Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Sep 11.
Published in final edited form as: Polit Groups Identities. 2015 Jun 2;3(3):454–468. doi: 10.1080/21565503.2015.1050415

Stuck between a rock and a hard place: The relationship between Latino/a’s personal connections to immigrants and issue salience and presidential approval

Gabriel R Sanchez 1,*, Edward D Vargas 2, Hannah L Walker 3, Vickie D Ybarra 4
PMCID: PMC4567257  NIHMSID: NIHMS708852  PMID: 26366323

Abstract

The Obama administration has simultaneously marketed the prospect of providing undocumented immigrants a pathway to citizenship through comprehensive immigration reform and overseen mass deportations of mostly Latino immigrants. While it is clear that immigration policy was highly influential to Latino voters in 2012, it remains unclear how this political hypocrisy is being interpreted by Latino voters. As deportations have risen steadily during the Obama administration, there has been little research on how deportations and personal connections to undocumented immigrants have influenced the political attitudes of the Latino/a electorate. Using a nationally representative survey of 800 registered Latino/a voters administered in 2013, we explore the relationships between personal connections to undocumented immigrants and issue salience among Latinos as well as Latinos’ views of President Obama. This study finds that registered Latino voters who know deportees and undocumented immigrants are more likely to report that they think the President and Congress should act on immigration policy versus all other policies. Moreover, Latino voters who know someone who is undocumented are less likely to have favorable views towards President Obama. This study has implications for our collective knowledge of how direct and indirect connections to policy outcomes influences the political behavior of the highly influential Latino/a electorate and how political and policy outcomes will be influenced in the future when a much higher proportion of the electorate have such connections.

Keywords: Deportation, Undocumented, Latino, Race and Ethnicity, Political Participation, Immigration

Introduction

Despite lack of movement on comprehensive immigration reform at the federal level, there has been tremendous policy action in this domain over the last several years. This has included a record number of immigration laws enacted at the state level and more deportations of immigrants during President Obama’s presidency than witnessed before in American history. This activity has had a noticeable impact on electoral politics, particularly during and after the 2012 election cycle, during which Latino voters identified only the economy and jobs as more important to their vote than immigration policy.

Consequently, scholars have explored the policy preferences of the electorate (Knoll et al. 2011; Rocha and Espino 2009), including the increased salience of immigration policy among Latinos (Suro and Escobar 2006; Lopez et al. 2011). However, we know less about the factors that have led Latinos not only to see immigration as a more important policy area, but also to become more liberal in their policy preferences as a collective. We focus our attention on this question: Is having a personal connection to immigrants, in the form of knowing someone personally who is undocumented and/or who has been detained/deported, associated with immigration policy salience and attitudes toward President Obama among Latinos? Results from our study will not only interest scholars of Latino politics, but also those interested in presidential approval, public opinion, and immigration policy.

Starting in 1996, three key pieces of federal legislation set the stage for higher numbers of deportations by substantially expanding the number of non-citizens, both legal and undocumented, eligible for deportation and removing the opportunity for judicial review (Hagan et al. 2008). First, the Anti-Terrorism and Effective Death Penalty Act (AEDPA) passed in 1996 virtually eliminated judicial review for all categories of immigrants subject to deportation. Second, the Immigration and Nationality Act (INA), also passed in 1996, strengthened the enforcement capacity of the then-Immigration and Naturalization Service by broadening the categories of noncitizens subject to deportation and expanding the offenses under which noncitizens could be deported. Third, the USA PATRIOT ACT of 2001 further expanded the pool of immigrants eligible for deportation and increased administrative powers available for use against noncitizens viewed as a threat to national security.

Although federal immigration authorities remain active in conducting workplace raids to apprehend noncitizens for deportation (Chaudhry et al. 2010), newer administrative programs providing for extensive partnerships between ICE (Immigration and Customs Enforcement) and local law enforcement now provide the primary mechanisms by which federal immigration authorities locate deportation-eligible immigrants in the interior of the United States (Golash-Boza and Hondagneu-Sotela 2013; Hagan et al. 2011). The Criminal Alien Program places ICE agents in state prisons to conduct immigrant screening., while the Secure Communities program allows cooperation between local law enforcement and ICE to identify potential criminal non-citizens among those booked into jail by local authorities (Capps et al. 2011). The 287(g) program, contained in the Immigration and Nationality Act of 1996 as a permissive but not required program was not used by local law enforcement until after 2001; it allows for training of local law enforcement to conduct immigration enforcement activities (Capps et al. 2011). Finally, increased apprehension and deportation of noncitizens is further potentiated by large expansion of detention capacity, due largely to the statutory quota inserted first in the 2006 Homeland Security annual spending bill now requiring ICE to keep an average of 34,000 immigration detainees per day in custody (Siskin 2012; Miroff 2013; Selway and Newkirk 2013).

The record number of deportations of the late Bush administration and the Obama administration has been driven in large part by the conflation of border enforcement and undocumented immigration with the terrorism threat to the U.S. held by the anti-immigrant public and elites following 9/11. Such fears were given life in the DHS Secure Border Initiative championed by Secretary Chertoff beginning in 2005, who said “We will continue to protect our nation from dangerous people by strengthening our border security efforts and continuing our efforts to gain effective control of our borders” (Chertoff 2005; 2008). In the absence of congressional immigration reform legislation, Chertoff’s Secure Border Initiative ended the practice of ‘catch and release,’1 increased detention bed capacity, and maximized the return on agreements with local law enforcement (Erlichman 2006; Beardsworth 2010).

