461438 461438American Politics ResearchLevin © The Author(s) 2011

APR41410.1177/1532673X12

Reprints and permissions: sagepub.com/journalsPermissions.nav

Article

Political Inclusion of Latino Immigrants: Becoming a Citizen and Political Participation

American Politics Research 41(4) 535­–568 © The Author(s) 2013 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1532673X12461438 apr.sagepub.com

Ines Levin1

Abstract Does becoming a citizen represent a major step toward the inclusion of Latino immigrants in the American polity? I study this question by comparing the behavior of immigrants who have acquired citizenship with that of immigrants who are not willing or not eligible to become citizens, focusing on nonelectoral political activities such as contacting government officials and working to solve problems with others informally or through existing groups and organizations. The data analysis is based on recent survey data from the 2006 Latino National Survey. I use matching methods to control for the nonrandom selection of respondents into citizenship status, and perform a sensitivity analysis to evaluate the robustness of findings to hidden bias. Results suggest that granting citizenship to nonnaturalized immigrants is a measure that, by itself, is unlikely to achieve full inclusion of Latino immigrants in the broader democratic process. Keywords Latino politics, political participation, citizenship, naturalization, matching methods

1

University of Georgia, Athens, GA, USA

Corresponding Author: Ines Levin, University of Georgia, Athens, GA 30602, USA. Email: [email protected]

536

American Politics Research 41(4)

Introduction Foreign-born Latinos are one of the fastest growing ethnic groups in the United States.1 The recent acceleration of immigration inflows from Latin American countries represents a new test of the view of America as a nation with remarkable ability to incorporate immigrants (Schildkraut, 2007). Scholarly and public opinion is divided when it comes to the consequences of Latino immigration (Fraga et al., 2006b; Huntington, 2004; Pew Research Center for the People & the Press, 2006). Contrasting cultures and conflicts of interests between the native-born population and newly arrived immigrants create apprehension in the native born, and may impede immigrant acculturation and political assimilation of the type predicted by the classical assimilation theory; leading instead to divergent paths of assimilation where immigrants exposed to antiimmigrant rhetoric and socioeconomic discrimination struggle to incorporate, or reject assimilation, and remain isolated from mainstream society (Michelson, 2001, 2003; Portes & Zhou, 1993). Some of the factors driving attitudes toward immigration include concerns about: the cost to taxpayers of allowing immigrants access to public benefits and social services (Alvarez & Butterfield, 1997; Citrin, Green, Muste, & Wong 1997); the effect of increased labor market competition on wages and availability of jobs (Hainmueller & Hiscox, 2010; Scheve & Slaughter, 2001); and particularly, whether Latino immigrants represent a threat to America’s identity (Schildkraut, 2007; Sierra, Carrillo, DeSipio, & Jones-Correa, 2000; Sniderman, Hagendoorn, & Prior, 2004). Scholars have dedicated much efforts to studying the degree to which Latino immigrants assimilate across generations—among other things, they have looked at immigrants’ linguistic assimilation; adherence to America’s values and principles of freedom, equality of opportunity, and economic self-reliance; and attachment to the American identity—and have found little evidence justifying the last concern (Bean, Brown, & Rumbaut, 2006; Citrin, Lerman, Murakami, & Pearson, 2007; de la Garza, Falcon, & Garcia, 1996; Schildkraut, 2007). However, challenges remain, as increased immigration inflows have led to a surge in the size of the population lacking adequate representation in the democratic process (Sierra et al., 2000). Inclusion in the democratic process is necessary for individuals to be able to protect their interests, express their preferences and opinions, and prevent abuse from government and society (Dahl, 2000, pp. 39-40). But the question of whether (and how) to incorporate immigrants into the democratic process does not have an easy answer. During the period of European immigration, political assimilation and voting rights were “the

Levin

537

American solution” (Dahl, 2000, p. 152) to the inclusion problem and immigrants were allowed to vote in federal elections (García Bedolla, 2009, pp. 24-25). In contrast, a large proportion of today’s immigrants are not allowed access to the franchise. With a few exceptions at the municipal level, only U.S. citizens can vote in national, statewide, and local elections, but many immigrants are not eligible to become citizens due to lack of legal status, temporary immigration status, or short length of residence in the United States.2 Moreover, a large proportion of immigrants who are eligible to naturalize do not apply for citizenship or are unable to complete the application process (DeSipio, 1987; Pachon, 1987; Yang, 1994); and a large proportion of those who do naturalize and become citizens do not vote (DeSipio, 1996, 2011).3 But participation in the electoral process—by voting or contributing money or work to political campaigns—is not the only channel that can be used to affect political outcomes. Immigrants who are not yet naturalized may try to make their interests and needs heard by different means—among other things, they can participate in groups or organizations, contact local government officials, and join protests or boycotts. Yet the fact that nonnaturalized immigrants may participate in nonelectoral activities does not seem to contribute much to solving the political inclusion problem. In the past, it has been found that nonnaturalized Latino immigrants lag behind their citizen counterparts in terms of involvement in nonelectoral political activities (Leal, 2002; Verba, Schlozman, & Brady, 1995). Why do naturalized Latino immigrants participate more than their nonnaturalized counterparts in nonelectoral political activities? Nonnaturalized immigrants sometimes stay away from politics because they feel it is not appropriate for immigrants to have a say in American politics. Moreover, nonnaturalized immigrants may lack interest in American politics and fear detainment or deportation, making them less likely to get involved in political activities. Going through the naturalization process may help compensate some of the disadvantages that immigrants have in terms of civic and language skills; and lead to an acceleration of the process of assimilation into mainstream society and to greater attachment to American identity, values, and democratic traditions (Alvarez, 1987; Yang, 1994). Thus, acquiring citizenship through naturalization may have positive impact on immigrants’ involvement in nonelectoral political activities. If this is the case, naturalization could be understood as the first step toward the inclusion of Latino immigrants in the broader democratic process—above and beyond the electoral arena. According to the U.S. Bureau of Citizenship and Immigration Services, to be eligible for naturalization a person must have been a permanent resident of

538

American Politics Research 41(4)

the United States (green card holder) for at least 5 years; be able to read, write, and speak English; have knowledge and understanding of U.S. history and government; and be a person of “good moral character, attached to the principles of the Constitution of the United States.” The specific procedure that must be followed includes submitting a naturalization application together with supporting documents and filing fees; passing a background check; succeeding in a test of English language and civic skills4; and taking the oath of allegiance in a public ceremony. Going through the naturalization process is not only costly in terms of time and money, it also requires minimal civic and English skills, ability to complete paperwork and follow bureaucratic procedures, and entails considerable “psychological costs” (North, 1987; Yang, 1994). Fears and uncertainty about the naturalization process may prevent some individuals from seeking citizenship (DeSipio, 1987; Pachon, 1987). The experience of successfully going through the naturalization process may create attachment to American institutions, raise awareness of civic duties, and increase knowledge of U.S. history and government.5 All or some of these factors may contribute to increased levels of political engagement after naturalization (Alvarez, 1987). Yet naturalization might not have mobilization effects. Observed differences in political engagement and participation rates between naturalized and nonnaturalized individuals may be explained by ex ante differences in individual attributes and contextual factors. On average, individuals who are eligible and choose to become citizens have longer length of residence in the host country (Bernard, 1936; DeSipio, 1987; Evans, 1988), higher socioeconomic status (Bernard, 1936; DeSipio, 1987), higher levels of English proficiency, and migrated for different reasons, relative to nonnaturalized immigrants. Moreover, they do not come from the same countries and regions, nor choose to settle in the same areas, as those who do not naturalize (DeSipio, 1987; Yang, 1994). Factors affecting the propensity to naturalize include distance to the country of origin, with individuals from Canada and Mexico being the least likely to naturalize (Grebler, 1966; Sierra et al., 2000); whether the country of origin allows dual citizenship (Jones-Correa, 2001); social context and urban concentration in the settlement region (Yang, 1994); and gender, due to Latinas’ higher involvement in community organizations (Pantoja & Gershon, 2006). After controlling for differences between those who naturalize and those who do not; is naturalization still positively related to civic engagement? There are at least two reasons why this might not be the case. First, acquiring citizenship status and voting rights through naturalization does not imply that immigrants will trust American institutions, be interested in American political affairs, or feel entitled or knowledgeable enough to vote

Levin

539

in local or federal elections. Many of those who seek naturalization do so for nonpolitical reasons, such as access to welfare benefits and restricted occupations, the conveniences of traveling abroad with a U.S. passport, and lower requirements for sponsoring the immigration of foreign-born relatives (Alvarez, 1987; Bloemraad, 2006a, 2006b; DeSipio, 2006; Jasso & Rosenzweig, 1986; Yang, 1994). Other immigrants naturalize for politically motivated reasons, or to acquire civil rights. For instance, restrictive immigration and welfare laws affecting the position of legal residents, and hostile antiimmigrant environment in some states, have created concerns and uncertainty among legal citizens about their vulnerable immigration status, and driven immigrants to pursue U.S. citizenship (Balistreri & Van Hook, 2004; DeSipio, 2011; Pantoja, Ramirez, & Segura, 2001, 2008; Sierra et al., 2000).6 However, since immigrants may naturalize for many reasons unrelated to civil rights, the sole acquisition of citizenship status does not necessarily imply greater civic engagement or that newly naturalized citizens will be drawn to participate in the democratic process. Second, scholars have found that longer contact with mainstream society and acculturation do not necessarily lead to greater levels of trust in government (Abrajano & Alvarez, 2010a; Michelson, 2001, 2003, 2007). Explanations include that new immigrants have not yet been exposed to the low levels of trust in government prevailing among the majority of Americans; they are less familiar with hostility and discrimination against immigrants and ethnic minorities because of their lower contact with mainstream American society; they are less aware of structural inequalities and lower opportunities for upward mobility available for ethnic minorities; and they are likely to hold an optimistic and idealistic view of the American society and political system, especially when comparing social and political conditions in the United States with those prevailing in some of their countries of origin (Abrajano & Alvarez, 2010a; Michelson, 2001, 2003, 2007). If nonnaturalized immigrants are more trusting due to their lower levels of assimilation, and if trust and civic engagement reinforce each other, as has been argued in the past (Brehm & Rahn, 1997; Putnam, 2000), then it is possible that the higher levels of trust among the nonnaturalized compensate for the lack of politically relevant resources and make them— everything else constant—as or more likely to participate in nonelectoral activities as their naturalized counterparts. In this article, I study whether the acquisition of the American citizenship through naturalization contributes to greater involvement of Latino immigrants in the democratic process, by comparing the level of civic engagement of Latino immigrants who have become citizens with that of immigrants who have not yet naturalized using data from the 2006 Latino National Survey (LNS; Fraga et al., 2006a). I focus on a series of nonelectoral activities that

