From Immigrant to Naturalized Citizen: Political Incorporation in the United States (The New Americans: Recent Immigration and American Society)

The New Americans Recent Immigration and American Society Edited by Steven J. Gold and Rubén G. Rumbaut A Series from...

Author: Catherine Simpson Bueker

12 downloads 1358 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

The New Americans Recent Immigration and American Society

Edited by Steven J. Gold and Rubén G. Rumbaut

A Series from LFB Scholarly

This page intentionally left blank

From Immigrant to Naturalized Citizen

Catherine Simpson Bueker

LFB Scholarly Publishing LLC New York 2006

Copyright © 2006 by LFB Scholarly Publishing LLC All rights reserved. Library of Congress Cataloging-in-Publication Data Bueker, Catherine Simpson, 1974From immigrant to naturalized citizen / Catherine Simpson Bueker. p. cm. -- (The new Americans) Includes bibliographical references and index. ISBN 1-59332-137-6 (alk. paper) 1. Immigrants--United States--Political activity. 2. Voting--United States--History. 3. Political participation--United States. I. Title. II. Series: New Americans (LFB Scholarly Publishing LLC) JV6477.B84 2006 324.9730086'912--dc22 2006009808

ISBN 1-59332-137-6 Printed on acid-free 250-year-life paper. Manufactured in the United States of America.

CONTENTS ACKNOWLEDGEMENTS……………………………………….......xi CHAPTER 1: Why Immigrant Political Incorporation? An Introduction.....................1 CHAPTER 2: Assessing Immigrant Political Activity................................................19 CHAPTER 3: The Effect of Individual Characteristics on Political Incorporation.....53 CHAPTER 4: The Role of Family...............................................................................93 CHAPTER 5: Does Community Matter?...................................................................121 CHAPTER 6: Gone, but Not Forgotten: The Enduring Influence of Home Society.149 CHAPTER 7: A Look Back, A Look Forward: Contributions & Future Research..181 SUPPLEMENTAL TABLES...……………………………………..193 REFERENCES……………………………………………………...213 INDEX…………………………………............................................225

v


LIST OF TABLES TABLE 2.1: Operationalization of Individual Characteristics..............23 TABLE 2.2: Operationalization of Family Characteristics…...............25 TABLE 2.3: Definition of Sub-Samples……………………………...32 TABLE 2.4: Operationalization of Community Characteristics……...35 TABLE 2.5: Operationalization of Country of Origin Characteristics………………………………………….38 TABLE 2.6: Descriptive Statistics of Independent Variables in SubSample A…………………………………………..…...44 TABLE 2.7: Descriptive Statistics of Independent Variables in SubSample C…………………………………………..…...45 . TABLE 2.8: Summary Statistics for Those Eligible to Naturalize, by Immigrant Group……………………….……………....49 TABLE 2.9: Summary Statistics for Those Eligible to Vote, by Immigrant Group…………………………………….....50 TABLE 3.1: The Odds of Citizenship Acquisition among Immigrants……………………………………………...69 TABLE 3.2: The Odds of Voter Turnout among Immigrants………..70 TABLE 3.3: The Odds of Citizenship Acquisition by Country of Origin…………………………...………………………79 TABLE 3.4: The Odds of Voter Turnout by Country of Origin…...…80 TABLE 3.5: The Odds of Voter Turnout, Immigrants versus the Native-Born………………………………………...…..86 TABLE 4.1: The Odds of Citizenship Acquisition by Marital Status………………………………………104 vii

viii

List of Tables

TABLE 4.2: The Odds of Voter Turnout by Marital Status.….…….109 TABLE 4.3: The Odds of Voter Turnout among Immigrants and the Native-Born by Marital Status……………...…………110 TABLE 4.4: The Odds of Citizenship Acquisition among Mexican Immigrants by Marital Status……………..…………..112 TABLE 4.5: The Odds of Voter Turnout among Mexican Immigrants by Marital Status………...……...……………………..114 TABLE 5.1: The Odds of Citizenship Acquisition for Immigrants in Communities with Selected Characteristics………..…138 TABLE 5.2: The Odds of Voter Turnout for Immigrants in Communities with Selected Characteristics…………..139 TABLE 5.3: The Odds of Citizenship Acquisition among Mexican Immigrants in Communities with Selected Characteristics………………………………………...143 TABLE 5.4: The Odds of Voter Turnout among Mexican Immigrants in Communities with Selected Characteristics…..…....144 TABLE 6.1: The Odds of Citizenship Acquisition among Immigrants with Selected Country of Origin Characteristics……...167 TABLE 6.2: The Odds of Voter Turnout among Immigrants with Selected Country of Origin Characteristics...…………173