Deportations of noncitizens from the interior of the U.S. (termed ‘removals’ by federal immigration authorities), have reached historic highs. The Department of Homeland Security reports over 4.6 million removals between 1997–2013,2 over twice the total number of all deportations before 1997 (Golash-Boza and Hondagneu-Sotelo 2013). Those removed by and large originate from Mexico – DHS reports that in 2013 over 240,000 Mexicans were removed, over five times the number of removals from the next highest country (Guatemala 47,000). Nine of the top 10 countries of origin targeted by removal are in Latin America, and include Mexico, Guatemala, Honduras, El Salvador, Dominican Republic, Ecuador, Brazil, Colombia, and Nicaragua. A number of scholars have noted the Mexican and Latino focus of current efforts. Lopez et al. (2011) find that 73% of removals are Mexican, although undocumented immigrants from Mexico represent only 58% of undocumented immigrants present in the U.S. Hagan et al. (2011) compare the current period of “mass deportations” to previous episodes of mass Mexican repatriation including Depression-era efforts and Operation Wetback in the mid-1950s. Golash-Boza and Hondagneu-Sotelo label the record-setting removals since 1997 a “racial removal” program concluding that the vast majority of removals have been Latino men (2013).

It is estimated that over 2.0 million family members have been separated from their families through deportations, and many of those deported are “long-term settlers whose deportation has dramatic effects on their families, disrupting relationships, remittances and settlement behavior” (Hagan et al. 2011, 1378). Family members in the U.S. separated from their households by detention and deportation often experience income loss, housing instability, and food insecurity, and children whose parents are deported may experience trouble in school and behavior changes (Chaurdry et al. 2010). Substantial numbers of U.S.-born children are impacted by deportation of their noncitizen parents (ICE 2012). At the community level, many immigrants now live in fear of deportation and family separation (Vargas 2010, Hagan et al. 2011).

Scholars have explored the policy preferences of the electorate during this period of unprecedented deportation action (see for example Knoll et al. 2011; Rocha and Espino 2009). However, we know less about how the outcomes associated with immigration policies enacted in recent years have influenced the Latino electorate.3 We turn our attention to this question by exploring whether personal connection to immigrants, in the form of knowing someone personally who is undocumented and/or who has been deported, is associated with immigration policy salience and attitudes toward President Obama among Latinos. This is an important area of research due to the increased influence of the Latino electorate on electoral politics, as well as the political implications of the unprecedented deportation numbers during the Obama administration’s term in office. In the data used for this analysis, for example, we find that nearly 65 percent of the sample knows someone who is undocumented, and 37 percent of the sample knows someone who has faced detention/deportation. We hypothesize that knowing someone who is undocumented or who has faced detention/deportation has the effect of making issues related to immigration salient to the individual, thus impacting a broader set of political attitudes. If we are correct, simply based on the large number of individuals with a connection to policy outcomes, immigration policy has the potential to impact the political perspectives of a large swath of an increasingly important sector of the electorate.

We first review Latinos’ overall attitudes toward immigration policy and immigrants. We then discuss what is known about presidential approval ratings, specifically among Latinos. Finally, we review the literature focused on how indirect contact with the criminal justice system influences political behavior, and here we contend that knowing someone who is deported is a type of such indirect contact. This review motivates our formal hypotheses regarding the relationship between this personal connection to immigration policy outcomes and Latino public opinion, and it grounds our work in three distinct yet related literatures. The results and conclusions are presented following our discussion of the data used for the paper. The results from our study will not only interest scholars of Latino politics but also those interested in presidential approval, public opinion, immigration policy, and criminal justice/criminology.

Latino attitudes toward immigration

Despite the common perception that Latinos in the U.S. support a very liberal immigration policy, analysis of Latino public opinion suggests that support for an expansionist immigration policy has not always been widespread among the Latino population. While early research found that Latinos were more likely than whites, Africans Americans, or Asians to support amnesty for illegal immigrants (Cain and Kiewiet 1987), other research has shown that a sizable portion of the Latino community believe that there are already too many immigrants coming to the United States annually (Binder et al. 1997; Cain and Kiewiet 1987; De la Garza et al. 1992; Espenshade and Hempstead 1996; Uhlaner and Garcia 2002).

Latinos generally have more favorable views of immigrants than non-Hispanic whites, but as is the case with non-Hispanic whites, those views are sensitive to the economic environment. Additionally, important differences exist among Latinos, and as might be expected, native-born Latinos are generally less favorable of expansionist policies than foreign-born Latinos. In national surveys of Latinos from 2002–2006, the Pew Hispanic Center consistently found Hispanics have favorable views of illegal immigrants (Suro and Escobar 2006). When presented with a series of options expressing views of the economic impacts of illegal immigrants on the U.S. economy, Latinos responding to the positive “undocumented immigrants help the economy by providing low cost labor” range from 68%–72% during these years with foreign-born Latinos expressing greater favorable support than native-born (78% vs. 64% in the 2006 survey). When asked about the impact of unauthorized immigrants on U.S. Hispanics, 50% said positive in 2007, dropping to 29% in 2010 and back up to 45% in 2013. The authors of these studies speculate this may be an impact of the economy – “it is possible that in 2010, a combination of a weak economy and a hardening political environment around immigration may have led to a more downbeat assessment of illegal immigration’s impact on U.S. Latinos” (Lopez and Gonzalez-Barrera 2013).

At the same time, immigration policy preferences have remained surprisingly consistent in reference to the proposal for a path to citizenship for undocumented immigrants in the United States. When asked about immigration policy options large majorities of Latinos since 2004 indicate a preference for a path towards legal status and eventual citizenship for undocumented immigrants. The Pew Hispanic Center reports 84% of Latinos surveyed in their 2004 and 2005 national surveys support the path to citizenship (Suro 2005) up to 86% in their more recent 2010 national survey of Latinos (Lopez, Morin and Taylor 2010).

