Verified by 
“An ethnic stereotype is a generalization made about an ethnic group, concerning a trait attribution, which is considered to be unjustified by an observer.” John Brigham (1971).
This is my favorite of all definitions (and it need not be restricted to ethnic stereotypes), because it exquisitely captures how laypeople and scholars typically use the word “stereotype”: as a damning indictment of someone else’s beliefs about a group. My beliefs are reasonable, rational, and appropriate – yours, at least when they differ from mine, are mere stereotypes. This definition conveys why social scientists perform research documenting cultural, ethnic, social class, racial, or ethnic differences, and, at the same time, condemn other people’s beliefs about group differences as irrational “stereotypes” steeped in bigotry. It helps us understand why proponents of multiculturalism believe that they know the “truth” about which group differences are important to understand and respect, and at the same time, rail against other people’s inaccurate stereotypes. Indeed, one of the easiest and most effective ways for Person A to derogate and dismiss Person B’s claims about a group is to say, “But that is just a stereotype.” Person B, now implicitly accused of being an “ist” of some sort, is most likely to just shut up and go away, and even if s/he doesn’t, has been discredited anyway.
So, this is a perfect definition that captures the psychological essence of how people use the term "stereotype." Unfortunately, however, it fails as a scientific definition, precisely because it is purely subjective. Its subjectivity leads it to logical incoherence: if I think your belief is accurate, then your belief is not a stereotype; but if someone else believes your belief is inaccurate, then your exact same belief is a stereotype. Subjectively, this is possible, but scientifically, it is impossible for something to simultaneously be and not be a stereotype. So, as a scientific definition, as amusingly beautiful as this definition is, it fails.
Nonetheless, what it captures psychologically is terribly important. One can see this in some of the comments on my prior blog post (http://www.psychologytoday.com/blog/rabble-rouser/201210/stereotype-ina…). Consider this comment from Dominic Amann, "For example -- what good does it do me to assume that all Asians are bad drivers?" Dominic is not alone in making this type of argument. It was made in 1991 by nothing less than the American Psychological Association (page 1063): "Once an individual is classified as a member of a social group … people tend to perceive members of the other group as all alike or to expect them to be all alike, which they never are." It has been made by many other scientists as well (Jussim, 2012).
Do people "perceive members of other groups as all alike?" Do people perceive "all Asians as bad drivers?" These are, as scientists say, "empirical questions." That means they can be answered by data. Check that. In this particular case, they have been answered by data. Although psychologists have been studying stereotypes for nearly 100 years, I have yet to identify a single empirical study that has identified a single person who holds a belief of the form "All of THEM are X" -- and you can fill in any THEM you like (race, sex, class, age, occupation, and many more) and any X you like (criminal, warm, intelligent, etc.). I have been laying this challenge down for over 20 years at conferences, in talks, and in print: "If you can identify a single study identifying a single such person, please let me know. And if you cannot, then please stop making these outrageous claims." In all that time, not a single person -- layperson or scientist -- has come forward to answer this challenge. It is repeated here. If you can identify a single such study, please provide the reference in a comment on this post.
I do not doubt that, in a world with over six billion people, somewhere there are some people who really do believe All of Them are X. They probably can be found disproportionately in hate groups, such as the Klan, neo-Nazis, and various ideological, religious, and nationalist extremist groups. But they are few and far between.
Undoubtedly stereotypes can lead to biases. Sometimes, they lead people to confuse the behaviors and characteristics of one individual from a group with those of another. For example, sometimes, I misremember which of my three children committed some act. The idea that this constitutes "perceiving my children as all alike," however, is absurd.
Many social science perspectives on stereotypes are exaggerated, inaccurate, rigidly resistant to change in the face of relentless disconfirming evidence, and maintain their conclusions by virtue of a very selective focus on studies and findings that confirm the a priori belief in the irrationality and badness of stereotypes. It is hard not to see some of this as irrational; some clearly is logically incoherent. In other words, stereotypes are routinely stereotyped by both laypeople and scientists!
References
American Psychological Association. (1991). In the Supreme Court of the United States: Price Waterhouse v. Ann B. Hopkins (Amicus curiae brief). American Psychologist, 46, 1061-1070.
Brigham, J. C. (1971). Ethnic stereotypes. Psychological Bulletin, 76, 15-38.
Jussim, L. (2012). Social perception and social reality: Why accuracy dominates bias and self-fulfilling prophecy. New York: Oxford University Press.
