just because a stereotype is a negative one, doesn't mean it isn't accurate.
A stereotype is simply an observation which turns into a generalization about a group and observation and generalization is the foundation of all knowledge.
Study here about how accurate stereotypes are
Some supplement articles about the accuracy of stereotypes
There are many different ways to test for the accuracy of stereotypes, because there are many different types or aspects of accuracy. However, one type is quite simple -- the correspondence of stereotype beliefs with criteria. If I believe 60% of adult women are over 5' 4" tall, and 56% voted for the Democrat in the last Presidential election, and that 35% of all adult women have college degrees, how well do my beliefs correspond to the actual probabilities? One can do this sort of thing for many different types of groups.
And lots of scientists have. And you know what they found? That stereotype accuracy -- the correspondence of stereotype beliefs with criteria -- is one of the largest relationships in all of social psychology. The correlations of stereotypes with criteria range from .4 to over .9, and average almost .8 for cultural stereotypes (the correlation of beliefs that are widely shared with criteria) and.5 for personal stereotypes (the correlation of one individual's stereotypes with criteria, averaged over lots of individuals). The average effect in social psychology is about .20. Stereotypes are more valid than most social psychological hypotheses.
What people call “stereotypes” are what scientists call “empirical generalizations,” and they are the foundation of scientific theory. That’s what scientists do; they make generalizations. Many stereotypes are empirical generalizations with a statistical basis and thus on average tend to be true. If they are not true, they wouldn’t be stereotypes. The only problem with stereotypes and empirical generalizations is that they are not always true for all individual cases. They are generalizations, not invariant laws. There are always individual exceptions to stereotypes and empirical generalizations. The danger lies in applying the empirical generalizations to individual cases, which may or may not be exceptions. But these individual exceptions do not invalidate the generalizations.
An observation, if true, becomes an empirical generalization until someone objects to it, and then it becomes a stereotype. For example, the statement “Men are taller than women” is an empirical generalization. It is in general true, but there are individual exceptions. There are many men who are shorter than the average woman, and there are many women who are taller than the average man, but these exceptions do not make the generalization untrue. Men on average are taller than women in every human society (and, by the way, there are evolutionary psychological explanations for this phenomenon, known as the sexual dimorphism in size, but that’s perhaps for a future post). Everybody knows this, but nobody calls it a stereotype because it is not unkind to anybody. Men in general like being taller than women, and women in general like being shorter than men.
But we agree that being impervious to data is a bad thing, right? Liberals routinely rail against conservatives' supposedly anti-scientific stands, right? Liberals, in sharp contrast, don't ever oppose data and science, do they?