Presidential approval

Past research exploring factors influencing presidential approval by the public has largely found that evaluations are rooted in a combination of aggregate and individual factors. Aggregate factors include unemployment and foreign policy, and individual factors include income, party identification, and issue salience (Wolf and Holian 2006; Hibbs 1982; MacKuen 1983; Ostrom and Simon 1985; Ostrom and Simon 1989; Edwards, Mitchell and Welch 1995). Early studies identified correlations in the aggregate that connected foreign policy and levels of unemployment to presidential approval (Ostrom and Simon 1985; Wolf and Holian 2006). Building on this research, scholars further explored the mechanisms for individuals that lead to presidential evaluations and support (Edwards, Mitchell and Welch 1995).

Research exploring the mechanisms by which issues such as war and the economy lead to presidential approval ratings largely centers on issue salience. The president takes, or fails to take, actions on specific issues important to members of the public, this is then reflected in presidential approval levels, and there are a few key issues that a majority of the public both finds important and associates with the president (Edwards, Mitchell and Welch 1995; Thomas, Sigelman and Baas 1984; McAvoy 2006). Examining the role of issue salience in presidential approval, Edwards et al. (1995) write, “As Converse pointed out three decades ago, people generally have only a few issues that are particularly important to them to which they pay attention. If a matter is not salient, it unlikely that it will play a major role in their evaluations of the president.” (111, 1995).

Presidential approval, then, is a dynamic function of citizen expectations of the president in a specific policy area and the extent to which the president has either met or failed to meet those expectations (Ostrom and Simon 1985, 1989). Importantly, something must first make an issue salient (Edwards, Mitchell and Welch 1995; Wolf and Holian 2006). Research exploring how issues become salient focuses on framing and priming, areas where the media takes center stage in making issues relevant to the public and shaping how the public thinks about the issue in reference to the president (Iyengar and Kinder 1987; Wolf and Holian 2006). Yet this research also identifies that when an issue is already highly salient, individuals are increasingly impervious to media effects (Wolf and Holian 2006). This suggests that issue salience, while potentially permeable to the media, is also individually determined.

In sum, evaluation of the president centers on issue salience, which is driven by individuals reflecting on the president’s performance on a given set of issues important to them and determining whether the president has met or failed to meet their expectations. What makes an issue salient, in a general sense, is some combination of media framing and priming along with individual experience. Connecting that issue to a president is a function of whether or not the public has an expectation that the president will act in that policy arena. This is important background for our analysis, as immigration policy fits these conditions for many Latino voters. We anticipate that a personal connection to the issue through knowing someone who is undocumented or who has faced detention/deportation will make this issue salient such that it influences evaluations of the president. We therefore expect to find an association between personal connections to deportation policy and Latino political behavior.

Latino presidential approval

Studies that explore Latino presidential approval are limited, but extant research does suggest that there are a few key policy issues of special import to the Latino community broadly (Sanchez et al. 2012; Stokes-Brown 2006; Wallace 2012; Branton 2007; Sanchez 2006). As Latinos have become a growing sector of the electorate, scholars have become increasingly interested in the circumstances in which Latinos vote as a group and what their unique policy interests might be. This scholarship has identified immigration and healthcare as issues of special import to the Latino community (Sanchez et al. 2012; Wallace 2012). Research in this vein was explicitly connected to presidential approval, as scholars endeavored to anticipate Latino support for the President’s 2012 bid for reelection, and worked from the perspective that President Obama had made certain promises on immigration reform during the 2008 election that he failed to follow through on in his first term. The authors suggest that because immigration is an issue salient to the community, failure to meet the expectations of the community led to lower levels of presidential approval, which might have negatively affected Obama’s chances with Latino voters in 2012 (Sanchez et al. 2012; Wallace 2012).

While the political context of the Obama presidency perhaps helped make immigration a salient issue for a broader coalition of Latino community members, other research suggests that this relationship is complex. Branton (2007) identifies that the relevance of immigration varies across the community, and does so by level of acculturation, where those further removed from the personal experience of immigration are less likely to hold the issue as salient. The larger point to take from Branton’s work in connecting it with presidential approval is that because issue salience can be highly individualized, experience with immigration policy makes it salient for members of the community. The literature (reviewed above) on issue salience suggests that an issue becomes salient via media framing, priming, or individual experience. Having a close relationship with someone who is threatened by immigration enforcement fits as an individual experience that may make an issue salient. Unlike issues to which the general public is broadly attuned, such as the economy or foreign policy, experience with immigration policy is highly specific, and consequences are potentially immediate if a close friend or family member is threatened with detention or deportation.

Thus, while immigration policy is unique to some of the major studies on issue salience and presidential approval, we see it as fitting squarely within conceptions of salience that tie individual experience to presidential evaluations in the instance that the president has made explicit promises around that issue. The promise of passage of comprehensive immigration reform in 2008 establishes an expectation of presidential action around this issue. While immigration has not always been salient to Latino voters in the past, we expect that members of the Latino community who have experienced a negative immigration event will be less likely to feel favorably towards the President than those who do not have a close immigration experience.

Political effects of indirect contact with the criminal justice system

Scholarship in sociology and law has effectively established that in the contemporary environment undocumented immigrations are criminalized (Ackerman and Furman 2013; Ackerman et al. 2014; Zatz and Smith 2012), that expanded immigration detention has brought immigration enforcement into the “prison-industrial complex” (Diaz 2012; Doty and Wheatley 2013; Welch 2000), and that immigration is functioning as a mechanism of social control similar to incarceration (Diaz 1012; Miller 2003). We extend the “criminalization of immigration” concept here and contend that in the absence of previous research examining the political impacts of knowing someone who is undocumented or has been detained/deported, we may rely on the closely analogous phenomena of knowing someone who has been incarcerated or had some kind of contact with the police to inform our expectations in this study. We draw on literature that explores the extant effects of the criminal justice system to speak to the more specific experience of knowing someone who is undocumented or who has been deported in thinking about how this might generate policy perspectives. Research on the political ramifications for individuals who have indirect contact with the criminal justice system is sparse and focuses on political participation outcomes, trust in government, and perceptions of police. This research finds that indirect contact leads to lower levels of voter turnout and degrades trust in government generally and trust in local police (Burch 2013; Vidales and Powe 2009; Lee, Porter and Comfort 2014; Hurwitz and Peffley 2005; Goffman 2014).