540

American Politics Research 41(4)

include contacting government officials and working to solve problems with others informally or through existing groups and organizations, as well as participation in community activities such as volunteering at schools.7 To control for prenaturalization differences in observed individual attributes and contextual factors that may explain part of the observed differences in political engagement, I use matching methods to balance the distribution of prenaturalization variables between naturalized and nonnaturalized individuals, and then use the matched sample to measure the causal effect of naturalization on participation in nonelectoral activities. Why is it important to study the effect of naturalization on civic engagement? Recent years have seen an increase in the number of immigrants seeking naturalization. Immigration laws granting amnesty to illegal immigrants, allowing them to become legal residents and naturalize after the usual 5-year period, such as the 1986 Immigration Reform and Control Act (IRCA), led to an increase in the number of individuals seeking naturalization (Pantoja et al., 2001). More recently—in the aftermath of the 2006 nationwide immigrant rights protests—naturalization drives by Latino civic organizations might have resulted in an increase in naturalization rates (Barreto, Manzano, Ramírez, & Rim, 2009a; Benjamin-Alvarado, DeSipio, & Montoya, 2008; DeSipio, 2011; Félix, González, & Ramírez, 2008). But many immigrants remain who are not eligible to naturalize; this has led to an increased awareness of the inclusion problem, and to a variety of proposals that include providing a clear pathway to legal status and naturalization for undocumented immigrants and temporary immigrant workers (Ewing, 2010; Galston, Pickus, & Skerry, 2009; Waslin, 2004). A serious consideration of this proposal requires studying whether naturalization is indeed an effective tool for facilitating immigrant incorporation, and whether more than a pathway to legal status and naturalization is needed to address the inclusion problem. Studying civic engagement among Latino immigrants is also important because doing so helps us learn more about the determinants of civic engagement more in general. To date, most studies of civic engagement explain involvement in political activities as a function of factors tied to demographic and socioeconomic attributes thought to affect individuals’ abilities to bear participation costs (Wolfinger & Rosenstone, 1980; Verba et al., 1995). In the case of immigrants—who start the process of socialization into American mainstream society later in life relative to the native born—it is also important to consider preimmigration political involvement, social and political conditions in the country of origin and settlement area (Pearson-Merkowitz, 2012; Rocha & Espino, 2010), as well as postimmigration socialization experiences in the host country

Levin

541

that might affect the development of political attitudes and behaviors (Alvarez & García Bedolla, 2003). As noted by Tam Cho (1999, p. 1144), “immigrants who travel through different socialization channels may have a very different cost and benefit structure relative to native-born Americans.” Studying the political behavior of Latino immigrants (not only those who have acquired citizenship, but also those who cannot or have not yet naturalized) allows learning a great deal about the relationship between participation, socialization experiences, and contextual factors, something that cannot be done by only looking at Latinos citizens or registered voters as is usually done in studies of political participation among Latinos. The structure of the article is as follows: First, I introduce the data used in the analysis, and discuss some of the main observable differences between the naturalized and nonnaturalized population. After that, I describe the methodological procedure used in this article and discuss the results of the quantitative analysis. The main finding of the article is that naturalization has no significant effects on most types of nonelectoral participation. While naturalized respondents are more likely to contact government officials, they are not more likely to join social or political groups, nor to participate in community building activities such as getting together with others to solve problems that need to be addressed—either informally or through organizations. Although the enfranchisement of a large proportion of the immigrant population surely contributes to drawing individuals into the electoral process—62.3% of naturalized respondents reported having voted in the 2004 presidential election— the results of this article suggest that naturalization alone is unlikely to have a major impact on the overall civic engagement of Latino immigrants.

A First Look at the Latino Immigrant Population I use data from the 2006 LNS (Fraga et al., 2006a, 2006b, 2012), a unique source of information about political attitudes and behaviors of Latino individuals living in the United States. The 2006 LNS is the most recent survey study of its kind and distinctive due to the large sample size and coverage of the U.S. native- and foreign-born Latino population. Jurisdictions considered in the analysis comprised 87.5% of the U.S. Latino population. Bilingual interviews were conducted by telephone between November 2005 and August 2006 in 16 states (including the District of Columbia), and states were chosen according to the size and recent evolution of the Latino population. In total, the study contains 8,634

542

American Politics Research 41(4)

interviews to Latino residents of any nationality. Since the objective of this article is to study the impact of naturalization, I restrict attention to 4,723 interviews conducted with foreign-born respondents.8 In Table 1, I present summary statistics of a broad set of individual attributes considered in the analysis. The first and second columns give average covariate values for naturalized (C = 1) and nonnaturalized (C = 0) respondents, respectively. Entries in the third column are average differences between both groups, and those in the fourth column are bootstrap p-values of a Kolmogorov−Smirnov test used to determine whether there were significant differences in the distribution of covariates across groups. Table 1 shows that foreign-born citizens differ significantly from immigrants who are not citizens in multiple aspects: sociodemographic attributes, exposure to the 2006 political environment, immigration experience, settlement area, reported reasons for immigrating to the United States, preimmigration experience, and region of origin. Naturalized immigrants tend to be older (47 years old, compared to 37 years old), have higher educational attainment (high school graduate, compared to less than a high school diploma), higher level of household income (around US$38,000, compared to about US$25,000), greater Internet access (52%, compared to 27%), are more likely to own their homes (67%, compared to 35%), and to participate in the military themselves or a family member (26%, compared to 8%), relative to their nonnaturalized counterparts. Due to their higher socioeconomic status and involvement in American institutions such as the military, naturalized Latinos are likely to possess higher levels of politically relevant resources and civic skills, which in turn might allow them to more easily bear participation costs and result in higher levels of political engagement (Verba et al., 1995). Since the survey was conducted during a period of heated debate centered around immigration reform and massive protests against anti-immigrant legislation being considered by Congress in 2006, I consider differences in interview dates—taking into account whether interviews were conducted before or after the start of the 2006 immigrant rights marches. I find that while 62% of the naturalized were interviewed on or after March 25—the beginning of the cycle of protests—only 56% of the nonnaturalized were interviewed on or after that date. In the multivariate analysis, it is important to control for these differences as individuals interviewed on or after March 25 might be more likely to experience stronger identification with the American identity (Silber Mohamed, 2010), which in turn might affect their level of civic engagement. Spanish language ethnic media played a key role in mobilizing Latinos ahead and after the start of the 2006 immigrant rights protests (Abrajano & Alvarez,

543

Levin Table 1. Summary Statistics (All Foreign-Born Respondents, N = 4,642).

Sociodemographic attributes                       2006 context   Immigration experience         Settlement area           Reasons for immigration           Preimmigration experience     Region of origin                 a

Female (%) Age (years old) Educational attainment (1-8) Income (1-7) Internet access (%) Homeowner (%) Employment status (1-3) Military (%) Catholic (%) Church attendance (1-5) Latino spouse Non-Latino spouse Interviewed March 25 or later (%) Spanish language media (1-3) U.S. education (%) Kid in U.S. school (%) Length of residency in the U.S. (years) Length of residency at home (years) Reliance on Spanish (1-3) North East (%) Central (%) South Atlantic (%) South Central (%) Mountain West (%) Pacific (%) Education (%) Reunification (%) Political (%) Brought by parents (%) Economic (%) Other (%) Voted before immigration (%) Political org. before immigration (%) Parents education (1-5) Dual citizenship Mexico: North (%) Mexico: West (%) Mexico: East and Center-North (%) Mexico: D.F. and Mexico State (%) Mexico: South and Center-South (%) Cuba (%) Central America and Dom. Rep. (%) South America (%)

Bootstrap p-value of Kolmogorov−Smirnov’s test.

