A vendor adjusts a pinata in the image of Republican presidential candidate Donald Trump on March 2 in Tijuana, Mexico. Trump has angered many Mexicans with his campaign rhetoric denigrating some immigrants as “rapists” who bring crime and drugs to the United States, and his promise to build a wall along the U.S.-Mexico border. (Gregory Bull/Associated Press)

Donald Trump’s famous rants against Mexican migrants as drug smugglers, criminals and rapists, and his promise to build a wall keeping them out, have apparently helped him become the presumptive Republican nominee. Like other opponents of immigration reform, he insists that his views stem from principles of fairness to aspiring immigrants who “play by the rules” or concerns about national security. But his ability to rouse supporters by talking about immigration in explicitly ethnic terms suggests that many white Americans’ opinions about immigration policy are motivated by unacknowledged prejudice against Latinos, as some research has suggested.

Our research finds that reality is more complex.

In a recent study, we tried to determine what is really behind many white Americans’ support for “getting tough” on illegal immigration. Do they oppose offering legal status to illegal immigrants because they fundamentally dislike Latinos as a group and reject policy reforms that would benefit them? Or do they tend to stereotype illegal Latino immigrants as unassimilated or lacking skills they need to succeed in the United States, such as speaking English or holding a steady job — and oppose legalization because they are concerned about how it would affect the culture or the U.S. economy?

We tested this with a survey about immigration and “Juan,” “Yuan” or “Johan.”

In June 2015, we conducted an online survey of non-Hispanic white voters in California through the firm Survey Sampling International. Respondents were randomly assigned to one of several experimental groups. All the groups read a short vignette about a hypothetical legalization program, much like those that have been debated in the past few years.

Then, after reading the vignette, respondents were asked whether a particular illegal immigrant should be included in the legalization program. About one-third of respondents were asked about a Mexican immigrant named “Juan.” Another third were asked about a Chinese immigrant named “Yuan.” The rest were asked about a German immigrant named “Johan.”

The experiment also included a second manipulation. All our respondents were told that the immigrant they read about had lived in the United States for two years. Half of them were also told that he speaks English and has worked steadily as a waiter.

If people opposed legalization because they dislike Latinos, we would expect respondents to favor the German and Chinese immigrants over the Mexican immigrant, regardless of his job status or language skills.

In fact, the additional information might even make anti-Latino bias worse. If people disliked Latinos, they might discount positive information and emphasize negative information about Juan, a well-known psychological process known as “motivated reasoning” – and would continue to favor offering legal status to Yuan and Johan but not Juan.

The campaign of the presumptive Republican nominee, Donald Trump, is sparking a surge in the number of citizenship applications and voter registrations among Hispanics fearful of his immigration policies. Since January, California alone has seen a boost of 218 percent in Democratic registration and among Hispanics, registration is up 123 percent. (Alice Li/The Washington Post)

Anti-Latino bias disappeared once our hypothetical immigrant could speak English and hold a job

But the immigrant’s work record and English fluency actually did make respondents as sympathetic to the Mexican immigrant as they were to the Chinese and German immigrant.

In the graph below, we show the percentage of respondents in each condition who said the immigrant they read about should be allowed into the legalization program.

When respondents learned nothing about the immigrant’s language skills or job status, they were indeed biased against the Latino. Compared with respondents who read about Yuan (red line) or Johan (green line), those who read about Juan (blue line) were roughly seven to eight percentage points less likely to say he belonged in the legalization program.

But that gap disappeared when we told our respondents that Juan spoke English and had been working for two years. The anti-Latino bias simply evaporated leaving those supporting legalization for all three immigrants statistically indistinguishable – 83 percent for Juan, compared with 81 percent for Yuan and 82 percent for Johan.

So why do so many white Americans oppose legalizing undocumented immigrants?

Discrimination against Latinos may grow not from hostility against an ethnic “outgroup,” but rather stereotypes about whether they will contribute to the United States or become a burden. In the absence of other information, whites in our sample rely on ethnic cues to “fill in the blanks” — assuming undocumented Latinos are uneducated, unassimilated and potential financial problems for U.S. society.

Of course, this is one experiment. But research seems to support our interpretation. Comparable experiments that don’t give any cues about skills, language and integration tend to find a lot of bias against Latinos. But studies that give clues about immigrants’ education, employment, legal status and assimilation often find little or no anti-Latino bias. Other studies find similar beliefs that Latino immigrants are here illegally and tend to be less skilled than others.

What can immigration reformers learn from these results?

What does this mean for those who hope to build public support for comprehensive immigration reform — including a pathway to citizenship for the millions of immigrants illegally living in the United States?

On the one hand, negative racial imagery may anchor resistance, furnishing a ready-made portrait of “the bad immigrants” that opponents can activate at will.

On the other hand, polling repeatedly shows that most Americans and many Republicans — even some of the people supporting Trump — are open to legalization programs, if those programs were to include stringent enforcement and such conditions as paying a fine, passing an English test and showing a solid employment record.

In other words, whites are open to policies that address the negative group stereotypes — and ensure that new U.S. citizens could genuinely join and contribute to the community.

But immigration reform proponents must still deal with the fact that illegal immigration and Latino ethnicity are closely linked in Americans’ minds. If the media were to cover Latinos in other contexts — not just in relation to immigration — the public might think less about immigration in ethnic categories, and therefore less likely to succumb to demagogues’ ethnic scapegoating.

Morris Levy is an assistant professor of political science at the University of Southern California.

Matthew Wright is an assistant professor of government in American University’s School of Public Affairs.