Vidales and Powe (2009) explore the use of local police to enforce immigration law in one community in Southern California. Within this community, the researchers were able to gather pre- and -post data on community members’ feelings about the police and the community alongside whether or not they had been stopped by the police. They found that Latino/a members of the community were less likely in the second time point to trust the police and more likely to say they felt unwelcome in their communities. This case study illustrates the potential effects of being embedded in a community where rates of contact with the criminal justice system are targeted at specific groups within the community.

This general drop in trust in government as a result of a negative experience with the criminal justice system is echoed in large nationally representative studies that look at the transitive effects of the system to those who are system-adjacent. Lee, Porter, and Comfort (2014) explore the political effects of having an incarcerated family member. Their analysis of a national dataset finds that having an incarcerated parent degrades trust in government, a relationship that is statistically significant. Similarly, other research exploring the racial divide in perceptions of police finds an assumption held by African Americans that police contact is procedurally unfair or unjust, where Whites hold the default assumption that it is fair (Hurwitz and Peffley 2005). They suggest that this is a result of the deep experience with criminal injustice held by the Black community compared to their White counterparts and further supports the notion that negative contact with the criminal justice system degrades trust in government, and more specifically, trust in police.

Finally, in addition to degrading trust in government, indirect contact with the criminal justice system drives down voter turnout. In her work, Trading Democracy for Justice: Criminal Convictions and the Decline of Neighborhood Political Participation, Traci Burch (2013) explores the effects of criminal justice intervention on community members who have not been incarcerated. She measures criminal justice intervention by looking at criminal convictions per census block, examining several hundred census blocks in two cities in the South. She finds this measure to be significantly predictive and negatively associated with voter turnout. This finding is echoed by Lee, Porter, and Comfort, who find that individuals with an incarcerated family member are less likely to turnout than their counterparts without an incarcerated family member (2014).

While it is unclear how indirect contact with the criminal justice system effects specific policy attitudes, it is clear that this experience has deep political ramifications. Individuals with indirect contact, across a variety of operationalizations of this concept, become alienated from the government by their experience, voting at lower levels and reporting lower levels of trust in government and local officials. It is not unreasonable to expect these findings to be reflected among Latinos who know someone who is undocumented, detained, or deported. We might expect this to drive down approval of the president in the instance where blame is assigned to him for failing to pass immigration reform as promised.

We propose two hypotheses: First, the personal connection hypothesis, that Latinos who know someone personally who is undocumented, as well as knowing someone who has been detained/deported will have a higher likelihood of identifying immigration as salient. We anticipate that the substantive impact of knowing someone who has been detained/deported will be greater than knowing someone who is undocumented, in that having been detained/deported represents actual punitive contact with law enforcement/corrections versus undocumented status which represents the potential for such punitive contact. Second, we offer the Obama responsibility hypothesis, which suggests that Latinos who personally know someone who is undocumented as well as knowing someone who has been detained/deported will have a lower favorability of President Obama due to assignment of partial responsibility to his administration. We anticipate that the substantive impact of knowing someone who has been detained/deported will be greater than knowing someone who is undocumented.

Data and methods

We utilize a unique survey conducted by Latino Decisions that was specifically focused on exploring Latino attitudes and personal interaction with immigration policy. The survey was fielded between February 15 and 26 of 2013. A total of 800 Latino registered voters were interviewed by live callers, with randomly selected calls to landline and cell-phone only Latino households, and each respondent was verified to be Latino through a screening question that opened the interview. The sample was designed to be nationally representative of the Latino electorate with respondents being drawn from states with the highest population density of Latino registered voters. Respondents could take the survey in English or Spanish, at their discretion, and were greeted in both English and Spanish by fully bilingual interviewers. The margin of error is +/− 3.5%4. All phone calls were administered by Pacific Market Research in Renton, Washington. The sample was weighted to reflect the 2010 Current Population Survey universe estimate of Latino registered voters nationwide.

The survey is ideal for our analysis given that it contains a measure of direct interaction with immigration policy. The survey items we draw from to create our measures of personal knowledge of immigration policy are: Do you personally know someone who has faced detention or deportation for immigration reasons? Now take a moment to think about all the people in your family, your friends, coworkers, and other people you know. Do you happen to know somebody who is an undocumented immigrant? We then create three mutually-exclusive dichotomous variables, know someone who has been detained/deported or undocumented (1=yes, 0=no), know someone that is undocumented only (1=yes, 0=no), and do not know an undocumented or detained/deported immigrant (1=yes, 0=no) which we specify as our reference category.

We utilize two main dependent variables for our analysis, one focused on immigration policy salience and another that measures favorability toward President Obama. We use the following to measure immigration salience: On the whole, what are the most important issues facing the [Hispanic/Latino] community that you think Congress and the President should address? Given that we are only interested in immigration versus all other policy areas, we create a binary variable (1=immigration policy, 0=all other policies) and estimate a logistic regression. For our second dependent variable, we model presidential approval with a measure asking respondents to identify how favorable they were toward the President at this point in his presidency (1=very unfavorable, 2=somewhat unfavorable, 3=somewhat favorable, 4=very favorable). Given the ordered structure of this outcome, here we estimate an ordered logistic regression that measures presidential favorability from negative to positive. To better understand the relationship between knowing someone who is undocumented and/or detained/deported compared to not knowing either, we control for age, language of interview, gender, education, income, partisanship, and country of origin.