LIST OF FIGURES FIGURE 1.1: Potential Paths of Immigrant Political Incorporation…...5 FIGURE 2.1: Estimated Direction of Coefficients…………………...42 FIGURE 2.2: Levels of Political Incorporation among Immigrants with Varying Levels of Education and Income…………..….46 FIGURE 3.1: The Odds of Naturalizing and Voting by Selected Socioeconomic Characteristics…….………………......72 FIGURE 3.2: The Odds of Naturalizing by Country of Origin……....73 FIGURE 3.3: The Odds of Voting by Country of Origin…………….75 FIGURE 3.4: The Odds of Women versus Men Naturalizing and Voting by Country of Origin…..…………………...….83 FIGURE 3.5: The Odds of Immigrants versus Native-Born Voting…87 FIGURE 4.1: The Odds of Naturalizing among Single and Single and Separated versus Married, Co-Residing Immigrants, by Gender……………………………………………...…106 FIGURE 4.2: The Odds of Women versus Men Naturalizing, by Marital Status….............................................................108 FIGURE 5.1: The Odds of Immigrant Naturalization in a Metro Area with Increasing Rates of Foreign-Born versus Non-Metro Residence...................................................................... 140 FIGURE 5.2: The Odds of Voter Turnout in a Metro Area with Increasing Rates of Foreign-Born versus Non-Metro Residence……………………………………………...141 FIGURE 6.1: The Effect of Rising GDP in the Country of Origin on the Odds of Naturalizing among Immigrants of Differing Educational and Economic Levels…………...……….170 FIGURE 6.2: The Odds of Immigrant Voter Turnout under Three Different Country of Origin Scenarios……………......175 ix


ACKNOWLEDGEMENTS There are numerous people and institutions I would like to thank for making this book a reality. Brown University provided me with significant financial support through a University Fellowship, allowing me to complete my research in a timely manner. Calvin Goldscheider and Michael White, both in the Department of Sociology at Brown, and Alan Zuckerman, in the Department of Political Science, all provided me with invaluable feedback and support throughout this project, constantly pushing me to think more deeply about the findings. Greg Elliott, of the Department of Sociology, lent unending hours assisting me in my statistical analysis. I would like to thank Leo Balk of LFB Scholarly Publishing, and Steven Gold and Ruben Rumbaut, editors of the series The New Americans: Recent Immigration and American Society, for initially contacting me and publishing my work. Finally, my proofreaders, David Simpson and John Bueker, are owed tremendous gratitude. As my father and husband, respectively, they walked the dangerous line of being both family and critics. I really do appreciate all they have done.

xi


CHAPTER 1

Why Immigrant Political Incorporation? An Introduction The United States is frequently viewed as a country of immigrants, a great experiment that has successfully incorporated each new wave of immigration into the various folds of American culture. This is a country founded by immigrants from Europe, coming voluntarily to seek religious freedom and economic opportunity, and largely built by involuntary migrants from Africa. Individuals have come to the United States from virtually every country and for every reason. Some have even viewed the historical and ongoing flows of migrants to American shores as the “defining attribute of American society” (Hirschman, Kasinitz & DeWind, 1999, p.1). Another way Americans commonly view their country is as a great democracy, a place where each individual has the freedom, right, and even obligation to voice his or her political opinion. At least theoretically, the rights and privileges of one citizen are no greater or lesser than those of any other. No citizen’s vote is any more or less powerful than that of another’s. The overarching question, therefore, is how do these two fundamental principles of American society interact with one another? This book is intended to improve our understanding of immigrant political incorporation, the process of integrating a previously disenfranchised individual into the polis. Do immigrants truly become politically incorporated into American society? Do they seek out the status of citizenship or remain outside of the polis? If they choose to naturalize, do they then voice their opinions, beliefs, and desires at the voting booth, or do they remain silent citizens? This book, based principally on four years of Current Population Survey data, is intended to provide greater insight into the process of political integration of immigrants, a field that has received less attention than other areas of immigrant incorporation. By examining immigrants from Mexico, Cuba, Canada, Britain, Italy, the former Soviet Union, China, Southeast Asia, India, and the Philippines, I explore how those with diverse individual characteristics, histories in 1

