Stereotypes can be damaging, so how can we stop children picking them up?

How can modern parents raise the next generation to be free from corrosive gender and racial stereotypes? By the time children start elementary school, gender and race shape their lives in many ways that parents might want to prevent. As early as first grade, girls are less likely than boys to think members of their own gender are “really, really smart.” And by just age three, white children in the United States implicitly endorse stereotypes that African-American faces are angrier than white faces.

These stereotypes go deeper than children’s beliefs – they can also shape a child’s behavior. By age six, girls are less likely than boys to choose activities that seem to require them to be really smart, which could contribute to the development of long-term gender differences in science and math achievement.

Why do stereotypes develop in such young children? As a professor of early cognitive and social development, I have seen my research reveal how surprisingly subtle features of language contribute to a child’s tendency to view the world through the lens of social stereotypes.

The problem of generalization

Many parents try to prevent the development of stereotypes in children by avoiding saying things like, “boys are good at math,” or “girls cannot be leaders.” Instead, parents might take care to say things that are positive, like “girls can be anything they want.”

But our research has found that, to the developing mind, even these positive statements can have negative consequences.

For young children, how we speak is often more important than what we say. Generalizations, even if they say only things that are positive or neutral, such as “Girls can be anything they want,” “Hispanics live in the Bronx” or “Muslims eat different foods,” communicate that we can tell what someone is like just by knowing her gender, ethnicity or religion.

In our research, published in Child Development, we found that hearing generalizations led children as young as two years old to assume that groups mark stable and important differences between individual people.

In this study, children were introduced to a new, made-up way of categorizing people: “Zarpies.” If they only heard statements about specific individuals, (e.g., “These Zarpies whisper when they talk”), children continued to treat the people as individuals, even though they were all marked by the same label and wore similar clothes. But if they heard the same information as a generalization (e.g., “Zarpies whisper when they talk”), they started to think that “Zarpies” are very different from everyone else. Hearing generalizations led children to think that being a member of the group determined what the members would be like.

In another recent study, we found that hearing these types of generalizations – even if none of them was negative – led five-year-old children to share fewer resources (in this case, colorful stickers) with members outside their own social group.

These findings show that hearing generalizations, even positive or neutral ones, contributes to the tendency to view the world through the lens of social stereotypes. It is the form of the sentence, not exactly what it says, that matters to young children.

From groups to individuals

Our research means that generalizations are problematic even if children do not understand them.

If a young child overhears, “Muslims are terrorists,” the child might not know what it means to be a Muslim or a terrorist. But the child can still learn something problematic – that Muslims, whoever they are, are a distinct kind of person. That it is possible to make assumptions about what someone is like just by knowing if they are Muslim or not.

Language that uses specifics – instead of making general claims – avoids these problems. Sentences like, “Her family is Hispanic and lives in the Bronx,” “This Muslim family eats different foods,” “Those girls are great at math,” “You can be anything you want,” all avoid making general claims about groups.

Using specific language can also teach children to challenge their own and others’ generalizations. My three-year-old recently announced that “Boys play guitar,” despite knowing many female guitar players. This troubled me, not because it matters very much what he thinks about guitar playing, but because this way of talking means that he is starting to think that gender determines what a person can do.

But there is a very easy and natural way to respond to statements like these, which our research suggests reduces stereotyping. Simply say, “Oh? Who are you thinking of? Who did you see play the guitar?” Children usually have someone in mind. “Yes, that man at the restaurant played the guitar tonight. And yes, so does Grandpa.” This response guides children to think in terms of individuals, instead of groups.

This approach works for more sensitive generalizations too – things a child might say, like “Big boys are mean,” or “Muslims wear funny clothes.” Parents can ask children who they are thinking of and discuss whatever specific incident they have in mind. Sometimes children speak this way because they are testing out whether drawing a generalization is sensible. By bringing them back to the specific incident, we communicate to them that it is not.

Every interaction counts

How much can this small change in language really matter? Parents, teachers and other caring adults cannot control everything that children hear, and exposure to explicitly racist, sexist or xenophobic ideas can also influence a child’s view of societal norms and values.

But children develop their sense of the world through minute-by-minute conversations with important adults in their lives. These adults have powerful platforms with their children. As parents and caregivers, we can use our language carefully to help children learn to view themselves and others as individuals, free to choose their own paths. With our language, we can help children develop habits of mind that challenge, rather than endorse, stereotyped views of the people around us.