Results

We begin our discussion of results by providing some descriptive statistics for our key explanatory variables that inform our analysis. Not only is the proportion of Latino registered voters who know someone who is undocumented and/or detained/deported high in our sample (70%), it is important to note that the number revealed here comes from a sample of registered voters, meaning all respondents are U.S. citizens who are eligible to vote. The mean age of our sample is 55 years of age, 60 percent of our sample took the survey in English, 57 percent of our sample was born in the U.S., and 57 percent of respondents were of Mexican origin. In terms of socio-economic status, 37 percent of our sample report that they made less than $40,000 a year, and the average education of sample was at least a high school graduate.

While the frequencies of our key explanatory variables are interesting and help motivate our larger analysis, by themselves they do not provide an answer to our question of whether these connections to immigration policy influence Latino political opinions. We turn to our multivariate analysis for answers to this question. We begin by exploring the impact of our primary explanatory variables, knowing someone who is undocumented and deported, and knowing someone who is undocumented only, compared to our reference category not knowing a deportee or an undocumented individual, on immigration policy salience. In line with our hypothesis, these variables are both significant and positive, indicating that respondents who have this personal connection to immigration and deportation policy are more likely to believe that immigration policy is the most important issue facing the Latino community that Congress and the President should address versus all other policies. Although both variables are significant, we find that knowing someone who has been detained/deported due to their immigration status has a larger substantive impact on issue salience than only knowing someone who is undocumented. More specifically, when we hold all other factors in the model constant, respondents who know someone who has been detained or deported are twice as likely to indicate that immigration policy is the most important issue facing the Latino community upon which Congress and the President should act. This finding is statistically significant at the 0.01 level. Moreover, those who report knowing someone who is undocumented are 4.4 percent more likely to believe immigration policy is the most salient issue Congress and the President should act on compared to those who do not know anyone who is undocumented or has been departed, holding all else constant. This finding is statistically significant at the 0.01 level. This reveals that the personal connection that Latino voters have to immigration policies and their outcomes may explain how immigration policy emerged as one of the most dominant themes of the 2012 election.

Among our control variables we find that those who did not report their income were less likely to believe immigration policy was the most important issue facing the Latino community versus all other policies in the poll compared to Latinos with the lowest incomes (the base category). Immigration policy is also more salient among foreign born Latinos than the U.S. born respondents in the survey, and among Democrats as opposed to Republicans.

The next step in our discussion of results is to explore the impact of our primary explanatory variables on favorability toward President Obama. Here we find partial support for our hypothesis, as knowing someone who is undocumented yields lower favorability scores for President Obama among Latino registered voters. In fact, the probability of having a favorable impression of President Obama decreases by a robust 40.5 percent for respondents who know someone who is undocumented, holding all else constant. This variable is significant at the 0.05 level. However, inconsistent with our hypothesis, knowing someone who has been deported or detained is not statistically significant in this model. While not an expected finding, we believe that the lack of significance for this variable may be related to the limitation of the measure to capture the nature of the relationship someone may have to a deportee, as well as the circumstances that may have led to the individual being deported or detained. For example, if the person the respondent knows was arrested for committing a crime which led to deportation this might not result in a drop in approval for the President in the same way having a family member deported for no other reason than their immigration status would. This finding suggests that the nuances of someone’s relationship with immigration and immigrants influences public opinion among Latinos, specifically views about the President. Among controls, both age and female are significant and positive, with higher presidential approval among Latinas and older Latinos. Foreign-born Latinos and non-Republican Latinos are also more supportive of the President at this point in his second term in office. Interestingly, Latinos of Mexican descent are 41.1 percent less likely to have favorable views of President Obama compared to non-Mexican origin Latinos.

Conclusion

The Obama presidency has been marked with inconsistent rhetoric and policy action concerning immigration policy. With a campaign promise to enact comprehensive immigration reform in 2008, Latinos were optimistic about seeing major movement on federal immigration policy. However, with the exception of the DACA (Deferred Action for Childhood Arrivals) executive action, the presidency has been marked largely by disappointment in regard to immigration policy. Most visible to Latinos, the Obama administration has deported more undocumented immigrants than President Bush. Given that most of these deportees are Latino and that our data indicates that many Latino registered voters are personally connected to Latino voters, we hypothesized that knowing someone who has been detained/deported or who is undocumented would influence Latino public opinion. Our results generally support our hypotheses, as these two measures lead to increased immigration salience, and knowing someone who is undocumented yields lower levels of favorability toward President Obama.

This study provides insights into how immigration policy during a period of heightened salience of the issue has marked electoral politics by focusing on a key electorate directly tied to immigration politics and policy outcomes. Our results indicate that the personal connection Latino registered voters have to undocumented immigrants influences the salience of immigration policy among the Latino electorate and shapes their views of the President. These findings have implications for multiple research areas we draw upon to inform our theory and hypotheses. The results advance the literature focused on Latino policy views toward immigration by providing an explanation for the salience of this policy area within the Latino electorate beyond ethnic identity (Sanchez 2006) and also provide an additional explanation for internal variation on Latino immigration policy. These new insights should motivate scholars of Latino politics and immigration policy to explore how personal connections to deportation and immigrant policy more broadly influence public opinion toward immigration policy.

Furthermore, our findings indicate that immigration fits the conditions for issue salience identified in the presidential approval literature. This is an important contribution given the relatively limited research in this area of study that focuses on the Latino electorate. As immigration is a highly publicized policy area, scholars should explore other questions related to issue salience and political behavior, including what role the media has had in the rise of immigration salience among Latinos as well as how the passage (or not) of comprehensive immigration reform will impact the legacy of President Obama among Latinos. Finally, by situating our study within the literature exploring implications for indirect contact with the criminal justice system, we have provided an opportunity for scholars to explore other lines of inquiry related to this unique conceptualization of deportation. We hope that this will help connect scholars whose work focuses on the criminal justice system to those working in immigration policy.