C=1

C=0

Diff.

p-Valuea

53.2 46.9 4.8 3.8 52.0 66.5 2.3 25.5 72.3 3.4 51.0 12.0 61.7 2.2 44.7 34.2 29.0

54.8 37.1 3.8 2.5 27.1 35.1 2.4 8.2 76.1 3.4 51.6 5.3 55.5 2.7 13.6 36.1 15.2

−1.6 9.7 1.0 1.3 24.9 31.4 0.0 17.3 −3.8 0.0 −0.6 6.7 6.2 −0.5 31.0 −1.8 13.9

.311 .000 .000 .000 .000 .000 .038 .000 .003 .022 .701 .000 .000 .000 .000 .210 .000

3.7 2.2 15.5 11.0 28.8 11.4 13.6 19.7 6.4 8.1 13.1 22.5 41.4 8.6 28.9 10.8 1.8 77.6 17.2 15.8 12.2 4.7 3.8

2.9 2.6 10.6 11.1 27.3 14.1 17.1 19.8 5.6 10.6 3.7 7.7 65.9 6.5 48.1 13.3 1.5 90.0 17.6 16.9 21.4 8.0 10.2

0.8 −0.4 4.9 −0.2 1.5 −2.7 −3.5 −0.1 0.8 −2.4 9.4 14.7 −24.5 2.1 −19.2 −2.6 0.3 −12.4 −0.4 −1.0 −9.2 −3.4 −6.4

.000 .000 .000 .868 .308 .011 .003 .966 .303 .010 .000 .000 .000 .010 .000 .017 .000 .000 .768 .352 .000 .000 .000

14.2 21.3 10.8

2.6 16.7 6.6

11.6 4.6 4.2

.000 .001 .000

544

American Politics Research 41(4)

2010b; Barreto et al., 2009a; Benjamin-Alvarado et al., 2008; DeSipio, 2011; Félix et al., 2008). While naturalized immigrants rely on both Spanish and English language media, their nonnaturalized counterparts are more reliant on Spanish language media only. In the multivariate analysis, I also control for the level of exposure to Spanish language media, as this might affect exposure to information about the 2006 immigrant rights protests and motivation to participate in nonelectoral political activities. Moving to the experience of living as an immigrant in the United States, measures of socialization into mainstream society differ considerably between both groups; those who naturalize are much more likely to have acquired some of their education in the United States (45%, compared to 14%), be married to a non-Latino spouse (12%, compared to 5%), have considerably longer length of residency in the host country (29, compared to 15 years), have spent more time leaving in the same home (4, versus 3 years), and are less likely to communicate in Spanish. These differences suggest that naturalized immigrants have more opportunities to integrate into the mainstream society by familiarizing themselves with American history and culture, and through contact with native-born citizens at home, school, and their local community; all or part of which may lead to higher levels of political involvement by naturalized citizens relative to their nonnaturalized counterparts. It is also the case that immigrants who obtain citizenship choose to settle in different areas, and come from different countries and regions, relative to nonnaturalized immigrants. While 31% of the latter live in south-central states or in the Mountain West, only 25% of those who naturalize settle in that region. Instead, naturalized immigrants are more likely to settle in north-eastern states. Since the immigration experience varies considerably depending on characteristics of the settlement area—such as concentration of Latino people, local laws, and presence of grassroots organizations that play an important role in mobilizing Latino residents (Bada, Fox, Donnelly, & Selee, 2010)— differences in the geographic distribution of naturalized and nonnaturalized are likely to have substantial impact on the political behavior of both groups. Coming from a country that does not allow dual citizenship is likely to affect the chances that an individual chooses to naturalize (Jones-Correa, 2001), as well as the likelihood of involvement in political activities as feelings of owed loyalty toward the home country may depress their levels of civic engagement in the United States (Staton, Jackson, & Canache, 2007). I find that while 78% of the naturalized come from a country that allows dual citizenship, as many as 90% of nonnaturalized Latino immigrants do so. The explanation of this seemingly counterintuitive difference is that 67% respondents considered in the analysis come from Mexico—a country that allows

Levin

545

dual citizenship—and Mexicans are considerably less likely to naturalize than immigrants from other national origins. After controlling for the lower likelihood of naturalization among Mexicans, individuals coming from countries that allow dual citizenship may be more likely to seek naturalization relative to those coming from countries that do not allow dual citizenship; therefore, it is important to control for this variable in the multivariate analysis. In the past, scholars have noted the importance of considering heterogeneities of national origin among Latinos, and have found that there are considerable differences in political behavior across national origin groups (Alvarez & García Bedolla, 2003; Diaz, 1996; Hero & Campbell, 1996). Accordingly, in studying the impact of naturalization, it is necessary to take into account that naturalization propensities vary considerably across immigrants from different national origins. While 74% of nonnaturalized respondents come from Mexico, only 54% of the naturalized do. Conversely, while only 3% of nonnaturalized respondents come from Cuba, 14% of the naturalized do. Since the sample size is small for most specific countries, I grouped immigrants who have neither Mexican nor Cuban origin into regions: Central America (18% of the sample, comprising 7 countries), and South America (8% of the sample, comprising 9 countries). The proportions of immigrants coming from these two regions are larger in the naturalized subsample (21% and 11%) than in the nonnaturalized subsample (17% and 7%). Immigrants from these two regions tend to naturalize more than Mexicans but less than Cubans.9 It is also important to take into account heterogeneities among immigrants coming from different Mexican regions. Studies of Mexican politics tend to pay special attention to regional factors due to subnational differences in political context and varying degrees of democratization across regions—in particular, between the wealthier and more industrialized North and the less developed South (Cortina, Gelman, & Lasala Blanco, 2008; Hiskey & Bowler, 2005; Klesner, 2007). Exposure to different political cultures before immigration may affect the propensity to seek naturalization and participate in politics after immigration, even after controlling for sociodemographic attributes and other factors. Since the immigrant sample includes many Mexican respondents, it is possible to consider differences in naturalization rates across Mexican regions. Individuals from the South are considerably less likely to naturalize (only 14% do), compared to other Mexican immigrants. Immigrants coming from the Center of the country are more likely to naturalize (21% do), but naturalization rates are still low compared to immigrants from Central America. Only immigrants coming from the North of the country have naturalization rates comparable to those prevailing among Central Americans (31%, compared to 34% among the latter). Due to these

546

American Politics Research 41(4)

differences, individuals coming from the Center and South of Mexico are considerably underrepresented in the naturalized sample (21% of naturalized respondents come from these regions, compared to 39% of nonnaturalized). When asked about the main reason for having migrated to the United States, 77% of those respondents who are not citizens say that they migrated to improve the economic situation or for family reunification, compared to only 50% of naturalized respondents who report similar reasons. The latter are more likely to have migrated for political reasons, or to have been brought to the United States by their parents as a child. These differences in immigration motives are not only likely to affect the likelihood of naturalization, but also willingness to assimilate into mainstream American society and participation in political activities. Finally, there are also important differences in preimmigration levels of political participation across groups. Particularly, nonnaturalized immigrants exhibit higher levels of political involvement in their home country: 48% reported that they voted before coming to the United States and 13% reported being active in a political party, political organization, or other type of organization before immigration, compared to 29% and 11% of participation in similar home-country activities among naturalized respondents. It is unclear whether preimmigration political activity should promote or discourage engagement in American politics. Still, since preimmigration experiences vary considerably across citizenship status, it is important to take these differences into account when studying the impact of naturalization on political participation. Table 2 reports average differences in nonelectoral political participation between naturalized and nonnaturalized immigrants. When asked whether they participated in the activities of a social, cultural, civic, or political group, 23% of the naturalized reported doing so, compared to only 10% of the nonnaturalized. Differences in participation rates are specially pronounced for groups that include White members—that is, whose members are mostly White, mostly Latino and White, or a mix of all ethnicities—in which case naturalized immigrants participated almost three times more than nonnaturalized immigrants. In contrast, when asked whether they would work together with others to solve issues or problems that need to be addressed, there were no considerable differences in participation rates between naturalized and nonnaturalized immigrants (each group answered affirmatively 76% and 79% of the time, respectively). Also, there were no significant differences in the channels used to solve imminent problems, either through existing groups or organizations, or by getting together informally. Finally, when asked whether they ever tried to get government officials to pay attention to

547

Levin

Table 2. Nonelectoral Participation (%, All Foreign-Born Respondents, N = 4,642).

Participation in groups   Including White members   Only minority members Solve problems with others   Through groups/organizations   Get together informally Contact government officials   Latino officials   Non-Latino officials School volunteerb

C=1

C=0

Diff.

p-Valuea

23.0 15.5 7.4 76.4 42.0 46.1 36.7 17.3 19.4 57.9

10.3 5.3 5.0 78.6 42.5 46.9 17.6 11.5 6.1 44.1

12.7 10.3 2.5 −2.2 −0.5 −0.8 19.1 5.8 13.3 13.7

.000 .000 .001 .099 .771 .626 .000 .000 .000 .000

a

Test of difference in proportions. Only includes respondents with kids attending school in the United States.

b

something that concerned them, 37% of the naturalized reported calling, writing a letter or going to a meeting, compared to only 18% of the nonnaturalized. Differences were particularly pronounced when it came to contacting non-Latino officials, where naturalized immigrants reported doing so three times more frequently than nonnaturalized immigrants. Finally, among immigrants who have kids attending school in the United States, naturalized immigrants were more likely to volunteer for school activities relative to nonnaturalized immigrants (58%, compared to 44%). Differences in participation rates cannot be fully attributed to differences in citizenship status, because the effect of naturalization is likely to be confounded by individual attributes that vary systematically between naturalized and nonnaturalized immigrants. In the next section I measure the causal effect of naturalization using matching methods to control for differences in observable individual attributes.