2


the United States, reasons for immigrating, and political and cultural situations in the country of origin either become politically incorporated in the United States or fail to do so. Such diversity allows us to see whether immigrant groups are affected differently by the same factors, whether the degree of political incorporation varies across groups, or whether, given enough time, all immigrant groups become similarly incorporated. THE IMPORTANCE OF POLITICAL INCORPORATION Understanding whether and how immigrants become incorporated into American society has received substantial attention from immigration scholars. But these studies have largely focused on three areas of incorporation: economic, educational, and social, with less attention given to political integration. The inattention of scholars to immigrant political incorporation gives no indication of its theoretical importance or its relevance to American society and to public policy. In fact, given the current immigration climate, understanding how immigrants become part of the polis is increasingly important. From 1991 through 2000, more than 9 million legally documented immigrants entered the United States (Office of Immigration Statistics, 2005), surpassing the number of immigrants who entered from 1901 through 1910, the period viewed as the height of U.S. immigration. Although the September 11th, 2001 terrorist attacks in the United States led to significant restrictions in entry to the United States, particularly for individuals from Middle Eastern countries of origin, considerable numbers of immigrants have continued to enter. Nearly 3 million documented immigrants entered the United States from 2002 through 2004 (Office of Immigration Statistics, 2005), and over 500,000 naturalized during 2004 (Rytina & Saeger, 2005), the most recent year for which statistics are available. At the individual-level, naturalization provides immigrants with a host of civic, economic, and social benefits (Johnson, Reyes, Mameesh, & Barbour, 1999). Politically, citizenship offers immigrants the right to vote, the right to hold most elected offices, and the rights and protections offered to the native-born (Jones-Correa, 2001). Citizenship also unlocks economic benefits (Lister, 1998; Pachon, 1987). In recent years, citizenship status has come to have greater economic implications, with many social welfare benefits limited to

Why Immigrant Political Incorporation?

3

American citizens. Under the 1996 Welfare Reform Act, economic aid in the form of cash transfers and food stamps has been tied to citizenship status. In the private sector, American citizenship is frequently a requirement to secure credit, and to obtain home and educational loans. Citizenship has implications beyond the individual. With the enactment of the Immigration Act of 1965, the United States altered its focus. Though labor needs continued to be a primary concern, country of origin preferences were jettisoned for family reunification goals (Massey, Arango, Hugo, Kouaouci, Pellegrino, & Taylor, 2005). As a result, skilled workers from all around the world and relatives of American citizens suddenly topped the list to receive visas (Johnson et al., 1999). This has meant that immigrants in the post-1965 period who have chosen to naturalize have been able to help relatives gain access to the United States, thus perpetuating immigration flows from their respective countries of origin. In 2004, 65% of all immigrants who legally entered the United States entered either as an immediate relative of a U.S. citizen, or under the family preference category (Office of Immigration Statistics, 2005), suggesting how individual-level behavior can affect migration flows at the societal level. In the United States, voting continues to be the most formalized, systematic, and frequent type of political activity and is arguably the most fundamental form of political participation in a democracy (Putnam, 2000). Who chooses to vote has implications for which politicians are elected, what policies are enacted, and how resources are allocated. Immigrants who mobilize and vote as a bloc can attain significant political power (Plotke, 1999). Choosing not to vote may have similarly important implications for the nation. Some argue that large groups of disenfranchised residents constitute a destabilizing influence on American democracy, by narrowing the percentage of people making decisions for the society as a whole (Harles, 1993; Plotke, 1999). These unincorporated individuals may turn into the political equivalent of an economic underclass. While the individual-level effects and benefits of voting may be far smaller, less tangible, and less direct than those which accrue from naturalizing, the implications at the societal level may be just as large.

