How do incarceration and arrest reporting biases affect comparisons between undocumented immigrants and US citizens?
Executive summary
Comparisons between undocumented immigrants and U.S. citizens depend heavily on which metric is used and how data are collected: arrest counts can be shaped by policing practices and include many minor offenses, while incarceration rates reflect convictions and sentencing regimes that themselves are uneven across citizenship status [1] [2] [3]. Robust state-level work in Texas — where immigration status is recorded at arrest — finds substantially lower felony-arrest rates for undocumented immigrants, but that finding coexists with documented measurement and enforcement biases that complicate national generalization [4] [5] [6].
1. Arrests are not a neutral measure — they reflect policing choices
Arrest data capture law-enforcement decisions about whom to stop, charge, and fingerprint, and include many minor infractions that researchers warn are more susceptible to police bias than serious crimes, so raw arrest tallies can overstate or misrepresent criminality across groups [1] [2]. The Texas studies exploit a legal requirement to check arrestees’ immigration status and thus produce a cleaner numerator for undocumented arrests than most jurisdictions provide, but that does not eliminate uneven policing practices that determine who is arrested in the first place [5] [6].
2. Incarceration is a different lens — convictions, sentencing, and court processes matter
Analysts who prefer incarceration as the indicator argue it better captures serious offending because it normally requires conviction; long-run work finds immigrants overall are substantially less likely to be incarcerated than the U.S.-born, particularly in recent decades [2] [7]. At the same time, several studies note that noncitizens often receive harsher treatment in sentencing and may be incarcerated longer for similar offenses, which can complicate interpretation of incarceration gaps as pure crime-rate differences [3] [8].
3. Denominator problems and population estimates can sway rates
Calculating arrest or incarceration rates requires estimating the size of the undocumented population — a notoriously difficult task — and small changes in that denominator can materially affect per‑capita comparisons; Texas authors test sensitivity and report their conclusions are robust to plausible alternative population estimates, but national analyses lack the same uniform arrest-status linkage [5] [9]. Where immigration status is not systematically recorded, researchers must rely on imperfect imputations or surveys, which introduces uncertainty into cross-group rate comparisons [10].
4. Immigration enforcement creates entanglements with criminal statistics
A large share of increased prosecutions of noncitizens historically reflects immigration-law violations rather than violent or property crimes, meaning immigration enforcement can inflate counts of “criminal” involvement by noncitizens in ways that are conceptually distinct from ordinary criminal offending [10]. This intersection also means jurisdictions that prioritize immigration cooperation will show different arrest and booking patterns than jurisdictions that do not, producing geographic heterogeneity in measured rates [9].
5. Empirical results: lower arrest and incarceration rates for immigrants, with caveats
Multiple recent syntheses and high-profile studies report that immigrants — including unauthorized immigrants in Texas — are arrested and incarcerated at lower rates than the U.S.-born for violent, drug, and property offenses, with Texas data showing undocumented people less than half as likely to be arrested for violent crimes and much lower for property offenses [4] [6] [11]. Advocacy and policy groups cite this body of evidence to argue that immigrant flows do not drive crime, while scholars caution that measurement choices (arrests vs. incarceration), local enforcement practices, and sentencing disparities must be acknowledged when interpreting those findings [12] [13] [3].
6. Politics, narratives, and the risk of misinterpretation
Public discourse often amplifies dramatic incidents and conflates immigration enforcement activity with ordinary crime, and political actors may have incentives to emphasize isolated events over population-level studies; the research community and criminal-justice advocates explicitly warn against attributing broad public-safety problems to immigrants without considering measurement bias and context [12] [13]. At the same time, researchers using nationwide incarceration series argue their measures avoid some policing biases but contend that sentencing differentials and data limitations still require careful interpretation [7] [3].
Bottom line: what the biases mean for comparisons
Arrest and incarceration biases push comparisons in different directions: arrests can over- or under-represent group contact with police depending on policing priorities and minor-offense enforcement [1] [2], while incarceration is shaped by charging, plea bargaining, and sentencing patterns that can disadvantage noncitizens [3] [8]. The consistent empirical thread across careful studies is that immigrants — including many analyses of undocumented populations in Texas — are not more criminal than U.S.-born people, but translating that research into definitive national claims requires acknowledging the measurement limits, jurisdictional variation, and the political incentives that shape which statistics get reported and amplified [4] [5] [12].