Causal Effect of Naturalization The analysis done in this article is based on observational survey data, rather than data collected during a controlled experiment. This is so because there was no control over the acquisition of citizenship of immigrants who answered the survey, and thus citizenship may vary systematically across respondents as a function of factors such as sociodemographic attributes and immigration experience. Obtaining reliable measures of causal effects based

548

American Politics Research 41(4)

on observational data requires controlling for prenaturalization differences between respondents who have become citizens and those who have not. The main difference between the methodological approach used in this article and that used in previous studies of the impact of naturalization is that, instead of using the traditional regression approach, I use matching methods to control for the nonrandom assignment of immigrants into citizenship status.10 Matching can help improve the quality of causal inferences relative to the regression approach, because it requires the researcher to think carefully about causal relationships, and also because it results in measures of causal effects that are less dependent on modeling assumptions (Ho et al., 2007a). A straightforward way of computing naturalization effects controlling for prenaturalization differences would be to stratify the sample based on individual attributes such as length of residence, English proficiency, and other variables that affect both naturalization and civic engagement, and that therefore we would wish to control for. Subsequently, we could compute the effect of naturalization within each strata, and then compute a weighted average of naturalization effects. An obvious limitation of this approach is that it is vulnerable to dimensionality issues, as it can lead to an excessive number of small-sized strata as it becomes necessary to control for additional covariates, such as sociodemographic characteristics and other relevant individual attributes. An alternative procedure that avoids the dimensionality problem is matching on the probability of naturalization conditional on covariates, P(ci = 1|xi), where xi is a vector containing the value of relevant prenaturalization variables for individual i. This procedure is called “propensity score matching” and can be used to produce samples of naturalized and nonnaturalized respondents where prenaturalization attributes are similarly distributed or “balanced” across groups, such that differences in group behavior can only be attributed to unequal citizenship status. In this article, I estimate the propensity score using a logistic regression.11 In general, nearest available propensity score matching consist of the following steps (Cochran & Rubin, 1973): (1) sort observations in the treatment group in random order; (2) starting with the first observation in sorted treatment group, match it to the observation in the control group that is closest in terms of propensity score; (3) consider the second observation in the sorted treatment group, and match it to the nearest neighbor in the control group among the remaining observations; and so on. To achieve better covariate balance across groups, I used exact matching for a subset of covariates, and within the subset of observations that were exact matches, observations were matched using nearest neighbor propensity score matching with a caliper of size k; such that for any observation in the treatment

Levin

549

group with propensity score Pi1, and observations in the control group with propensity score Pj0, I searched for observations in the control group such that Pi1- Pj0 < k.12 The success of propensity score matching should not be evaluated according to how well the propensity score model explains observed naturalization levels, but according to how well matching on P(c=1|x) helps balance the distribution of covariates among naturalized and nonnaturalized individuals (Ho et al., 2007a). To evaluate balance improvement, I first examined the distribution of the propensity score in the treatment and control group before and after matching. The left-most plot in Figure 1 gives the distribution of propensity scores among nonnaturalized and naturalized individuals in the original sample. Clearly, immigrants who have acquired citizenship are not comparable to those who have not, because ex ante (i.e., given observed individual attributes that are causally prior to naturalization) they are much more likely to acquire citizenship through naturalization than their nonnaturalized counterparts. To solve this issue, I used the propensity score procedure described above to match individuals who are as similar as possible in all their attributes, except for their citizenship status. The right-most plot in Figure 1 gives the distribution of the propensity scores among nonnaturalized and naturalized individuals in the matched sample.13 This plot suggests that the matching procedure was highly successful in balancing the likelihood of naturalization. Since balance is a “multivariate concept” (Ho et al., 2007), it is important to achieve balance not only in a low-dimensional measure like the propensity score, but in all relevant aspects of the multivariate distribution of covariates (Ho et al., 2007a; Sekhon, 2009). Therefore, I also evaluated balance by comparing moments of the distribution of covariates included in the matching procedure.14 Table 3 reports postmatching summary statistics for the same set of individual attributes discussed in the summary statistics section.15 The structure of Table 3 is similar to that of Table 1, except that there is one additional column (the first one) that gives the postmatching balance improvement, computed as the reduction in difference in covariate means across groups, divided by the original difference. According to the balance statistics shown in Table 3, all covariates underwent extensive balance improvement (always larger than 39%) and there was a substantial reduction in observable differences between both groups (all p-values increase considerably).16 Using the matched sample of naturalized and nonnaturalized respondents, I proceeded to estimate the causal effect of naturalization. Table 4 gives average differences in participation rates across immigrant types. After matching, I find that there are no significant differences in

550

Figure 1. Distribution of Propensity Scores.

551

Levin

Table 3. Balance Improvement (Matched Foreign-Born Respondents, N =1,680). B.I. (%) C = 1 Sociodemographic attributes                       2006 Context   Immigration experience    

Female (%)

Age (years old) Educational attainment (1-8) Income (1-7) Internet access (%) Homeowner (%) Employment status (1-3) Military (%) Catholic (%) Church attendance (1-5) Latino spouse Non-Latino spouse Interviewed March 25 or later (%) Spanish language media (1-3) U.S. education (%) Kid in U.S. school (%) Length of residency in the U.S. (years)   Length of residency at home (years)   Reliance on Spanish (1-3) Settlement area North East (%)   Central (%)   South Atlantic (%)   South Central (%)   Mountain West (%)   Pacific (%) Reasons for immigration Education (%)   Reunification (%)   Political (%)   Brought by parents (%)   Economic (%)   Other (%) Preimmigration experience Voted before immigration (%)   Political org. before immigration (%)   Parents education (1-5) Region of origin Dual citizenship   Mexico: North (%)   Mexico: West (%)   Mexico: East and Center-North (%)   Mexico: D.F. and Mexico State (%)   Mexico: South and Center-South (%)   Cuba (%)   Central America and Dom. Rep. (%)   South America (%) a

Bootstrap p-value of Kolmogorov–Smirnov’s test.

C=0

Diff.

p-Valuea

39.5

54.6

53.7

1.0

.711

92.9 91.1 95.0 99.0 98.9 97.4 95.2 93.8 82.9 100.0 89.3 92.4 95.7 98.8 87.0 97.0

42.7 4.2 3.3 41.8 57.4 2.4 17.6 74.6 3.4 54.2 7.7 60.6 2.4 31.2 39.5 23.8

42.0 4.1 3.2 41.5 57.7 2.4 16.8 74.9 3.4 54.2 7.0 61.1 2.4 30.8 39.3 23.4

0.7 0.1 0.1 0.2 −0.4 0.0 0.8 −0.2 0.0 0.0 0.7 −0.5 0.0 0.4 0.2 0.4

.179 .094 .750 .933 .909 .930 .664 .934 .784 1.000 .612 .857 .150 .711 .940 .097

98.6 90.2 58.7 100.0 75.9 51.4 89.7 100.0 69.0 61.1 92.4 95.2 98.1 88.5 91.3 53.5 92.8 95.2 100.0 20.6 97.4 96.4 94.4 96.9 81.8 91.5

3.4 2.4 13.3 11.3 23.6 13.3 17.3 21.2 5.7 10.6 7.1 16.0 52.7 7.9 34.9 11.2 1.6 86.4 17.5 19.3 16.7 5.2 5.8 5.7 21.4 8.3

3.5 2.4 11.3 11.3 23.9 14.6 17.6 21.2 6.0 9.6 6.4 16.7 53.2 8.1 33.2 10.0 1.6 87.0 17.5 20.1 16.9 5.4 6.2 5.4 20.6 8.0

0.0 0.0 2.0 0.0 −0.4 −1.3 −0.4 0.0 −0.2 1.0 0.7 −0.7 −0.5 −0.2 1.7 1.2 0.0 −0.6 0.0 −0.8 −0.2 −0.1 −0.4 0.4 0.8 0.4

.838 .122 .216 1.000 .888 .453 .888 1.000 .865 .535 .607 .719 .863 .896 .451 .465 .500 .743 1.000 .692 .932 .950 .807 .766 .671 .817

552

American Politics Research 41(4)

Table 4. Nonelectoral Participation (Matched Foreign-Born Respondents, N = 1,680).

Participation in groups   Including White members   Only minority members Solve problems with others   Through groups/organizations   Get together informally Contact government officials   Latino officials   Non-Latino officials School volunteerb

C=1

C=0

Diff.

p-Valuea

16.8 10.7 6.1 76.3 41.9 46.4 30.0 15.6 14.4 52.1

14.6 9.6 5.0 77.0 44.2 44.0 22.1 12.5 9.6 49.1

2.1 1.1 1.1 −0.7 −2.3 2.4 7.9 3.1 4.8 3.0

.254 .519 .393 .773 .375 .352 .000 .079 .003 .485

a

Test of difference in proportions. Only includes respondents with kids attending school in the United States.

b

participation in the activities of social, cultural, and political groups. This is not only the case for groups that only include minority members, but also for groups that have White members. Similarly, differences in respondents’ willingness to solve problems with others—either through existing groups or organizations, or informally—remain nonsignificant. Nonetheless, even after accounting for prenaturalization differences, naturalized citizens are 8 percentage points more likely to contact officials relative to immigrants who have not yet acquired citizenship. This difference is particularly strong when it comes to contacting non-Latino officials. Finally, after matching, there is no significant difference in the likelihood of involvement in parental school activities between naturalized and nonnaturalized immigrants. Even though the matching procedure successfully reduced bias in all covariates across naturalized and nonnaturalized groups, some differences remain for certain covariates. To control for any remaining bias in observed covariates, I conducted a postmatching model-based adjustment of treatment effects—that is, I estimate logistic regressions in the matched data set, controlling for the same covariates included in the matching procedure. Compared to regression models estimated in unmatched data, model-based adjustment in matched samples leads to estimates of treatment effects that are less dependent on the assumed functional form (Ho et al., 2007a; Rosenbaum & Rubin, 1983), and this is why they can be used to control for

553

Interviewed March 24 or later Female Age Educational attainment Income Internet access Homeowner Employment status Military Catholic Church attendance Latino spouse Non-Latino spouse U.S. education Kid in U.S. school Length of residency in the U.S. Length of residency at home Reliance on Spanish Spanish media Dual citizenship Why immig.: reunificationb Why immig.: politicalb Why immig.: brought by parentsb

Naturalization

 

β

σ

0.35 −0.39 0.00 0.23 0.06 0.22 0.28 −0.24 0.36 −0.36 0.35 −0.19 0.19 0.49 0.07 0.00 0.09 −0.15 −0.24 0.15 0.33 −0.56 −0.25

