Still in her pajamas on a weekend morning in March of 2006, criminologist Charis Kubrin read with interest an article about an idea foreign to many people: With more immigration may come a decrease, not an increase, in crime.

The opinion article, in the New York Times, was by sociologist Robert Sampson, and described his research in Chicago. He had found lower rates of violence among predominantly Mexican American communities with many recent immigrants than in Chicago communities of blacks and whites. First- and second-generation immigrants in the study were, respectively, 45 percent and 22 percent less likely to commit violent crimes than third-generation Americans, he wrote.

Kubrin, now a researcher at the University of California, Irvine, emailed Sampson to congratulate him on the piece and to ask for a copy of his academic paper. His answer surprised her: Hers was the first nice response he’d received.

Kubrin was taken aback. At the time, a good deal of research had rejected an oft-proposed link between immigrants and increased crime, finding instead that the concentration of immigrants in a community had no effect or even correlated with slightly reduced crime rates. But many email responses Sampson had gotten — some vitriolic — insisted, despite the data, that more immigrants meant more crime.

Kubrin stayed at her computer all day, reviewing the recent research on immigration and crime as well as data on the public’s perception of the issue. The two views seemed far apart. By the evening, she had decided to get involved. “I naively thought that we need more studies and better studies,” she says.

Kubrin started by expanding on Sampson’s work, designing studies to examine the relationship between immigrant concentration and crime in cities around the country. After publishing some of those studies, she was invited by the US National Academy of Sciences to review the scientific literature dating back to the 1920s on the question of how immigration affects crime. The bulk of the evidence agreed with Sampson’s work: Immigration did not increase crime — if anything, it was linked to less crime.

Then Kubrin went further. She and researcher Graham Ousey of the College of William and Mary in Virginia performed a meta-analysis — pooling the results of about 50 studies with about 550 findings among them, and using statistical methods to number-crunch the data. The two spent four years coding details of the studies and running analyses.

“You do all this work and the conclusion is right where we thought we’d be based on our previous research. This, to me, is very definitive.” Charis Kubrin

Overall, the pair found a weak link between immigration and decreased crime. Most findings included in the meta-analysis showed no correlation between immigration and crime, but where effects existed, immigration was 2.5 times more likely to be linked to less crime than it was to be linked to more.

“You do all this work and the conclusion is right where we thought we’d be based on our previous research,” Kubrin says. “This, to me, is very definitive.”

But when she turns on the news, she often feels like the rest of the world isn’t listening.

The disconnect has motivated her to speak up. In the last few years, she’s made an effort to talk publicly about all of her research (she also has an extensive body of work on criminal justice reform and the use of rap lyrics as evidence in criminal court cases). She makes it a point to take media calls, looks for opportunities to share her work in local presentations and has published a number of opinion pieces in major news outlets.

Spurred by the tough stance on immigration that has been a focus of Donald Trump’s presidency, Kubrin recently doubled down on her resolve to share her data on immigration and crime — particularly with people the least likely to accept it. She’s been mostly pleased with the results of her efforts to escape “my echo-chamber,” as she puts it. She enjoyed a lively question-and-answer session with UC Irvine’s Conservative Student Union this past fall. She’s had meaningful email exchanges with people writing to question her assertions.

But some reactions have been unpleasant, even scary. She’s been harassed on Twitter. Critics have filled her inbox with hate mail and devoted blog posts to “how stupid I am,” Kubrin says. She keeps a folder titled “Victimization” on her computer where she saves messages that make her especially nervous: an email suggesting she should watch her back, for example, and a set of aggressive messages from the same writer that pinged her mailbox in quick succession.

In 2018, most US adults think undocumented immigrants are no more likely than US citizens to commit serious crimes, a PEW survey finds. But a strong minority disagrees. Those who disagree are most likely to be over 65 and conservative. 

Kubrin’s experiences aren’t unique. Science is a work in progress — findings are made, results are (or aren’t) reproduced, nuances reveal themselves, knowledge slowly builds. But even when the data converge on concrete conclusions, some people remain unconvinced, especially in cases where belief in a narrative becomes enmeshed with politics, ideology or fear.

Both sides of the political aisle can engage in this. Plant geneticist Pamela Ronald of the University of California, Davis, communicates about the benefits of genetic engineering for food security and sustainable agriculture while working to counter the misperception that genetically engineered crops are unsafe to eat. Ronald says she has been called a shill and been accused of lying for financial gain, and must turn over her emails every six months or so to activists who have used the Freedom of Information Act to target scientists at public universities.

Like Kubrin, Ronald keeps going. She wants the genuinely curious to be able to talk with scientists and doesn’t want her students to be discouraged from their own outreach projects down the road. For her trolls, she says, “that really is the goal — to prevent scientists from communicating. So I think it's important to continue.”

“Scientists will always sound like they’re waffling and all they’re doing is being respectful of the scientific method.” Paul Offit

Paul Offit, a Children’s Hospital of Philadelphia pediatrician specializing in infectious diseases and an expert on vaccines, immunology and virology, is also resolute in continuing to engage. Offit has long been the target of harassment for his efforts to dispel the notion, tested and refuted over and over again, that vaccines cause autism. He says that in some ways scientists’ training — to be cautious and nuanced and to qualify findings — is at odds with the ideal for clear public communication. Speaking scientifically, for example, no epidemiological study can prove cause and effect. And no study sample size can be infinite, so scientists are taught to never say never.

“That makes for weak language,” Offit says. “Scientists will always sound like they’re waffling and all they’re doing is being respectful of the scientific method.”

Sampson, whose opinion article inspired Kubrin’s research, also uses cautious language when talking about the science around immigration and crime. Though he firmly believes that the data show immigrants have not made America less safe, he is careful to not stray from conclusions he can draw directly from evidence.

And the data aren’t perfect, he adds. To track crime, researchers must rely on official crime reports, victimization surveys or confidential questionnaires that ask people whether they have committed violence or other criminal activities. All the methods have their shortcomings. Official reports may be skewed by a reluctance of undocumented or recent immigrants to go to the police (most US immigrants are in the country legally). And surveys and interviews depend upon the honesty of respondents; this, too, may vary in different communities based on factors such as fear of repercussions.

Some researchers have avoided these problems by focusing on homicides — crimes that rarely go unreported, Sampson says. Data from cities across the United States show that immigration and murder do not increase together.

It’s telling that the data from multiple sources converge despite the limitations of individual studies, says Sampson. Though he thinks criminologists should continue to refine immigration-crime studies, he says it’s pretty evident that research doesn’t support the claim that immigration increases crime. “For now, if you had to tell a policy maker what is the state of the art — that is, what is the best conclusion we can draw given the information that we have at hand — the answer is pretty clear,” he says.

Kubrin herself feels so confident in her findings on immigration and crime that “I’m bored,” she says. She’s ready to move on to deeper questions about why immigration might sometimes be associated with lower crime. She thinks it’s likely that immigrants may be self-selected for grit and determination. The supportive enclaves characteristic of many immigrant communities also might help people stay on the straight and narrow, she says, and family factors, such as differences in divorce rates, may play in as well.

She says it’s hard to know if she’s making much headway in convincing people through her outreach, though. “It just feels like trying to patch up a dam with a little Band-Aid,” she says. “It’s so difficult to get public perception to turn on this.”