This page includes a section that was inappropriately cut off from a reading on
electronic reserve.  The reading is a selection from an article titled simply
Accuracy by me (Dr. Jussim), and it appears in section III.

The cut off section starts about halfway down the first paragraph
below.


    Stereotype accuracy and level of analysis: Conclusion.  Claims suggesting that stereotypes are inaccurate because they do not apply to all individual members of a group (Allport, 1954; APA, 1990; Fiske, 1998; Hamilton, et al., 1990; Nelson, 2002; Stangor, 1995) are both true and false.  The claim that stereotypes cannot possibly apply to all individual members of a group is completely true.  The suggestion that this renders stereotypes inaccurate is, however, unjustified because it confounds two levels of analysis (population and individual).  A claim about a population cannot be evaluated against the characteristics of an individual, or even subsets of individuals.  Consistency between the level of the perception and the level of the criterion must be maintained when assessing accuracy by comparing: beliefs about populations (stereotypes) to characteristics of those population groups; beliefs about differences between small groups of individuals to the actual differences between those small groups of individuals; and beliefs about an individual to the characteristics of that individual.  
    The one exception: Absolutist stereotypes.  Absolutist stereotypes -- beliefs that all members of a group have some attribute -- will indeed almost always be false, because there are almost always wide variations among individuals.  A single exception invalidates an absolutist belief.  Just as a belief that the temperature in all locations in Alaska is always below freezing will be disconfirmed by a single reading of 33 degrees fahrenheit in Juneau on July 15th at 1pm, a belief that all Germans are efficient will be disconfirmed by discovery of a single inefficient German.  
    The vast accumulated empirical evidence on stereotypes, however, has yet to report a single person who holds absolutist stereotypes.  Instead, the evidence indicates that most stereotypes are quantitative and probabilistic, not absolute (e.g., Judd, et al., 1995; McCauley & Stitt, 1978; Swim, 1994).  Probabilistic stereotypes, which permit many exceptions and wide variability, can only be evaluated by comparison to population-level criteria.  People who hold absolutist stereotypes undoubtedly exist, and probably comprise significant portions of extremist groups such as the Ku Klux Klan and neo-Nazis.  Nonetheless, such people are atypical of the participants in most scientific research on stereotypes.