0.16 0.16 0.01 0.05 0.05 0.17 0.18 0.10 0.18 0.16 0.07 0.18 0.28 0.20 0.16 0.01 0.07 0.10 0.11 0.30 0.34 0.44 0.32

0.12 0.15

Any

0.32 −0.42 0.00 0.23 0.08 −0.01 0.36 −0.30 0.13 −0.43 0.24 −0.21 0.32 0.55 0.07 0.00 0.09 −0.27 −0.28 −0.29 0.55 −1.10 −0.08

0.06

β 0.20 0.20 0.01 0.06 0.06 0.20 0.21 0.12 0.22 0.19 0.08 0.22 0.31 0.24 0.20 0.01 0.09 0.11 0.13 0.35 0.40 0.59 0.38

0.18

σ

Nonminority

Groups

σ

β

σ

β

σ

0.27 −0.26 0.01 0.21 0.00 0.48 0.02 −0.05 0.60 −0.01 0.54 −0.14 −0.31 0.25 0.14 0.00 0.09 0.11 −0.10 1.26 −0.04 0.10 −0.43

0.25 0.25 0.01 0.07 0.08 0.26 0.26 0.15 0.26 0.27 0.12 0.27 0.50 0.31 0.25 0.02 0.11 0.16 0.18 0.64 0.52 0.61 0.52

−0.30 0.01 −0.02 0.18 0.06 0.12 0.06 −0.06 0.05 0.17 0.22 −0.10 −0.19 0.25 0.53 0.01 0.09 0.46 −0.11 0.14 0.02 0.00 0.09 0.13 0.14 0.01 0.04 0.04 0.14 0.14 0.08 0.17 0.14 0.05 0.14 0.25 0.18 0.14 0.01 0.06 0.09 0.10 0.27 0.33 0.38 0.31

−0.22 0.13 0.00 0.11 0.01 0.20 0.13 −0.15 0.28 −0.03 0.12 −0.20 −0.06 0.20 0.09 −0.01 0.03 0.26 −0.15 0.05 0.14 0.09 0.03 0.11 0.11 0.01 0.03 0.03 0.12 0.12 0.07 0.14 0.12 0.04 0.12 0.22 0.15 0.11 0.01 0.05 0.07 0.08 0.22 0.27 0.33 0.26

σ

0.01 0.09 −0.02 0.01 0.02 −0.02 −0.07 0.17 −0.24 0.25 0.08 0.04 −0.09 −0.07 0.15 0.01 0.01 0.08 0.02 0.12 0.22 −0.01 0.26

0.11 0.11 0.01 0.03 0.03 0.12 0.12 0.07 0.14 0.12 0.04 0.12 0.21 0.15 0.11 0.01 0.05 0.07 0.08 0.22 0.27 0.33 0.25

0.11 0.10

β

Some way Groups/Org. Informally

0.16 0.23 −0.04 0.12 −0.10 0.10

β

Minority

Solve problems with others

Table 5. Regression Adjustment (Matched Foreign-Born Respondents, N = 1,680).

β

σ

β

0.16 −0.07 −0.01 0.15 0.09 0.40 0.10 0.13 0.60 −0.07 0.08 −0.08 0.15 0.30 0.06 0.02 0.12 0.16 0.03 0.01 0.02 −0.08 −0.29

0.13 0.13 0.01 0.04 0.04 0.13 0.14 0.08 0.15 0.14 0.05 0.14 0.24 0.17 0.13 0.01 0.06 0.08 0.09 0.25 0.30 0.37 0.29

0.23 −0.15 0.00 0.06 0.03 0.38 −0.19 0.14 0.40 −0.04 0.19 0.01 −0.24 0.26 0.13 0.00 0.19 0.11 0.14 0.44 0.03 0.05 −0.40

0.16 0.16 0.01 0.05 0.05 0.16 0.17 0.10 0.19 0.17 0.07 0.17 0.33 0.21 0.16 0.01 0.07 0.10 0.12 0.37 0.36 0.44 0.35

0.15

σ

Latino

0.43 0.12 0.25

Any σ

0.01 0.05 −0.02 0.22 0.12 0.27 0.43 0.07 0.59 −0.05 −0.06 −0.16 0.33 0.18 −0.01 0.03 −0.01 0.20 −0.11 −0.30 0.04 −0.17 −0.05

0.17 0.18 0.01 0.05 0.05 0.18 0.19 0.11 0.19 0.19 0.07 0.20 0.29 0.22 0.18 0.01 0.08 0.11 0.12 0.32 0.41 0.50 0.38

0.48 0.16

β

σ 0.19 0.20 0.02 0.06 0.06 0.19 0.20 0.11 0.25 0.21 0.08 0.22 0.36 0.26 —− 0.02 0.09 0.13 0.14 0.37 0.46 0.62 0.43

0.17

(continued)

−0.02 0.70 −0.03 0.20 0.03 −0.12 0.37 0.02 0.12 0.68 0.29 −0.23 −0.02 0.52 —− 0.02 0.19 −0.01 0.01 −0.56 −0.03 0.01 0.03

0.14

β

Non-Latino School volunteera 

Contact government officials

554

0.2

σ

0.6

0.35 0.41 0.23 0.27 0.08 1.03 0.5

−0.06 −0.88 0.20 0.37 −0.09 −7.54

β

b

a

σ

1.7

−2.4 2.4

β

1.0 −0.6

σ 0.24 0.29 0.13 0.17 0.06 0.63

β

2.4

−0.10 −0.13 0.14 0.03 −0.04 −0.71

β

β

Any σ

β

2.4

6.8

2.1

1.9

1.3

0.31 0.42 0.18 0.22 0.08 0.93

σ

Latino

4.9

0.10 0.71 −0.07 −0.23 −0.03 −3.57

β

β

2.1

3.4

0.36 0.59 0.41 −0.21 0.21 −0.07 0.29 0.11 0.08 0.11 0.96 −3.37

σ

3.8

0.40 0.49 0.21 0.29 0.10 1.13

σ

Non-Latino School volunteera 

Contact government officials

0.23 −0.01 0.26 −0.07 0.28 0.09 0.32 −0.54 0.13 0.17 0.15 0.31 0.17 0.25 0.19 0.50 0.06 −0.04 0.06 −0.07 0.62 −3.66 0.72 −4.90

σ

Some way Groups/Org. Informally

0.45 0.21 0.29 0.48 0.67 0.26 0.35 0.49 0.28 0.30 0.15 0.04 0.32 0.28 0.22 0.35 0.12 −0.04 0.07 0.06 1.47 −0.99 0.75 −1.90

σ

Minority

Solve problems with others

Note. Settlement area and region of origin are also included in the regression adjustment. Only includes respondents with kids attending school in the United States. The baseline category is education.

0.7

0.9

β

Adjusted effect of naturalization

σ

Nonminority

0.30 0.07 0.36 0.51 0.19 0.05 0.22 0.49 0.07 0.13 0.87 −2.65

β

Any

Groups

0.01 Why immig.: economicb Why immig.: otherb 0.10 Voted before immigration 0.14 Political org. before immigration 0.50 Parents education 0.07 Constant −3.46

 

Table 5. (continued)

Levin

555

remaining bias without undoing the benefits of measuring causal effects using matching methods instead of a more traditional approach. Table 5 reports the results of the regression adjustment. For each political activity, figures in the first column are logistic regression coefficients, and figures in the second column are standard deviations. Also, the last row gives the adjusted effects of naturalization, measured as the change in simulated probability of participation resulting from a one-step increase in the naturalization indicator. When these adjusted effects are compared to those reported in Table 4, it becomes clear that estimated effects remain very similar after the regression adjustment. Naturalized immigrants are considerably more likely to contact government officials—and particularly nonLatino officials—but there are no significant differences in participation in other nonelectoral political activities or involvement in parental school activities.17 As in any observational study where researchers have no control over treatment assignment, controlling mechanisms like propensity score matching presume that selection is based on observable characteristics. However, it may be the case that naturalization is driven by unobservable individual attributes, which also affect propensities to participate in political activities. These unobservable attributes may include prenaturalization political interest and group consciousness, or determinants of legal status that are not captured by variables included in the matching procedure. When the selection-onobservables assumption is not satisfied the matching procedure may reduce observed differences between the treatment and control group, but increase bias in unobservable attributes. Since the estimated propensity score is treated as a fixed measure during the matching procedure, propensity score matching does not account for the uncertainty regarding whether the procedure was successful in reducing bias in unobservables. To evaluate the robustness of estimated effects to unobservables, I assessed sensitivity to hidden bias using Rosenbaum Bounds sensitivity analysis (Rosenbaum, 1995, Chapter 4).18 Table 6 reports Rosenbaum lower and upper bounds for p-values associated with effect of naturalization on the likelihood of contacting government officials.19 Γ values indicate the size of the omitted factor. For example, when Γ = 1.30, unobservable attributes may cause someone to be 30% more likely to contact officials, relative to someone else who looks similar based on observables. If the impact of naturalization remains significant for large values of Γ, we can conclude that effects measured using matching are robust to the existence of relevant unobservable attributes. Results given in Table 6 suggest that the impact of naturalization on the likelihood of contacting government officials (of any ethnicity) may become insignificant when Γ = 1.30 and that the

556

American Politics Research 41(4)

Table 6. Sensitivity Analysis. Contact government officials   Γ 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40 1.45 1.50

Any

Latino

Non-Latino

L

U

L

U

L

U

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

0.000 0.001 0.003 0.010 0.026 0.058 0.111 0.189 0.289 0.404 0.524

0.040 0.018 0.007 0.003 0.001 0.000 0.000 0.000 0.000 0.000 0.000

0.040 0.080 0.142 0.225 0.326 0.436 0.546 0.650 0.739 0.813 0.870

0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

0.001 0.004 0.009 0.018 0.034 0.060 0.096 0.144 0.204 0.273 0.349

Note. The table gives Rosenbaum bounds for p-values of the effect of naturalization on the likelihood of contacting government officials. “L” columns give minimum p-values and “U” columns give maximum p-values. Γ values indicate the size of the omitted factor.

impact of naturalization on the likelihood of contacting non-Latino government officials may become insignificant when Γ = 1.35. The degree of sensitivity to hidden bias suggested by these values of Γ is common in the social sciences, and cautions us that it is almost always the case (within an observational framework) that estimated causal effects may be sensitive to the presence of unobservables that affect both the treatment and outcome of interest.

