In the last several years, growing popularity of anti-immigration policies—ranging from public support for building a wall between the U.S. and Mexico to  the outcome of the Brexit vote in the United Kingdom—has been largely attributed to native-born Whites feeling threatened by a rise in ethnic diversity across the Western world. But to what extent does research on diversity as a whole support the idea that more diversity leads to backlash from Whites?

Evidence from social science seems to both refute and support the idea that greater ethnic diversity leads to increased feelings of threat amongst Whites. Proponents of “contact theory” claim that increasing levels of diversity within a community actually reduces prejudice. When Whites are able to build personal relationships with people belonging to a particular ethnic group, and get to know them as close friends or neighbors, contact theory predicts that they will develop a more positive view of members of that ethnic group. However, other studies, such as a famous paper by Harvard’s Robert Putnam, seem to support “threat theory”: as Whites have increasing contact with members of other ethnic groups, they tend to withdraw from their communities and become less trusting of people belonging to other ethnicities.

Until now, these two theories have competed with each other, leaving a lack of clarity about the degree to which ethnic diversity truly leads Whites to feel more threatened. A recently published paper by political scientists Eric Kaufmann and Matthew Goodwin attempts to reconcile these two theories and determine conclusively which one is actually correct. They found that the relationship between ethnic diversity and feelings of threats amongst native-born Whites is mixed: for small communities, contact seems to reduce threat, while in larger communities, more diversity seems to increase it. This suggests that increasing diversity by itself does not necessarily lead to greater wariness amongst Whites; diversity interacts with the size of a community in producing threatened feelings.

The researchers conducted a meta-analysis: a statistical method that allows scientists to examine the accuracy of a hypothesis using hundreds studies. Meta-analyses have proven especially useful in the field of medicine where individual studies about the beneficial or harmful effects of a particular food or drug tend to yield contradictory results, making it difficult for doctors and the public to adhere to certain guidelines. Kaufmann and Goodwin’s meta-analysis drew on 171 studies, all conducted after 1995, and all of which looked at the impact of ethnic diversity on support for immigration.

They divided up the studies according to a number of characteristics, including the size of the geographic area that was under examination. Doing this revealed the importance of geographic size for understanding the impact of diversity on perceptions of threat. They found that when ethnic diversity increased in relatively small communities, such as those consisting of between 5,000 and 10,000 people, the increasing diversity seemed to reduce perceptions of threat amongst native-born Whites. However, studies that looked at the impact of increased diversity in large, metropolitan areas (i.e. between 50,000 to 500,000 people) found that as ethnic diversity increases, Whites report greater opposition to immigration policies and tend to give more support to political parties that are anti-immigration.

Perhaps contact theory is only true in smaller sized neighborhoods because that is where people are most likely to develop close, personal relationships with people who have different backgrounds than themselves. In larger cities, people may be more likely to segregate themselves into their own racial groups. Or perhaps the largeness of a community inherently reduces people’s ability to trust strangers who appear different than themselves. However, this is all speculation, as the study cannot offer a definitive explanation as to what is causing the findings. To answer that question, the researchers plan to undertake additional research that looks more deeply at the conditions under which diversity either does or does not produce a sense of threat amongst Whites.

In the meantime, the findings have important implications for the immigration debate in countries across the world. For Whites in larger communities, being in relatively close proximity to ethnically diverse neighbors is probably not sufficient to reduce their feelings of threat towards immigration. For one thing, meaningful interracial contact is often hard to come by due to high levels of residential segregation in most communities. For example, Kaufmann and Goodwin point out how in England and Wales, most Whites live in communities that are overwhelmingly comprised of other Whites. It also seems unlikely that things will naturally change over time even though younger generations tend to espouse more support for immigration. This is because longitudinal studies suggest that as people age, they tend to shift their political attitudes towards becoming more conservative towards immigration. These realities make it practically very difficult to change people’s attitudes towards diversity.

To create real change, concerted efforts would need to be made to desegregate people and increase the amount of personal contact they have with people who belong to ethnic groups different than their own. But such efforts would not be easy given that most of us tend to naturally prefer and gravitate towards people who are similar to ourselves. Therefore, it seems likely that debates about immigration are likely to rage on, especially as levels of ethnic diversity continue to increase across major cities all over the world.