Naturalization disparities found in new Gould study

Significant group disparities in naturalization — the process where individuals born outside of the United States become U.S. citizens — based on applicants’ race, ethnicity, gender and religion were found in an empirical study published March 12. The study was co-authored by Emily Ryo, a professor of law and sociology at the Gould School of Law, and Reed Humphrey, a USC alumnus and Ph.D. candidate at the University of Washington.

In their investigation of naturalization applications decided by the U.S. Citizenship and Immigration Services between October 2014 and March 2018, researchers found that non-white and Hispanic applicants are less likely to be approved than non-Hispanic applicants. The study also found that male applicants are less likely to be approved than female applicants, and applicants from Muslim-majority countries are less likely to be approved than applicants from other countries.

Further, research data demonstrates that different factors, including race, gender and religion, combine to produce a certain group hierarchy in terms of approval probabilities.

The study found the probability of approval for Black men is 89.27%, compared to that of white women at 94.15%. The probability of approval for Black individuals from Muslim-majority countries is around 86.43%, compared to 95.56% for white individuals from non-Muslim majority countries.

Ryo’s inquiry into records of naturalization application denials revealed that a considerable portion of administrative data for the category is missing.

“What really surprised me was just how scarce, or maybe I should say, nonexistent, administrative data was in this area,” Ryo said. “For example, we just couldn’t even find any individual-level data that was available that would allow us to look at agency decisions to approve or deny naturalization applications.”

Ryo filed a request under the Freedom of Information Act for data on naturalization applicants with the USCIS in August 2018. However, officials were slow to cooperate with Ryo’s request, she said, which prompted her to file a federal complaint against USCIS with the assistance of American Oversight, a nonprofit organization working on government transparency issues, in July 2020. The litigation lasted more than a year.

“What requesters under that act are allowed to do after they’ve exhausted administrative appeal is to actually file a lawsuit, and that’s their remedy for inadequate production of records,” Ryo said. “I was just following what’s laid out under FOIA as my right, in terms of trying to obtain the records that I had requested initially from USCIS.”

Ryo concluded that, despite laws requiring USCIS to provide explanations to denied applications, 20% of denials in the data were absent. Another 14% in the data set were attributed to unexplained reasons, such as “Other” or “Secondary Evidence.” 

The incompleteness of adjudication data makes it hard to explain the disparities among naturalization applications, Ryo said.

“To the extent that we want to understand and explain the disparities, it would be really important for the government to be accurately documenting the adjudicators’ reasons for denial in the first place,” she said.

In a separate article, “Denying Citizenship: Immigration Enforcement and Citizenship Rights in the United States,” Ryo penned a recommendation for governments to better document and more readily release information about naturalization denials to increase public understanding of the CBP’s decision-making process.

In addition to the missing data, another conspicuous challenge for the researchers are the inconsistencies in coding that existed in the database. It took an “enormous” amount of time for the researchers to decipher and reconcile discrepancies over time, a process that researchers called “data cleaning,” before they could proceed with their analysis.

“There were changes in how different variables were encoded over time and there was no documentation for that,” Humphrey said. “We really just had to be really careful that we were interpreting things correctly because [the USCIS] didn’t provide a lot of documentation on what things meant.”

The study garnered attention from USCIS who invited Ryo to brief the study, which Ryo said she is “thrilled” about. 

“That indicates willingness on the part of the agency that I think is really admirable, and willingness to learn and to see how they can improve their system,” Ryo said.