Conclusion In recent decades, the accelerated growth in the number of immigrants coming from Latin American countries and of the overall size of the Latino population has motivated greater interest in Latino political behavior. Scholars have studied several aspects of Latino political participation. They have compared the behavior of Latinos in the United States with that of other ethnic groups, and found that Latinos tend to participate less than Anglos— although differences tend to disappear after controlling for attributes such as socioeconomic status, nativity, and English usage (Leighley & Vedlitz, 1999;

Levin

557

Tam Cho, 1999; Uhlaner, 1989; Verba et al., 1995); and there are considerable differences in behavior across national origin groups (Diaz, 1996; Hero & Campbell, 1996). Other scholars have studied the effect of sources of citizenship—naturalization or nativity—on political participation by comparing the behavior of foreign-born naturalized Latinos with the behavior of native-born Latinos. While some find that naturalized Latinos participate less than the native born (DeSipio, 1996), others argue that differences depend on the political context (Pantoja et al., 2001). Pantoja et al. (2001) find that individuals who naturalize for political reasons during periods of heated and polarized antiimmigrant rhetoric—such as when proposition 187 passed in California in 1994—are more likely to turn out to vote relative to native-born Latinos. While comparing the behavior of naturalized and native-born Latinos can be used to learn about the effect of sources of citizenship; learning about the effect of acquisition of citizenship requires comparing the behavior of nonnaturalized and naturalized immigrants, as was done in this article. Understanding the effect of acquisition of citizenship through naturalization on political participation is important for at least two reasons. First, in order to improve the incorporation prospects of undocumented Latino immigrants, temporary workers, and recently arrived legal immigrants, scholars, and advocates have made a variety of proposals, including: granting legal residents immediate voting rights (García Bedolla, 2006, p. 63); providing a clear pathway to legal status and naturalization for undocumented and temporary immigrants (Ewing, 2010; Galston et al., 2009); and allowing immigrants to vote in municipal elections, so that they can affect those collective decisions that affect them most directly, such as the selection of school board or city council members (García Bedolla, 2006, pp. 61-62; Shapiro, 2003, pp. 52-53). Studying the effects of naturalization can help evaluate these proposals. Second, as discussed in the Introduction, studying the determinants of political participation among immigrants can contribute to acquiring a more complete understanding of the determinants of civic engagement more in general, as it requires researchers to go beyond the consideration of standard factors such as socioeconomic status, and consider other individual characteristics such as differences in pre- and post-immigration experiences. Still, with a few exceptions, little has been done to date to study the effect of acquisition of citizenship on civic engagement. What is the effect of naturalization on political inclusion of Latino immigrants? Even if naturalized Latinos are less likely to turn out to vote relative to their native-born counterparts (as found by DeSipio, 1996), it is clear that naturalization has electoral benefits for the Latino immigrant community.

558

American Politics Research 41(4)

Thanks to having gone through the naturalization process, more than half of today’s Latino immigrants have acquired the right to vote in local and federal elections in the United States, and a smaller but still considerable proportion choose to exercise their right to vote—76.5% of naturalized immigrants interviewed in the 2006 LNS reported being registered to vote, and 62.3% reported having voted in the 2004 presidential election. Thus, the naturalization process already contributes to drawing many immigrants into a fundamental part of the democratic process: elections. Increasing the proportion of Latinos who are eligible to vote, by easing naturalization procedures or extending the suffrage to noncitizens, might be beneficial for Latino immigrant as this can help make elected officials more responsive to the needs of the Latino immigrant constituency.20 However, the results of this article suggest that the effect of naturalization on civic engagement among Latino immigrants might be limited to the electoral arena. I found that after acquiring citizenship, Latino immigrants are not more likely to get involved in community-building activities such as involvement in social or political groups; trying to solve problems through existing organizations and informal gatherings; or parental volunteering at schools. The mobilizing effect of naturalization is only evidenced in greater willingness to contact government officials. Thus, granting citizenship status to nonnaturalized immigrants might not be enough for bringing about broader civic and political incorporation of the Latino immigrants—unless this measure is supplemented with other policies directed at promoting immigrant civic engagement. Why does the acquisition of citizenship through naturalization not lead to greater overall levels of community and political engagement? Characteristics of the social context faced by Latino immigrants might represent an obstacle to their political incorporation even after having successfully gone through the naturalization process. According to Portes and Zhou (1993, p. 83), “there are three features of the social contexts encountered by today’s newcomers that create vulnerability to downward assimilation. The first is color, the second is location, and the third is the absence of mobility ladders.” Experiences of racial discrimination, lack of access to politically relevant resources (Verba et al., 1995), and absence of “civic infrastructure” in settlement areas (Pearson-Merkowitz, 2012; Rocha & Espino, 2010), are all factors that might be conductive to reduced levels of trust in government and society (Abrajano & Avarez, 2010a; Michelson, 2001, 2003, 2007), and lower civic engagement of Latino immigrants regardless of citizenship status. Easing naturalization requirements or granting voting rights to noncitizens, might not be

Levin

559

enough to solve the inequalities created by the adverse social context faced by many Latino immigrants. To better understand the interaction between naturalization and social context faced by immigrant communities, future research should explore the relationship between naturalization and political participation among immigrants of different ethnic backgrounds that undergo different immigration experiences, such as Asians, and compare results to those found for immigrants of Latin American origin. It is important to take into account that the data used in this article was collected in a context of massive immigrant right protests on a nationwide scale. Millions of Latinos, immigrants and native born, took the streets to demonstrate against House bill H.R. 4437, a bill that if signed into law would have criminalized undocumented immigrant status. In the past, it has been found that antiimmigrant legislation and rhetoric stimulate political participation by naturalized Latino immigrants, to the extent that they turned out to vote more than native-born Latinos (Pantoja et al., 2001). Pantoja et al. (2008) argue that the 2006 cycle of protests could also affect the long-term behavior of U.S. Latinos, by increasing their political engagement, and motivating them to naturalize and participate in political activities. It is likely that political context does not only affect differences in behavior between naturalized and native-born Latinos, but also between nonnaturalized and naturalized Latino immigrants—for instance, it is possible that the politicized environment did more to mobilize nonnaturalized Latinos than those who had acquired citizenship before 2006, concealing differences in participation levels between naturalized and nonnaturalized immigrants. To acquire a better understanding of the effect of naturalization on the behavior of Latino immigrants, future research should take into account the influence of political context by comparing the behavior of nonnaturalized and naturalized Latino immigrants over different time periods. Acknowledgment I thank Michael Alvarez, Roderick Kiewiet, Melissa Michelson, Betsy Sinclair, Alex Street, conference participants, and anonymous reviewers for their helpful comments.

Author's Note Previous versions of this article were presented at the 2010 Annual Conference of the Midwest Political Science Association, and the 2012 Max Weber Fellows June Conference at the European University Institute.

560

American Politics Research 41(4)

Declaration of Conflicting Interests The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author received no financial support for the research, authorship, and/or publication of this article.

Notes   1. According to the 2008 American Community Survey, approximately 15% of the people living in the United States in 2008 were of Latino origin. Between 2000 and 2008, the increase in the foreign-born Latino population represented 35% of the total U.S. population growth (Dockterman & Velasco, 2010).   2. Locations that allow noncitizen voting include municipalities in Chicago, Maryland, Massachusetts, and New York City (García Bedolla, 2006, p. 62; Hayduk, 2004).   3. There are several reasons why citizens—either native born or naturalized— may not vote. They may choose not to vote due to lack of interest in the election or indifference toward the candidates; they may be unable to vote due to being busy, out of town, or other life circumstances; or they may be kept from voting due to election administration issues. Recently, there has been much debate about election administration requirements introduced by some states that could make voting more difficult, such as the requirement of showing photo identification at the polls prior to voting. Scholars have found that photo ID requirements have been unequally applied (Atkeson, Bryant, Hall, Saunders, & Alvarez, 2010), and others have argued that these requirements could be specially damaging to Latino immigrants and other ethnic minorities (Barreto, Nuño, & Sanchez, 2009b; Sobel & Smith, 2009)—although it is worth noting that current evidence is inconclusive regarding the extent to which photo ID requirements lead to lower turnout among minority voters (Alvarez, Bailey, & Katz, 2007; Mycoff, Wagner, & Wilson, 2009).  4. Not all applicants pass the citizenship test (North, 1987; Yang, 1994). According to a recent report, the overall pass rate of the examination among first-time takers was 87.1% in 2003-2004, and increased to 95.8% in 2010 after a new test was introduced in 2008 (see ICF International, 2011). Alvarez (1987) analyzed the experiences of 38 recently naturalized citizens and found that “the examination process is a mystery to most individuals who choose to become citizens. Unsureness about the exam itself, stories from

