By Christopher Inkpen, Jonathan Holt, John Bollenbacher, Heather Deforge, Lilly Yu, Pranav Athimuthu, Nicole Jasperson, Renata Zablocka, Nick Kruskamp

This report, funded by the National Institute of Justice (award #: 2020-R2-CX-0027) and prepared by RTI International, describes the results of a National Institute of Justice-funded research study that uses advanced analytical methods and novel datasets to explore the complex relationship between immigration, crime, and crime reporting at the neighborhood level. The study, which employs crime and crime reporting data from ten jurisdictions across the United States paired with a synthetic population that estimates the unauthorized immigrant population, aims to provide an in-depth analysis at the Census tract level. Analyses focus on unauthorized immigration and its correlation with drug, property, and violent crime rates, while accounting for crime reporting in traditional and emerging immigrant destinations along with sites with low foreign populations. BackgroundL  Despite persistent political discourse linking immigration and increases in crime, most academic research contradicts this notion, showing either a negative or null relationship between immigration and crime. At the individual level, first-generation immigrants tend to have lower arrest rates than native-born citizens. Yet this trend diminishes with subsequent generations, as the children of first-generation immigrants (i.e., second generation immigrants) are often arrested at similar rates to children of native-born citizens. However, few studies assess the relationship between documentation status and offending. Macro-level analyses that focus on crime and immigration in specific areas reveal that areas with higher immigrant populations often experience lower crime rates or that the prevalence of immigrants in an area is not associated with an increase in arrests. Yet these studies frequently omit distinctions in documentation status, as these data are often unavailable. Further, many macro-level analyses are conducted at the county, state, or city level, which may obscure relationships observed at local levels. This study also attempts to control for the nuances of crime reporting among immigrant populations. Immigrant neighborhoods, especially those in emerging destinations, show lower rates of crime reporting. Trust in the police and fear of deportation are potentially significant factors influencing the likelihood of underreporting crimes. highlighting the importance of community-police relations. Strong police-community relations are crucial for public safety as they foster trust and cooperation, leading to more accurate crime reporting, effective law enforcement, and safer spaces. Methodology and Data Sources This study uses a variety of data sources to analyze the relationship between the presence of unauthorized immigrants in a Census tract and corresponding crime rates. Synthetic Population Development Traditional methods for estimating the unauthorized immigrant population in the United States rely on either demographic accounting or model-based survey imputation techniques. Defined simply, demographic accounting involves subtracting the estimated number of legally present immigrants from the total foreign-born population recorded in U.S. Census Bureau surveys. These techniques may also employ logical edits to large survey datasets that use characteristics like age, education, and place of birth to infer unauthorized status. Model-based survey imputation techniques combine data from different survey datasets to estimate unauthorized status in nationally representative surveys, using statistical techniques to merge information from surveys with immigration queries and those with extensive geographic detail. This study builds on the work of model-based imputation methods by developing models that predict unauthorized status in a survey dataset and applying them to a synthetic population of the United States, based on U.S. Census Bureau survey datasets. This approach allows for granular estimates of unauthorized immigrant populations at the Census tract level. By combining data from the Survey of Income and Program Participation (SIPP) and the American Community Survey (ACS), the study developed a robust model to predict unauthorized status and produced Census tract-level estimates of the unauthorized immigrant population for 2019. Validation efforts for these estimates included comparisons with county- and state-level estimates from sources like the Migration Policy Institute and Pew Research Center along with scaling local estimates to meet state-level figures 

Research Triangle Park, NC: RTI International, 2024. 53p.

By Michael T. Light

Leveraging the Computerized Criminal History System (CCH), which provides case processing information for all arrests recorded in Texas between 2011 and 2018, this study explores the relationship between unauthorized immigration, crime, and recidivism. The first section compares the criminality of undocumented immigrants to legal immigrants and native-born U.S. citizens between 2012 and 2018 in Texas. The second section establishes the foundational empirics for general criminological literature on the immigration-homicide nexus. Key findings include: 1) Immigrants generally exhibit lower rates of serious violent crime in California and Texas. This is true for overall rates of violence and homicide. 2) Violent crime rates among immigrants in California are lower than among immigrants in Texas, and the relative gap between native and foreign-born individuals is considerably larger in California. 3) In both states, there is substantial heterogeneity in the immigration-homicide relationship by race/ethnicity and national origin. Generally speaking, immigrants from Asian countries have especially low rates of homicide offending. 4) Relative to the U.S.-born population, the criminal histories of immigrants arrested for violent crimes are both less extensive and less severe. Section 3 answers important questions about the extent to which immigrant criminality changed during the Trump administration. The authors find no evidence, descriptive or otherwise, to suggest that the transition from the Obama administration to the Trump administration had a meaningful effect on immigrant criminality, whether measured as violence, property, drug, or traffic offenses. Section 4 examines recidivism among the undocumented population and details the data limitations that caution against strong conclusions on this issue. Most notably, criminal justice databases rarely have information as to whether the defendant was eventually deported. As a result, researchers do not know if an individual restrains from recidivating or is simply removed from the country and is thus no longer at risk of recidivating.

Madison WI: University of Wisconsin 2022. 79p.