A stereotype is when we make a generalization about the behaviours of a group and then assume that a member of that group will show that behaviour.  A stereotype is a way of interpreting the world around us and is a cognitive schema.  There are many theories as to how stereotypes are formed.  One theory is the social identity theory.  We form our stereotypes when compare our in-group to an out-group.  Another theory is that stereotyping is a cognitive bias called illusory correlation, when we see a correlation between a group and a behaviour that is not really there. 

Social identity theory says that our identity and behaviour is influenced by the groups we belong to.  Social identity theory says that we choose a group to belong to based on traits that we think are similar to our own.  This is called social categorization.  SIT argues that we divide the world into in-groups and out-groups.  We identify with our group and begin to conform to the norms of our group.  We also compare our in-group to the out-group.  We see our in-group as superior.  We also see the out-group as all sharing common traits.  This “out-group homogeneity” is the basis for stereotyping.  Often these traits are negative.  This is a way of justifying that our in-group is better than the out-group and raises our self-esteem

Hilliard and Liben wanted to see if making one’s in-group more salient would lead to the formation of stereotypes in children. Their study looked at two different preschools.  Both schools had a “gender neutral” policy. In both schools, the children did a pre-test to measure their level of gender stereotyping.  They then had the teachers in one school make children more aware of their gender by grouping them by sex, posting their work on different bulletin boards, and using gendered language like, “boys and girls.”  The other group was a control.  After two weeks, the researchers found that the children in the “high salience” condition had a significant increase in their gender stereotypes than the control.

The study was a true experiment, so it can show a cause and effect relationship between in-group salience and stereotype formation. The two schools were randomly assigned to conditions and the “gendered behaviour” of the teacher was manipulated by the researcher. The study was done in a naturalistic setting (school), so it has high ecological validity. The use of a pre-test/post-test allowed the researcher to see change over the two-week period.  However, since the children also went home, the internal validity is questionable.  It was not possible to control  for other variables that may have had an influence on the children.  However, it was only a two week period, so it is unlikely that such variables would have a significant effect on the children.  Also, since the children’s parents had chosen to put their children in a gender-free school, it is likely that their parents would not have an effect on the development of these stereotypes.

Illusory correlation is a cognitive bias.  As humans, we always try to explain “why” something happens.  This means that we often see a cause and effect relationship where there isn’t one.  Hamilton and Gifford wanted to see if knowing that a group was a minority would have an effect on stereotyping.  The researchers should a group of American university students a series of slides.  Each slide had a positive or negative behaviour for a member of group A or B.  There were twice as many comments for A than B, but the proportion of good to bad behaviours was the same.  When the participants were asked to rate the traits of each group, they said that A was more positive. They also recalled more negative traits for group B than for group A.  The researchers argued that since the group is smaller, negative behaviours are more distinctive – and we assume that this characteristic of the group.

The study manipulates the IV (size of the group), so the study is an experiment.  The researchers used A and B for the two groups to avoid any previous stereotypes influencing the participants.  This control helps to eliminate extraneous variables.  However, the study is very artificial and does not reflect how we might make stereotypes in the real world.  Often stereotypes are developed by direct interaction with people, not from just hearing a statement about them.  In addition, one’s emotions, intelligence, or even social identity may play a key role in the formation of these illusory correlations. Finally, it could be that the fact that they were told it was a smaller group had an effect  on the participants simply because of the connotation of a "minority."  When they did the experiment again and didn't tell participants that group B was a minority, the illusory correlation was not observed.

Studying the formation of stereotypes is difficult to do in a naturalistic setting. There are too many extraneous variables that would lower the internal validity of the research.  It is also not possible to “see” a person’s thinking. Psychologists have to make assumptions about the process which resulted in a stereotype.  It is also difficult to measure someone’s “in-group bias” or the level of “salience” – even in a lab situation.  However, stereotyping seems to be a natural cognitive process.  In order for us to simplify our understanding of the world around us, we create schema that generalize about a group. So, when talking about “Americans”, everyone has a general sense of who we are talking about.  This oversimplification of the world means that we have to consciously try not to stereotype and to recognize people’s individuality.