I suppose the classic example from the history of science is the way that people with different ethnic backgrounds scored in relatively predictable ways in an IQ test. Here the stereotype, which definitely isn't true, was that people of a particular ethnic background were more intelligent than others. However, this wasn't what the test actually showed. What it showed was that people of a particular ethnic background were better at doing IQ tests than others. This definitely was true, but some still considered that unacceptable, considering it to be the application of a racial stereotype. It wasn't. All it was saying is that the test was designed with a certain cultural background in mind and someone with that background would do better. The danger here is the knee-jerk response that if a statement fits a group that is often stereotyped, then that statement must be false, offensive, lazy and disgusting.
I ought really to take one step back and ask what a stereotype is. According to my dictionary it is 'A preconceived and oversimplified idea of the characteristics which typify a person, situation, etc.; an attitude based on such a preconception.' So, by implication a stereotype can't be totally true, because it is oversimplified. So it would be a stereotype to say that all people with naturally red hair (to choose a minority I am a) a member of and b) who those who defend minorities don't care about) burn easily in the sun. I would suggest, however, that it would not be a stereotype to say that most people with naturally red hair burn easily in the sun, because there are good genetic reasons why this is likely to be the case, and because there is reasonable experiential evidence for this to be true.
In the case of my correction, I had said that chocolate seems to have a particular appeal for female consumers. Now, I would say that similarly, it would be a stereotype to say 'all women love chocolate', but that to say 'seems to have a particular appeal' is not a stereotype, because it describes the appearance - and I would challenge anyone to give me evidence that this is not the case. As a simple example, in last week's episode of Broadchurch, Olivia Colman's character said to her son 'I love you more than chocolate.' There's an obvious implication there.
So the stereotype would be 'all women love chocolate' or 'women love chocolate more than men do.' But I think it is entirely factual to suggest that women express the appeal of chocolate more than men do, just as, for instance, more men express the appeal of football than do women. The whole point is that this is not a generalisation. There are men who don't like football - I'm one. For that matter I have a male friend who goes on about his liking for chocolate. But we are exceptions. Expressing the particular appeal of football is primarily a male activity, just as making comments like Coleman's character does about chocolate is primarily a female activity, is not a stereotype, it is a fact.
So then we have to ask, is it okay to suppress facts because we don't like them? Well, it's certainly not a scientific thing to do. It can be a social thing to do - it's called a white lie, and I can only think this is why some people get het up about this kind of thing - but I'm really not sure that white lies have a place in science.