4


Therefore the importance of better understanding immigrant political incorporation should be apparent, given the substantial consequences for the individual, the community, and the larger society. Unfortunately, we continue to know relatively little about the factors which promote the processes of naturalizing and voting. As discussed, citizenship brings an individual many concrete benefits, far beyond just access to the voting booth. Although voting may be viewed by some as a precious right and duty, as the very embodiment of citizenship, and as the most fundamental political act in a democracy (Putnam, 2000), it provides fewer immediate and tangible benefits than does the acquisition of citizenship. Individuals may receive a psychic reward for having participated in the electoral process; and with sufficient group mobilization, voting may translate into the election of a preferred candidate or the implementation of favored policies. However, such outcomes and benefits rely upon the turnout of large numbers of likeminded voters, and mere participation does not ensure them. Thus, although naturalizing is one step in the process of political participation, and is a necessary condition for voting in the United States, as Figure 1 illustrates, it is far from sufficient indication of full political participation. Are immigrants becoming citizens and part of the polis, or are they remaining outsiders? Do those who become citizens vote and do they thereby succeed in making their political voices heard, or do they fail even to go to the voting booths? Given the differing incentives behind the two acts, it is likely that the causal structures of naturalizing and voting are distinct. Exploring voting trends, as well as citizenship trends, allows us to understand the three potential courses of action open to immigrants: not becoming a citizen and thus being precluded from formal political participation; becoming a citizen, but failing to vote; and becoming a citizen and voting. Further, in the wake of the 1965 Immigration Act, patterns of immigration have changed dramatically. The immigrants of today arrive disproportionately from Asia and Latin America; not a single European country is listed among the top ten contributors of immigrants to the United States in recent years (Massey et al., 2005; Office of Immigration Statistics, 2005). Substantial portions of recent


5

FIGURE 1.1: Potential Paths of Immigrant Political Incorporation Immigrant

Citizen

Voter

Non-Citizen

Non-Voter

immigrants arrive from non-democratic regimes, with countries such as China, Vietnam, and Cuba consistently ranking among the top ten countries of origin. A common concern among both the American public and policy makers has been the lack of familiarity of some immigrant groups with the American political system, specifically, and democratic governance, more generally. Do immigrants learn the ways of democratic society and become incorporated, politically, or do they remain ignorant of and uninvolved in American democracy? Some scholars (Black, Niemi, & Powell, 1987; Finifter & Finifter, 1989) argue that immigrants from democratic societies are able to transfer their previous experience to the new democratic society in which they find themselves. If this is indeed the case, though not familiar with the American system of governance, immigrants should be able to transfer their previous experience and knowledge to American electoral procedures. At the same time, Americans are arguably raised with a very specific set of political values, “Examinations of the republic’s political culture typically maintain that Americans are to a remarkable degree united on the tenets of a central political value system…most often defined in terms of a variety of liberalism” (Harles, 1993, p.2). If this is the case, the transfer of previous democratic experience may not work, leaving even larger numbers of immigrants unfamiliar with this country’s political norms

6


and requiring all immigrants, even those from democracies, to be politically re-socialized. Therefore, it is essential to move beyond just examining whether immigrants naturalize and vote to understand which immigrants do so and why. By exploring these different paths of political incorporation, as measured by naturalizing and voting, we can better understand whether certain groups have different propensities to naturalize and/or vote, whether voting is the natural outgrowth of becoming an American citizen, and why there may be differentials in these political behaviors across groups. These explorations contribute to the literature by increasing what we know about political incorporation, a theoretically less developed area. What are the multi-leveled factors that affect the process of political incorporation? Are citizenship and voting similar political processes, or are they different practices motivated by diverse factors? Do the processes of political integration follow patterns similar to those identified in the realms of economic or educational incorporation, or does civic integration operate differently? PREVIOUS STUDIES The majority of research examining immigrant incorporation has focused on economic, educational, and social assimilation. These three areas have received such attention largely because they are viewed as potentially having the greatest consequences for American society. If immigrants cannot support themselves financially, will the burden fall on the native-born? Are immigrants “catching up” educationally, or will they develop into an underclass? Are immigrants learning the English language, or will we become a multilingual society? In short, many studies have focused on the extent to which American society is being altered by immigrants, rather than on how immigrants are being altered by American society. Historically, the economic, educational and social areas have been used to measure the level of assimilation or incorporation of immigrant groups. Milton Gordon (1964) identified seven phases of assimilation, beginning with acculturation or the acquisition of such mainstream characteristics as language, food and dress, and ending in civic assimilation or the acquisition of the political norms and values of the host society. Between the first step of acculturation and the last step of