Levin

561

acquaintances about poor experiences, myths about rejecting peoples from certain countries, trouble with English language and other factors make the exam a black box.” For a discussion of the new citizenship test from a comparative perspective, see Etzioni (2007).   5. Alvarez (1987) found that “individuals who have recently experienced naturalization are knowledgeable about the Constitution and the system of government here” and that “much of this knowledge is internalized and expressed as part of a new identity.”   6. Restrictive immigration laws can increase immigrants’ incentives to seek naturalization. Examples of such laws include the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), which restricted immigrants’ access to welfare benefits; the 1996 Illegal Immigrant Reform and Immigration Responsibility Act (IIRIRA), which increased deportation risks for immigrants committing felonies and made sponsorship of relatives more difficult for legal immigrants; and antiimmigrant ballot measures such as California’s Proposition 187, passed in 1994 and later ruled unconstitutional by a federal court (Balistreri & Van Hook, 2004; Pantoja et al., 2001; Sierra et al., 2000).   7. Although parental volunteering at schools is a community activity that is not necessarily done or political purposes, it can have implications for political outcomes and affect the development of political attitudes among immigrants. According to DeSipio (1996, p. 202), “the literature on the political adaptation of turn-of-thecentury immigrants portrays [parental involvement in] schools as many immigrants first introduction to U.S. politics.”   8. I consider the behavior of all foreign-born respondents, except those with a U.S. born father or mother. The incidence of missing values was relatively low for most variables considered in the analysis. To prevent dropping observations due to high nonresponse in a few variables such as household income, I imputed missing values using a multivariate procedure implemented through R’s package mice (van Buuren & Groothuis-Oudshoorn, 2011).   9. Specifically, 34% of those coming from Central America or Dominican Republic, and 44% of those coming from South America, have acquired citizenship, compared to 26% among Mexicans and 72% among Cubans. 10. In addition, compared to other studies such as Leal (2002), I use data from a more recent survey and control for a broader set of individual attributes that may affect both the likelihood of becoming an American citizen and the likelihood of participating in politics. Also, I conduct a sensitivity analysis to evaluate the robustness of findings to unobservable differences in individual attributes such as legal status and political interest.

562

American Politics Research 41(4)

11. Specifically, for every individual I computed a predicted propensity score conditional on a vector of prenaturalization attributes, xi, and a vector of logistic 1 regression coefficients, β, such that P(ci = 1| xi ) = 1 + e-xi β 12. The “caliper” specifies the maximum acceptable distance—in terms of propensity score—for matching two units. The size of the caliper was set to 10% of the standard deviation of the estimated propensity score. This caliper size is small compared to the one used in most studies (usually around 25%) and implies that I only matched naturalized and nonnaturalized immigrants that had very similar exante likelihood of becoming American citizens. 13. The matching procedure was implemented using R package MatchIt (Ho et al., 2007b); repeating it until considerable balance improvement was achieved in all covariates. 14. In estimating causal effects it is important not to control for variables that affect outcomes but are themselves affected by treatment assignment, because doing so may lead to biased estimates of treatment effects. Thus, I did not control for individual attitudes that may have been affected by citizenship status, such as interest in politics, trust in government, or group consciousness, which can be characterized as “posttreatment” variables. 15. To improve balance in all moment of the distribution of individual attributes, I also matched on polynomial terms of ordered covariates. Also, to help match individuals who migrated at a similar age, the propensity score model included the interaction between age and length of residence in the United States. 16. To further assess balance improvement, I constructed pre- and postmatching Q−Q plots for ordered covariates included in the analysis (see Online Appendix A1). These plots show that balance was achieved not only on covariate means, but also on other moments of the distribution of covariates included in the matching procedure. 17. To evaluate the extent to which the results of the matching procedure were sensitive to the grouping of respondents into regions of origin, I replicated the analysis after making a series of changes to the region of origin variable. Specifically, I kept the following national origins separate: Cubans, Dominicans, and Salvadorans; and grouped the remaining respondents into the following smaller regions: Other Central Americans (Costa Rica, Guatemala, Honduras, Nicaragua, Panama), North of South America (Bolivia, Colombia, Ecuador, Peru, and Venezuela), and South of South America (Argentina, Chile, Paraguay, and Uruguay). The classification of Mexican respondents remained the same (North, West, East and Center-North, D.F. and Mexico State, and South and Center South). These changes resulted in a new region-of-origin indicator with

Levin

563

11 categories, compared to 8 in the original grouping. The results of this replication—which, overall, reinforce the conclusions of the original analysis—are available in Online Appendix A2. 18. The sensitivity analysis was implemented using R package rbounds (Keele, 2008). 19. I omitted other activities where the effect of naturalization was statistically indistinguishable from 0. 20. Vernby (2013) shows that the introduction of noncitizen suffrage in Swedish municipalities lead to greater public spending on social services in geographic regions with relatively higher concentration of noncitizen immigrants; and something similar might happen in the United States following policies such as easing naturalization requirements or allowing for noncitizen voting in local elections.

References Abrajano, M. A., & Alvarez, R. M. (2010a). Assessing the causes and effects of political trust among U.S. Latinos. American Politics Research, 38, 110-141. Abrajano, M. A., & Alvarez, R. M. (2010b). New faces, new voices: The Hispanic electorate in America. Princeton, NJ: Princeton University Press. Alvarez, R. M., Bailey, D., & Katz, J. N. (2007). The effect of voter identification laws on turnout. Caltech/MIT Voting Technology Project, Working Paper No. 57. Alvarez, R. M., & Butterfield, T. L. (1997). The resurgence of nativism in California? The case of Proposition 187 and illegal immigration. Social Science Quarterly, 81, 167-179. Alvarez, R. M., & García Bedolla, L. (2003). The foundations of Latino voter partisanship: Evidence from the 2000 election. Journal of Politics, 65, 31-49. Alvarez, R. R. (1987). A profile of the citizenship process among Latinos in the United States. International Migration Review, 21, 327-351. Atkeson, L. A., Bryant, L. A., Hall, T. E., Saunders, K. L., & Alvarez, R. M. (2010) A new barrier to participation: Heterogeneous application of voter identification policies. Electoral Studies, 29, 66-73. Bada, X., Fox, J., Donnelly, R., & Selee, A. (2010). Context matters: Latino immigrant civic engagement in nine U.S. Cities, Reports on Latino immigrant civic engagement (National report). Washington, DC: Woodrow Wilson International Center for Scholars. Balistreri, K. S., & Van Hook, J. (2004). The more things change the more they stay the same: Mexican naturalization before and after welfare reform. International Migration Review, 38, 113-130. Barreto, M. A., Manzano, S., Ramírez, R., & Rim, K. (2009a). Mobilization, participation, and solidaridad: Latino participation in the 2006 immigration protest rallies. Urban Affairs Review, 44, 736-764.

564

American Politics Research 41(4)

Barreto, M. A., Nuño, S. A., & Sanchez, G. R. (2009b). The disproportionate impact of voter-ID requirements on the electorate—New evidence from Indiana. PS: Political Science & Politics, 42(1), 111-116. Bean, F. D., Brown, S. K., & Rumbaut, R. G. (2006). Mexican immigrant political and economic incorporation. PS: Political Science & Politics, 4, 309-313. Benjamin-Alvarado, J., DeSipio, L., & Montoya, C. (2008). Latino mobilization in new immigrant destinations: The Anti–H.R. 4437 Protest in Nebraska’s Cities. Urban Affairs Review, 44, 718-735. Bernard, W. S. (1936). Cultural determinants of naturalization. American Sociological Review, 1, 943-953. Bloemraad, I. (2006a). Becoming a citizen in the United States and Canada: Structured mobilization and immigrant political incorporation. Social Forces, 85, 667-695. Bloemraad, I. (2006b). Citizenship lessons from the past: The contours of immigrant naturalization in the early 20th century. Social Science Quarterly, 87, 927-953. Brehm, J., & Rahn, W. (1997). Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41, 999-1023. Citrin, J., Green, D. P., Muste, C., & Wong, C. (1997). Public opinion toward immigration reform: The role of economic motivations. Journal of Politics, 59, 858-881. Citrin, J., Lerman, A., Murakami, M., & Pearson, K. (2007). Testing Huntington: Is Latino immigration a threat to American identity? Perspectives on Politics, 5, 31-48. Cochran, W. G., & Rubin, D. B. (1973). Controlling bias in observational studies: A review. Sankhya A, 35, 417-446. Cortina, J., Gelman, A., & Blanco, N. L. (2008). One vote, many Mexicos: Income and vote choice in the 1994, 2000, and 2006 presidential elections. Unpublished manuscript. Dahl, R. A. (2000). On democracy. New Haven, CT: Yale University Press. de la Garza, R. O., Falcon, A., & Garcia, F. C. (1996). Will the real Americans please stand up: Anglo and Mexican-American support of core American political values. American Journal of Political Science, 40, 335-351. DeSipio, L. (1987). Social science literature and the naturalization process. International Migration Review, 21, 390-405. DeSipio, L. (1996). Making citizens or good Citizens? Naturalization as a predictor of organizational and electoral behavior among Latino immigrants. Hispanic Journal of Behavioral Sciences, 18, 194-213. DeSipio, L. (2006, March 2-3). From naturalized citizen to voter: Context of naturalization and electoral participation in Latino communities. Paper presented at Immigration Policy After 9/11: U.S. and European Perspectives. Center for European Studies, University of Texas at Austin. DeSipio, L. (2011). Immigrant incorporation in an era of weak civic institutions: Immigrant civic and political participation in the United States. American Behavioral Scientist, 55, 1189-1213.