Finally, from an historical perspective, as the electorate changes in response to the contemporary immigration influx, voters with direct connections to immigration will come to make up a larger and more consequential portion of the overall electorate. Given that large numbers of U.S.-born children of immigrants are still under 18 and not yet eligible to vote, we can anticipate that the future electorate will contain a much higher portion of voters with personal connections to the immigration experience and to deportation policy. Among Latinos, immigrants and their children made up 55 percent of eligible Latino voters in 2012, up from 49 percent in 1996 (Oakford 2014). To the extent that personal connections to immigration policy motivate political behavior, these behavioral impacts will become more consequential for overall U.S. political and policy outcomes.

Table 1.

Summary Statistics for Analysis (n=800)

Variable Mean Std. Dev. Min Max
Deported and Undocumented 0.341 0.475 0 1
Undocumented 0.358 0.480 0 1
Neither Deported or Undocumented 0.286 0.452 0 1
Policy Preferences1 0.488 0.500 0 1
Views of President Obama2 3.239 1.065 1 4
Age 54.746 16.370 19 91
Spanish Language 0.401 0.490 0 1
Female 0.514 0.500 0 1
Education3 3.895 1.534 1 6
Inc. Missing 0.175 0.380 0 1
Inc. Less $39k 0.368 0.482 0 1
Inc.$40k–$80k 0.131 0.338 0 1
Inc. $60–$80K 0.099 0.299 0 1
Inc. $80k+ 0.228 0.419 0 1
Foreign Born 0.429 0.495 0 1
Republican 0.126 0.332 0 1
Democrat 0.586 0.493 0 1
Independent 0.288 0.453 0 1
Mexican Origin4 0.569 0.495 0 1
Open in a new tab
1

Policy Preferences (0= All Other Preferences, 1=Immigration)

2

Views of Pres. Obama (1=Very Unfavorable, 2=Somewhat Unfav., 3=Somewhat Fav., 4=Very Favorable,)

3

Education (1=Grade 1–8, 2=Some HS, 3=HS, 4=Some College, 5=College Grad, 6=Post-Grad)

4

Mexican Origin (0=Non-Mexican Origin, 1=Mexican Origin)

Table 2.

Logistic Coefficients for Regression of Knowing Someone That Is Undocumented or Deported on Immigration Being The Most Important Issue Facing the Latino Community That Congress and the President Should Be Addressing vs. All Other Policies using a 2013 Latino Decision survey.

Variables Model 1
β Odds Ratio
Deported & Undocumented1 0.743*** 2.102***
Undocumented 0.671*** 1.956***
Inc. Missing2 −0.600* 0.549*
Inc. $40k–$60k 0.218 1.244
Inc. $60–$80K 0.116 1.123
Inc. $80k+ −0.281 0.755
Education 0.086 1.090
Female 0.101 1.106
Age −0.003 0.997
Spanish Language −0.184 0.832
Foreign Born 0.582** 1.790**
Democrat3 0.532* 1.703*
Independent 0.537 1.711
Mexican Origin −0.289 0.749
Constant −1.142* 0.319*
Observations 578
Adjusted R-Sq. 0.0445
Open in a new tab
***

p<0.01

**

p<0.05,

*

p<0.1

β=Logistic regression coefficients

Reference Categories:

1

Not Knowing Someone Detained/Deported or Undocumented

2

Income: Less than $39k

3

Republican

Table 3.

Ordered Logistic Coefficients for Regression of Knowing Someone That is Undocumented or Deported on Having a Favorable Impression of President Obama using a 2013 Latino Decisions survey.

VARIABLES Model 2
B Odds Ratio
Deported & Undocumented1 0.076 1.079
Undocumented −0.519** 0.595**
Inc. Missing2 −0.319 0.727
Inc. $40k–$60k 0.389 1.476
Inc. $60–$80K −0.417 0.659
Inc. $80k+ −0.203 0.817
Education 0.008 1.008
Female 0.355* 1.426*
Age 0.015** 1.015**
Spanish Language 0.348 1.416
Foreign Born 0.607*** 1.836***
Democrat3 3.356*** 28.670***
Independent 1.977*** 7.221***
Mexican Origin −0.534*** 0.586***
Constant 3.127*** 22.817***
Observations 603
Adjusted R-Sq. 0.044
Open in a new tab
***

p<0.01,

**

p<0.05,

*

p<0.1

β=Ordered logistic regression coefficients

Reference Categories:

1

Not Knowing Someone Detained/Deported or Undocumented

2

Income: Less than $39k

3

Republican

Footnotes

1

Ending ‘catch and release’ is reflected both in the rise of deportations and reduction in ‘voluntary returns’ in which “immigrants apprehended at the border were largely allowed to voluntarily return to Mexico-without any penalty.” (Light et al. 2014, 5). While it is likely that many new ‘removals’ would in previous years have been classified as ‘returns,’ the change in practice is far more than simply a change in semantics. The classification change has resulted in increased penalties including increased chances of federal conviction for those attempting to re-enter after deportation (Light et al. 2014).

2

The authors’ calculations using Homeland Security data – 2010 Yearbook of Immigration Statistics for FY 1997-FY 2010, and annual HSD announcements of year-end removal statistics for FY 2011- FY 2013. http://www.dhs.gov/immigration-statistics

3

While there is much discussion of this research area among bloggers, particularly through the Latino Decisions blog, there has been little published work in this area.

4

Margin of error was calculated using a 95% confidence level, and is calculated using the formula, 0.98/√n= 0.98/√800=0.0346 or 3.5%.

Contributor Information

Gabriel R. Sanchez, University of New Mexico.

