Brief 12: Attitudes Towards Immigrants Among Boston-area Commuters

The author conducted a field experiment in which native Spanish speakers were placed at train stations in homogeneously white suburbs.

Link to Full Study

Category: Human Rights

Date of Publication: Friday, March 20, 2015

EGAP Researcher: Ryan Enos

PDF: Brief12BostonEnosTrainsexperiment1.pdf

Click to Download the Data

Geographical Region: North America

Research Question:

Does intergroup contact cause exclusionary attitudes?

Preparer: Seth Ariel Green



Social scientists have long been interested in the question of whether exposure to certain out-groups (racial, ethnic, gay and lesbian, handicapped, etc.) increases or decreases prejudice. Pettigrew and Tropp (2006) analyzed 713 independent samples from 515 studies on the subject, and found that intergroup contact “typically reduces prejudice.” Paluck and Green (2009), however, note that of the hundreds of studies that measure prejudice reduction techniques, only a small number are randomized controlled trials, the gold standard of scientific research. Enos’s study is a randomized controlled trial.

The Boston suburb communities in which Enos surveyed train-riders were predominantly white, with an average of 4.4% of Hispanic residents in 2010. His sample was commuters from nine separate train stations.

Research Design:

In each of the nine train stations Enos studied, he chose two rush hour trains that were close in time and randomly assigned one to treatment and one to control. Prior to the intervention, Enos distributed surveys asking commuters a variety of questions, including three specifically about immigration and nationhood: 1) Should immigration be increased, decreased or kept the same? 2) Should illegal immigrants without criminal records be allowed to stay in the United States? and 3) should English be declared the national language of the United States? He then hired pairs of native Spanish speakers, blind to the purpose of the experiment, to simply wait on the platform prior to treatment trains’ arrival, without specific instructions about speaking to or interacting with fellow commuters. They did this for ten working days.

After three days, Enos randomly picked half of those who had filled out initial surveys to fill out the same survey again. After ten days, Enos solicited the other half to fill out the same survey as well. He compared survey responses both between the two groups in treatment and also to a group that had not been exposed to Spanish-speaking individuals.


On all three questions, commuters who were exposed to the Spanish-speakers were more likely to think that immigration should be decreased, that illegal immigrants should not be granted citizenship, and that English should be the national language. The effects were especially large, and statistically significant, on the first two questions, while on the third, the effects were smaller, but most people in both conditions agreed that English should be made the official language prior to the treatment. When Enos limited his analysis just to respondents who reported that they waited for their train on the platform rather than in their cars, the results were stronger. For those surveyed after 10 days of treatment, however, effects were less pronounced, although all still in an exclusionary direction. Enos writes that this suggests that people grew more comfortable with the presence of Hispanics.

This shift in exclusionary attitudes was not accompanied by a shift towards more conservative politics generally.

Graph 1: Time Effects

Legend: Solid circle: 3-day treatment
Open circle: 10-day treatment
Policy Implications:

The clearest implication of this research is that exposure to immigrants will cause an exclusionary reaction and that anti-immigrant attitudes are likely not the result of generally conservative political orientations.  This implies that as immigration in the United States increases beyond traditional immigrant communities, a general backlash may be expected.    However prejudiced attitudes tend to diminish after repeated exposure, so when considering the issue of immigration in the United States, policy-makers and legislators ought to consider this study’s evidence that initial hostility does not necessarily equate to long-term problems, and that attitudes tend to mellow over time. Therefore, policies that encourage interpersonal contact and interaction should be encouraged in order to increase harmony.

Another interesting finding is that merely hearing people speak in Spanish, without any interpersonal contact, changes attitudes. For policy-makers, it is worth remembering that integration interventions that might seem unobtrusive may nonetheless activate prejudiced attitudes.


Enos, Ryan, 2014. “Causal Effect of intergroup contact on exclusionary attitudes.” Proceedings of the National Academy of Science, vol. 111, no.10 pp. 3699-3704

Paluck, Elizabeth Levy, and Green, Donald P., 2009. “Prejudice Reduction: What Works? A Review and Assessment of Research and Practice.” The Annual Review of Psychology, 60:339-67

Pettigrew, Thomas F., and Tropp, Linda R., 2006. “A Meta-Analytic Test of Intergroup Contact Theory. Journal of Personality and Social Psychology. Vol. 90, No. 5, pp. 751-783