Back story: Using data to understand who gets asylum

Back story: Using data to understand who gets asylum


“Returned” is a multi-part series launched by the Union-Tribune this year that investigates the U.S. asylum system. The first installment, published in February, told the story of a woman from Nicaragua who is still waiting to find out whether she’ll be protected by the United States. This is the second installment, which explores disparities in outcomes in the system through a year-long analysis of a decade of immigration court records.

Immigration reporter Kate Morrissey and Watchdog and Data reporter Lauryn Schroeder discuss their data analysis and the interactive components of the project.

Q: Why did you decide to do this story?

A: Kate had been thinking for a long time about doing an analysis of immigration judge work histories using data from the Transactional Records Access Clearinghouse, a Syracuse University-based research group that is cited frequently in immigration reporting. But then the Trump administration changed the way grant rates are calculated so that TRAC used one denominator while the agency responsible for immigration courts used another, making a story based on statistics from the two confusing and difficult to write. After we were able to get a copy of the database used to track immigration court records, we decided it would be better to do that analysis ourselves so that we could be transparent about our approach and more fully understand what is happening in the system. Because the dataset is so complicated and large, Lauryn got involved early on to use a special kind of code to read and analyze the data.

Q: Did you find anything that surprised you that didn’t make it into the story?

A: Unlike with deportation rates, the help of an attorney did not seem to have much of an impact on an asylum seeker’s chances at actually winning asylum, according to our analysis. This is likely because of the time period that we examined. During the final years of the Obama administration, officials put policies in place that allowed some asylum seekers to close their immigration cases without having judges make decisions on the merits of their applications, and people with attorneys appear to have been much more likely to go for those options. Those options often allow asylum seekers to stay in the United States. Earlier studies that used time periods prior to those policies have found that attorneys do have a large impact on an asylum seeker’s chances of success.

Q: Was there anything you wanted to find out from the data and couldn’t?

A: We had hoped to be able to narrow down our analysis exclusively to people who had requested asylum when they arrived at the border, since most recent changes to the asylum system are intended to deter those asylum seekers in particular. However, the information that the government tracks about each case did not allow us to do that completely. It would’ve also been helpful and interesting to look at race, age and gender in our analysis, but that information was also not available.

Q: What were some of the challenges in putting together a data story like this?

A: To start, the data files are massive — more than 50 million rows in all. On top of that, we found errors and inconsistencies, including misaligned rows and columns, and uninterpretable entries in incorrect fields. Before any analysis could take place, we spent weeks working with statisticians and immigration experts to mold the tables and get case information in an understandable format.

Q: You created an online interactive simulation that invites readers to experience — in a small way — what it’s like to go through the asylum system. What do you hope readers will take away from it?

A: With lots of help from our colleague Ruby Gaviola, we created an experience that asks readers to make many of the same decisions that an asylum maker would make. We hope to help readers understand how much happenstance — where the cousin who has promised to take care of you lives in the U.S. or which detention center happens to have bed space — as well as a bit of random chance, play a large role in whether you end up in front of a judge who is likely to order you deported or one who is likely to grant you asylum. Of course, many other factors — including what happened to you in your home country, how clearly you can explain what happened to you, whether you have a lawyer, whether that lawyer has a lot of asylum experience and which attorney the government assigns to oppose your case — can all play some role in whether you stay in the United States.

Q: What can we expect from the next installment in the series?

A: Part III will look more specifically at asylum claims from Honduras, particularly at cases that have long fallen into a gray area of asylum law — people whose lives are in danger but who do not fit neatly into the legal definition of refugee.