Edward D. Vargas, University of Wisconsin-Madison

Hannah L. Walker, University of Washington

Vickie D. Ybarra, University of New Mexico

References

  1. Ackerman Alissa R, Furman Rich. The Criminalization of Immigration and the Privatization of the Immigration Detention: Implications for Justice. Contemporary Justice Review. 2013;16(2):251–63. [Google Scholar]
  2. Ackerman Alissa R, Sacks Meghan, Furman Rich. The New Penology Revisited: The Criminalization of Immigration as a Pacification Strategy. Justice Policy Journal. 2014;11(1):1–20. [Google Scholar]
  3. Binder Norman E, Polinard JL, Wrinkle Robert D. Mexican American and Anglo Attitudes toward Immigration Reform: A View from the Border. Social Science Quarterly. 1997;78(2):324–37. [Google Scholar]
  4. Branton Regina. Latino Attitudes toward Various Areas of Public Policy: The Importance of Acculturation. Political Research Quarterly. 2007;60(2):293–303. [Google Scholar]
  5. Burch Traci. Trading Democracy for Justice: Criminal Convictions and the Decline of Neighborhood Political Participation. Chicago, IL: The University of Chicago Press; 2013. [Google Scholar]
  6. Cain Bruce E, Roderick Kiewiet D. Latinos and the 1984 Election: A Comparative Perspective. In: de la Garza Rodolfo., editor. Ignored Voices: Public Opinion Polls and the Latino Community. Austin, TX: Center for Mexican American Studies; 1987. [Google Scholar]
  7. Capps Randy, Rosenblum Marc R, Rodriguqz Cristina, Chishti Muzaffar. Delegation and Divergence: A Study of 287(g) State and Local Immigration Enforcement. [May 29, 2014];Migration Policy Institute. 2011 www.migrationpolicy.org/pubs/287g-divergence.pdf.
  8. Chaudry Ajay, Capps Randy, Pedroza Juan Manuel, Casteneda Rosa Maria, Santos Robert, Scott Molly M. Facing our Future: Children in the Aftermath of Immigration Enforcement. [May 29, 2014];The Urban Institute. 2010 http://www.urban.org/UploadedPDF/412020_FacingOurFuture_final.pdf.
  9. Chertoff Michael. Testimony of Secretary Michael Chertoff before the House Subcommittee on Homeland Security Appropriations. 2008. Apr 10, [Google Scholar]
  10. Chertoff Michael. Statement of Secretary Michael Chertoff US Department of Homeland Security before the United State Senate Committee on Homeland Security and Government Affairs. 2005. Jul 14, [Google Scholar]
  11. De la Garza Rodolfo O, DeSipio Louis, Chris Garcia F, Garcia John, Falcon Angelo. Latino Voices: Mexican, Puerto Rican, and Cuban Perspectives on American Politics. Boulder, CO: Westview Press; 1992. [Google Scholar]
  12. Diaz Jesse. Prison and Immigration Industrial Complexes: The Ethnodistillation of People of Color and Immigrants as Economic, Political, and Demographic Threats to US Hegemony. International Journal of Criminology and Sociology. 2012;1:265–84. [Google Scholar]
  13. Doty Roxanne Lynne, Wheatley Elizabeth Shannon. Private Detention and the Immigration Industrial Complex. International Political Sociology. 2013;7(4):426–43. [Google Scholar]
  14. Edwards GC, Mitchell W, Welch R. Explaining Presidential Approval: The Significance of Issue Salience. American Journal of Political Science. 1995;39(1):108–34. [Google Scholar]
  15. Erlichman Jeff. Secure Border Initiative Program Guide. [May 29, 2014];Trezza Media Group. 2006 http://www.trezzamediagroup.com/uploads/sbi_supplement.pdf.
  16. Espenshade Thomas J, Hempstead Katherine. Contemporary American Attitudes toward U.S. Immigration. International Migration Review. 1996;30(2):535–70. [PubMed] [Google Scholar]
  17. Goffman Alice. On the Run: Fugitive Life in an American City (Fieldwork Encounters and Discoveries. Chicago, IL: The University of Chicago Press; 2014. [Google Scholar]
  18. Golash-Boza Tanya, Hondagneu-Sotelo Pierrette. Latino Immigrant Men and the Deportation Crisis: A Gendered Racial Removal Program. Latino Studies. 2013;11(3):271–92. [Google Scholar]
  19. Hagan Jacqueline, Eschbach Karl, Rodriguez Nestor. U.S. Deportation Policy, Family Separation, and Circular Migration. International Migration Review. 2008;42(1):64–88. [Google Scholar]
  20. Hibbs Douglas A. On the Demands for Economic Outcomes: Macroeconomic Performance and Political Support in the United States, Great Britain, and Germany. Journal of Politics. 1982;44:426–62. [Google Scholar]
  21. Hurwitz J, Peffley M. Explaining the Great Racial Divide: Perceptions of Fairness in the U.S. Criminal Justice System. The Journal of Politics. 2005;67(3):762–83. [Google Scholar]
  22. Iyengar Shanto, Kinder Donald R. News that Matters: Television and American Opinion. Chicago, IL: University of Chicago Press; 1987. [Google Scholar]
  23. Knoll Benjamin R, Redlawsk David P, Sanborn Howard. Framing Labels and Immigration Policy Attitudes in the Iowa Caucuses: “Trying to Out-Tancredo Tancredo. Political Behavior. 2011;33:433–454. [Google Scholar]
  24. Lee H, Porter LC, Comfort M. Consequences of Family Member Incarceration: Impacts on Civic Participation and Perceptions of the Legitimacy and Fairness of Government. Annals- American Academy of Political and Social Science. 2014;651(1):44–73. doi: 10.1177/0002716213502920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Light Michael T, Lopez Mark Hugo, Gonzalez-Barrera Ana. The Rise of Federal Immigration Crimes. Washington, D.C: Pew Research Center’s Hispanic Trends Project; 2014. Mar, [Google Scholar]
