Policing immigrant communities: Latino perceptions of police involvement in immigration enforcement

ABSTRACT

Using data from a survey of 2004 Latinos in four urban counties in the USA, this paper considers a question that has not been systematically investigated: how has increasing police involvement in immigration enforcement impacted the perceptions of the police that are held by immigrant and non-immigrant Latinos? Survey results indicate that many Latinos report fear of police, contributing to their social isolation and exacerbating their mistrust of law enforcement authorities. A substantial portion of Latino respondents report that they would be less likely to voluntarily contact the police if they are the victim of a crime, or to provide information about a crime, because they fear that police would use this contact as an opportunity to investigate their immigration status or that of their friends and family members. We use regression analysis to further analyse the determinants of these responses. Our findings suggest that negative encounters with police involving questions of immigration status and perceived unfair treatment, as well as vulnerabilities due to immigration and documentation status, contribute to social isolation and hesitancy to report crimes to police. These findings have implications for cooperation between police and Latino communities, particularly since local authorities have been enrolled in immigration enforcement.

KEYWORDS:

• Immigration enforcement
• immigration control
• context of migration
• community policing
• secure communities programme

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Pub. L. No. 104-208, 110 Stat. 3009-646 (1996) 8 U.S.C. § 1103(a)(10). In 2003, the responsibilities of the Immigration and Naturalization Service were transferred to three entities within the US Department of Homeland Security. US Immigration and Customs Enforcement is the entity that currently is empowered to enter into MOUs with state and local law enforcement agencies.

2 8 U.S.C. § 1357(g).

3 8 U.S.C. § 1357(g)(1).

4 As unweighted responses were used for the regression analysis, the descriptive statistics reported below may differ from the weighted estimates of the survey.

5 The other choices included jobs and the economy, health care, education, moral values, crime and drugs, and affordable housing.

6 Several diagnostic tests were performed for both models. Breusch-Pagan tests for heteroskedasticity failed to reject the null hypothesis in both cases. Variance inflation factors for all covariates fell below the rule-of-thumb of four. Durbin Watson tests failed to reject the null hypothesis, indicating that autocorrelation was not an issue.