7

civic assimilation, Gordon argued that immigrant groups would travel through structural integration, amalgamation, identification, attitudinal assimilation, and behavioral assimilation, with each of the steps leading inevitably to the next, until the immigrant group was indistinguishable from the host population. Though Gordon suggested the process might take several generations to complete, he argued that assimilation followed a “straight-line,” with each phase leading to the next, in an ordered fashion for all groups. Given enough time, immigrants, regardless of country of origin, would take the same path toward assimilating with the reference group, White Anglo-Saxon Protestants. Immigrants would acquire the English language, begin to live in more integrated neighborhoods, identify as American, and affiliate with mainstream, rather than ethnic, organizations. Political incorporation, called civic assimilation in Gordon’s work, would translate into embedded democratic values and indistinguishable rates and types of political participation among all immigrant groups, given equal amounts of time in the United States. Gordon wrote prior to the 1965 changes in immigration policy, thus largely confining his analysis to European immigrants. The changes in migration patterns to Asian and Latino-dominated flows in the post-1965 period have led many scholars to compare today’s immigrants with their European predecessors of the nineteenth and early twentieth centuries (Perlmann & Waldinger, 1999). Will the current immigrants and the new second generation follow the same “straight-line” patterns of assimilation as their forerunners did, or will they follow a different path? In recent years, scholars have developed theories of segmented assimilation and have tested Gordon’s straight-line assimilation theory against the new patterns of migration to the United States. This “new” theory allows for diversity in immigrant assimilation and incorporation by immigrant group, order and type of incorporation, and area of settlement (Alba & Nee, 1999; Portes, 1995a; Zhou, 1999). Segmented assimilation theorists have identified three potential paths of incorporation: upward assimilation towards the white middle-class, downward assimilation towards the native-born minority underclass, and a third option whereby immigrant groups retain some of the

8


country of origin characteristics, while assimilating to the mainstream society in other ways (Zhou, 1999). Most interest and attention has focused on Gordon’s first two phases of assimilation: cultural and structural. Studies of cultural assimilation largely focus on language acquisition, while structural assimilation is generally measured by entry into mainstream institutions, such as schools, middle-class jobs, integrated neighborhoods, and intermarriage. Though they do not directly relate to patterns of political incorporation, the findings from these studies serve as roadmaps for such an investigation. Economic Incorporation In the economic sphere, research has been dominated by questions about the ability of post-1965 waves of immigrants to reach economic parity with the native-born, as compared with the pre-1965 immigration flows. Some studies have found that the more recent waves remain consistently behind the native-born economically, and have a more difficult time “catching-up” than did their predecessors (Borjas, 1994, 1998; Frenette & Morissette, 2005; Raijmin & Tienda, 1999). These “lesser quality” immigrants migrate with lower levels of human capital than did earlier waves, placing them at a consistent economic disadvantage. Scholars argue that this human capital disadvantage is then passed down to their children and even grandchildren, delaying economic success for a full three generations, and potentially creating a permanent underclass. A study conducted by Alba, Lutz, and Vesselinov (2001), specifically designed to test such findings, reaches very different conclusions. The researchers find that over several generations, economic differences between the foreign-born and the native-born disappear, supporting straight-line assimilation theory. Others come to even more optimistic conclusions, arguing that differences will disappear over a period of decades rather than generations (Perlmann & Waldinger, 1999, 2000). Given enough time in the United States, today’s immigrants will reach economic parity with the native-born and are not at any greater disadvantage than were earlier waves of immigrants. Although it may take 15 or 20 years to reach economic equality, this is simply the “cost of migration,” and should not be taken as an indication of immigrant quality or of the ability of immigrants to