Levin

565

Diaz, W. A. (1996). Latino participation in America: Associational and political roles. Hispanic Journal of Behavioral Sciences, 18, 154-174. Dockterman, D., & Velasco, G. (2010). Statistical portrait of the foreign-born population in the United States, 2008. Pew Latino Center. Retrieved from http:// www.pewhispanic.org/2010/01/21/statistical-portrait-of-the-foreign-born-population-in-the-united-states-2008/ Etzioni, A. (2007). Citizenship tests: A comparative, communitarian perspective. The Political Quarterly, 78, 353-363. Evans, M. D. R. (1988). Choosing to be a citizen: The time-path of citizenship in Australia. International Migration Review, 22, 243-264. Ewing, W. A. (2010). The many facets of effective immigration reform. Society, 47, 110-117. Félix, A., González, C., & Ramírez, R. (2008). Political protest, ethnic media, and Latino naturalization. American Behavioral Scientist, 52, 618-634. Fraga, L. R., Garcia, J. A., Hero, R., Jones-Correa, M., Martinez-Ebers, V., & Segura, G. M. (2006a). Latino National Survey (LNS) [Computer file]. ICPSR20862-v3. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2010-03-26. doi:10.3886/ICPSR20862. Fraga, L. R., Garcia, J. A., Hero, R., Jones-Correa, M., Martinez-Ebers, V., & Segura, G. M. (2006b). Su casa es nuestra casa: Latino politics research and the development of American political science. American Political Science Review, 100, 515-521. Fraga, L. R., Garcia, J. A., Hero, R., Jones-Correa, M., Martinez-Ebers, V., & Segura, G. M. (2012). Latinos in the new millennium an almanac of opinion, behavior, and policy preferences. New York, NY: Cambridge University Press. Fraga, L. R., & Segura, G. M. (2006). Culture clash? Contesting notions of American identity and the effects of Latin American immigration. Perspectives on Politics, 4, 279-287. Galston, W., Pickus, N., & Skerry, P. (2009). Breaking the immigration stalemate: From deep disagreements to constructive proposals. Brookings-Duke Immigration Policy Roundtable. Brookings Institution and Kenan Institute for Ethics. Retrieved from http://www.brookings.edu/research/reports/2009/10/06-immigration-roundtable García Bedolla, L. (2006). Rethinking citizenship: Noncitizen voting and immigrant political engagement in the United States. In T. Lee, S. K. Ramakrishnan, & R. Ramírez (Eds.), The Political incorporation of immigrants in the United States: Transforming politics, transforming America (pp. 51-70). Charlottesville: University of Virginia Press. García Bedolla, L. (2009). Latino politics. Malden, MA: Polity Press. Grebler, L. (1966). The naturalization of Mexican immigrants in the United States. International Migration Review, 1, 17-32.

566

American Politics Research 41(4)

Hainmueller, J., & Hiscox, M. J. (2010). Attitudes toward highly skilled and lowskilled immigration: Evidence from a survey experiment. American Political Science Review, 104, 61-84. Hayduk, R. (2004). Noncitizen voting rights: Extending the franchise in the United States. National Civic Review, 4, 57-62. Hero, R. E., & Campbell, A. G. (1996). Understanding Latino political participation: Exploring the evidence from the Latino National Political Survey. Hispanic Journal of Behavioral Sciences,18, 129-141. Hiskey, J. T., & Bowler, S. (2005). Local context and democratization in Mexico. American Journal of Political Science, 49, 57-71. Ho, D. E., Imai, K., King, G., & Stuart, E. A. (2007a). Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political Analysis, 15, 199-236. Ho, D. E., Imai, K., King, G., & Stuart, E. A. (2007b). MatchIt: Nonparametric preprocessing for parametric causal inference. Journal of Statistical Software, 42. Retrieved from http://gking.harvard.edu/node/4355/rbuild_documentation Huntington, S. P. (2004). Who are we? The challenge to America’s national identity. New York, NY: Simon and Schuster. ICF International. (2011). U.S. Citizenship and Immigration Services’ records study on pass/fail rates for naturalization applicants. Retrieved from http://www.uscis. gov/USCIS/files/Records_Study_for_the_Naturalization_Test.pdf Jasso, G., & Rosenzweig, M. R. (1986). Family reunification and the immigration multiplier: U.S. immigration law, origin-country conditions, and the reproduction of immigrants. Demography, 23, 291-311. Jones-Correa, M. (2001). Under two flags: Dual nationality in Latin America and its consequences for naturalization in the United States. International Migration Review, 35, 997-1029. Keele, L. J. (2008). rbounds: An R Package for sensitivity analysis with matched data. Columbus: Ohio State University. Retrieved from http://www.polisci.ohio-state. edu/faculty/lkeele/rbounds.html Klesner, J. L. (2007). The 2006 Mexican elections: Manifestation of a divided society? PS: Political Science and Politics, 40, 27-32. Leal, D. L. (2002). Political participation by Latino non-citizens in the United States. British Journal of Political Science, 32, 353-370. Leighley, J. E., & Vedlitz, A. (1999). Race, ethnicity, and political participation: Competing models and contrasting explanations. Journal of Politics, 61, 1092-1114. Michelson, M. R. (2001). Political trust among Chicago Latinos. Journal of Urban Affairs, 23, 323-334. Michelson, M. R. (2003). The corrosive effect of acculturation: How Mexican Americans lose political trust. Social Science Quarterly, 84, 918-933.

Levin

567

Michelson, M. R. (2007). All roads lead to rust: How acculturation erodes Latino Immigrant trust in Government. Aztlán: A Journal of Chicano Studies, 32, 21-46. Mycoff, J. D., Wagner, M. W., & Wilson, D. C. (2009). The empirical effects of voterID laws: Present or absent? PS: Political Science & Politics, 42, 121-126. North, D. S. (1987). The long grey welcome: A study of the American naturalization program. International Migration Review, 21, 311-326. Pachon, H. P. (1987). An overview of citizenship in the Hispanic community. International Migration Review, 21, 299-310. Pantoja, A. D., & Gershon, S. A. (2006). Political orientations and naturalization among Latino and Latina immigrants. Social Science Quarterly, 87, 1171-1187. Pantoja, A. D., Ramirez, R., & Segura, G. M. (2001). Citizens by choice, voters by necessity: Patterns in political mobilization by naturalized Latinos. Political Research Quarterly, 54, 729-750. Pantoja, A. D., Ramirez, R., & Segura, G. M. (2008). Commentary on “Citizens by choice, voters by necessity: Patterns in political mobilization by naturalized Latinos.” Political Research Quarterly, 61, 50-52. Pearson-Merkowitz, S. (2012). The limits of the homogeneity model: Segregation and civic engagement in Latino communities. American Politics Research, 40, 701-736. Pew Research Center for the People & the Press. (2006). No consensus on immigration problem or proposed fixes: America’s immigration quandary. Retrieved from http://www.people-press.org/2006/03/30/americas-immigration-quandary Portes, A., & Zhou, M. (1993). The new second generation: Segmented assimilation and its variants. Annals of the American Academy of Political and Social Science, 530, 74-96. Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community. New York, NY: Touchstone Books/Simon & Schuster. Rocha, R. R., & Espino, R. (2010). Segregation, immigration, and Latino participation in ethnic politics. American Politics Research, 38, 614-635. Rosenbaum, P. R. (1995). Observational studies. New York, NY: Springer-Verlag. Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70, 41-55. Scheve, K. F., & Slaughter, M. J. (2001). Labor market competition and individual preferences over immigration policy. Review of Economics and Statistics, 83, 133-145. Schildkraut, D. J. (2007). Defining American identity in the twenty-first century: How much “there” is there? Journal of Politics, 69, 597-615. Sekhon, J. S. (2009). Opiates for the matches: Matching methods for causal inference. Annual Review of Political Science, 12, 487-508.

568

American Politics Research 41(4)

Shapiro, I. (2003). The state of democratic theory. Princeton, NJ: Princeton University Press. Sierra, C. M., Carrillo, T., DeSipio, L., & Jones-Correa, M. (2000). Latino immigration and citizenship. PS: Political Science and Politics, 33, 535-540. Silber Mohamed, H. (2010). Eventual identity: Latinos, immigration politics and the Spring 2006 protests. Paper presented at the Western Political Science Association 2010 Annual Meeting Paper. Retrieved from http://papers.ssrn.com/sol3/ papers.cfm?abstract_id=1581029. Sniderman, P. M., Hagendoorn, L., & Prior, M. (2004). Predisposing factors and situational triggers: Exclusionary reactions to immigrant minorities. American Political Science Review, 98, 35-49. Sobel, R., & Smith, R. E. (2009). Voter-ID laws discourage participation, particularly among minorities, and trigger a constitutional remedy in lost representation. PS: Political Science & Politics, 42, 107-110. Staton, J. K., Jackson, R. A., & Canache, D. (2007). Dual nationality among Latinos: What are the implications for political connectedness? Journal of Politics, 69, 470-482. Tam Cho, W. K. (1999). Naturalization, socialization, participation: Immigrants and (non-)voting. Journal of Politics, 61, 1140-1155. Uhlaner, C. J., Cain, B. E., & Kiewiet, D. R. (1989). Political participation of ethnic minorities in the 1980s. Political Behavior, 11, 195-231. van Buuren, S., & Groothuis-Oudshoorn, K. (2011). MICE: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3), 1-67. Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and equality: Civic voluntarism in American politics. Cambridge, MA: Harvard University Press. Vernby, K. (2013). Inclusion and public policy: Evidence from Sweden’s Introduc­tion of Noncitizen Suffrage. American Journal of Political Science, 57, 15-29. Waslin, M. (2004). Immigration reform: Comprehensive solutions for complex problems. Paper presented at the 2004 Meeting of the Latin American Studies Association, Las Vegas, NV. Retrieved from http://lasa.international.pitt.edu/members/ congress-papers/lasa2004/files/WaslinMichele_xCD.pdf Wolfinger, R. E., & Rosenstone, S. J. (1980). Who votes? New Haven, CT: Yale University Press. Yang, P. Q. (1994). Explaining immigrant naturalization. International Migration Review, 28, 449-477.

Author Biography Ines Levin is an Assistant Professor in the Department of Political Science at the University of Georgia.