  26. Lopez Mark Hugo, Gonzalez-Barrera Ana. Latinos’ View Illegal Immigration’s Impact on Their Community Improve. Washington, D.C: Pew Research Center’s Hispanic Trends Project; 2013. [May 29, 2014]. http://www.pewhispanic.org/files/2013/10/Latinos-Views-of-Impact-of-Illegal-Immigration-on-Their-Community-Improve.pdf. [Google Scholar]
  27. Lopez Mark Hugo, Gonzalez-Barrera Ana, Motel Seth. As Deportations Rise to Record Levels, Most Latinos Oppose Obama’s Policy. [May 29, 2014];Pew Hispanic Center. 2011 http://www.pewhispanic.org/files/2011/12/Deportations-and-Latinos.pdf.
  28. Lopez Mark Hugo, Morin Rich, Taylor Paul. Illegal Immigration Backlash Worries, Divides Latinos. [May 29, 2014];Pew Hispanic Center. 2010 http://pewhispanic.org/files/reports/128.pdf.
  29. MacKuen Michael B. Political Drama, Economic Conditions, and the Dynamics of Presidential Popularity. American Journal of Political Science. 1983;27:165–92. [Google Scholar]
  30. Mcavoy G. Stability and Change: The Time Varying Impact of Economic and Foreign Policy Evaluations on Presidential Approval. Political Research Quarterly. 2006;59(1):71–83. [Google Scholar]
  31. Miller Teresa A. Citizenship and Severity: Recent Immigration Reforms and the New Penology. Georgetown Immigration Law Journal. 2003;17(4):661–66. [Google Scholar]
  32. Miroff Nick. The Washington Post. 2013. Oct 13, Controversial Quota Drives Immigration Detention Boom. [Google Scholar]
  33. Oakford Patrick. The Latino Electorate by Immigrant Generation: The Rising Influence of Children of Immigrants. [August 1, 2014];Center for American Progress. 2014 Jun 12; http://cdn.americanprogress.org/wp-content/uploads/2014/06/Oakford-LatinoElectorate-brief-FINAL2.pdf.
  34. Ostrom CW, Simon DM. Promise and Performance: A Dynamic Model of Presidential Popularity. The American Political Science Review. 1985;79(2):334–58. [Google Scholar]
  35. Ostrom CW, Simon DM. The President’s Public. The American Political Science Review. 1989;79(2):334–358. [Google Scholar]
  36. Regina B. Latino Attitudes toward Various Areas of Public Policy: The Importance of Acculturation. Political Research Quarterly. 2007;60(2):293–303. [Google Scholar]
  37. Rocha Rene R, Espino Rodolfo. Racial Threat, Residential Segregation, and the Policy Attitudes of Anglos. Political Research Quarterly. 2009;62(2):415–26. [Google Scholar]
  38. Sanchez Gabriel R, Medeiros J, Sanchez-Youngman S. The Impact of Health Care and Immigration Reform on Latino Support for President Obama and Congress. Hispanic Journal of Behavioral Sciences. 2012;34(1):3–22. [Google Scholar]
  39. Sanchez Gabriel R. The Role of Group Consciousness in Latino Public Opinion. Political Research Quarterly. 2006;59(3):435–446. [Google Scholar]
  40. Selway William, Kewkirk Margaret. Congress Mandates Jail Beds for 34,000 Immigrants and Private Prisons Profit. [May 29, 2014];Bloomberg. 2013 Sep 24; http://www.bloomberg.com/news/2013-09-24/congress-fuels-private-jails-detaining-34-000-immigrants.html.
  41. Siskin Alison. Immigration-Related Detention: Current Legislative Issues. [May 29, 2014];Congressional Research Service. 2012 http://www.fas.org/irp/crs/RL32369.pdf.
  42. Stokes-Brown Atiya Kai. Racial Identity and Latino Vote Choice. American Politics Research. 2006;34(5):627–52. [Google Scholar]
  43. Suro Robert, Escobar Gabriel. 2006 National Survey of Latinos: The Immigration Debate. [May 29, 2014];Pew Hispanic Center. 2006 http://www.pewhispanic.org/2006/07/13/the-2006-national-survey-of-latinos-the-immigration-debate/
  44. Suro Robert. Attitudes toward Immigrants and Immigration Policy: Surveys among Latinos in the U.S. and in Mexico. [May 29, 2014];Pew Hispanic Center. 2005 http://www.pewhispanic.org/2005/08/16/attitudes-toward-immigrants-and-immigration-policy/
  45. Thomas DB, Sigelman L, Baas LR. Public Evaluations of the President: Policy, Partisan, and ‘Personal’ Determinants. Political Psychology. 1984;5(4):531–42. [Google Scholar]
  46. Uhlaner Carole Jean, Chris Garcia E. Latino Public Opinion. In: Norrander Barbara, Wilcox Clyde., editors. Understanding Public Opinion. Washington, DC: CQ Press; 2002. [Google Scholar]
  47. Vargas Edward. PhD thesis. Bloomington: Indiana; N.d. Mixed-Status Families and Public Assistance: The Impact on Both States and Families. [Google Scholar]
  48. Vidales G, Day KM, Powe M. Police and Immigration enforcement: Impacts on Latino(a) Residents’ Perceptions of Police. Policing. 2009;32(4):631–53. [Google Scholar]
  49. Wallace SJ. It’s Complicated: Latinos, President Obama, and the 2012 Election. Social Science Quarterly. 2012;93(5):1360–83. [Google Scholar]
  50. Welch Michael. The Role of the Immigration and Naturalization Service in the Prison-Industrial Complex. Social Justice. 2000;27(3):73–88. [Google Scholar]
  51. Wolf Michael, Holian David. Polls, Elite Opinion, and the President: How Information and Issue Saliency Affect Approval. Presidential Studies Quarterly. 2006;36(4):584–605. [Google Scholar]
  52. Zatz Marjorie S, Smith Hilary. Immigration, Crime, and Victimization: Rhetoric and Reality. Annual Review of Law and Social Science. 2012;8:141–59. [Google Scholar]

RESOURCES