9

succeed economically in the United States. Mexican migrants are the exception to this finding, consistently remaining at the bottom of the economic ladder. Just as Mexican immigrants are located at the bottom, migrants from India are consistently found at the top, suggesting that today’s migration flows resemble an hourglass shape in terms of the human capital immigrants bring with them. This hourglass shape is the result of a number of factors, specifically selectivity and public policy. Though migration is always selective, with those who migrate being distinct from those who do not, selectivity may be particularly pronounced among economic migrants who tend to be younger, more aggressive, and have higher levels of human capital (Stark, 1991). However, meeting American labor needs is one of the primary goals of current U.S. immigration policy, with those needs being met by individuals at both ends of the human capital spectrum. The result is that levels of human capital are highly correlated with country of origin. Immigrants from Europe, as well as those from certain Asian countries, such as the Philippines, Japan, and India have similar (or even higher) rates of education and income as their native-born counterparts. Immigrants from Vietnam, Mexico and other Latin American and Asian countries have substantially lower levels of income and education than their native-born counterparts, frequently finding themselves in low SES occupations with little room for advancement. For example, Indians who migrate to the United States have tended to come as highly skilled laborers, with 72% of economically active Indian immigrants holding professional or managerial jobs. In contrast, less than 19% of Dominicans hold such positions. Mexicans have similarly low rates, generally migrating as agricultural workers. As a result of the selectivity growing out of migration policy, immigrants from certain countries of origin tend to become more economically incorporated than immigrants from other countries of origin, who remain economically marginalized (Portes & Rumbaut, 1996). Although previous studies have focused on individual levels of human capital and their effect on economic incorporation, we know that this is a process influenced by factors at multiple levels. Family and social networks play a role both in the country of origin and the country of destination, bearing on both the initial decision to migrate

10


and the choice of settlement area (Boyd, 1989; Foner, 1999; Goldscheider, 1995; Palloni, Massey, Ceballos, Espinosa, & Spittel, 2001; Pessar, 1999). Family contacts help migrants to find jobs and places to live, encouraging economic incorporation. Community characteristics may similarly encourage economic integration, with immigrants residing in ethnic enclaves potentially having greater access to jobs, start-up capital, and additional resources than their counterparts living in more integrated areas (Light, 1984; Light & Bonacich, 1988; Portes, 1987). Rates of self-employment and small business ownership tend to be higher in immigrant communities (Sanders & Nee, 1987; Waldinger, 1996). At the same time, living in ethnically segregated areas may translate into higher levels of exploitation and lower levels of income than found among those in the mainstream economy (Chiswick & Miller, 2005; Sanders & Nee, 1987). Finally, a growing body of research has found economic characteristics in the home country to effect the relative permanence of migration (Constant & Massey, 2002; Lindstrom, 1996; Lindstrom & Lauster, 2001). Individuals who have investment opportunities and job prospects in their home communities may migrate to the United States for limited periods of time, returning to the country of origin once a financial goal has been met. These immigrants tend to be less incorporated economically and socially than those with longer-term settlement plans. At the same time, recent research has found that among first generation Mexican migrants in California, home ownership, a sign of socioeconomic incorporation, is positively associated with remittances, an act previously thought to suggest less socioeconomic integration (Marcelli & Lowell, 2005). These most current studies suggest that characteristics in the community of origin and an immigrant’s relationship with that community influence economic incorporation at the individual level, but the relationship is not as unidirectional as one might think. “Looking” towards the home society does not necessarily suggest a lack of incorporation in the host society. Educational Incorporation Educational incorporation is another area that has received a substantial amount of attention from researchers. How do immigrant children


11

measure up to their native-born counterparts in terms of academic achievement? Are they less likely to complete high school? Are they more likely to become burdens on society? Studies of students in California have found that immigrant children have generally performed better than their native-born peers, assuming comparable socio-economic characteristics. Interestingly, the differences in GPA between the native and foreign-born become less dramatic with increased exposure, suggesting a downward assimilation towards the native-born (Portes & Rumbaut, 2001; Rumbaut, 1999; Rumbaut & Portes, 2001; Zhou, 1999). In terms of high-school enrollment, immigrants who enter at younger ages and have longer-term exposure to American society have rates of high-school completion comparable to the native-born (Hirschman, 2001). Mexicans, again, counter the trend, as do Dominicans and recent Cubans, all of whom have lower rates of highschool enrollment. Just as in the studies of economic incorporation, these findings suggest two distinct trends: country of origin matters and length of exposure matters. Where an immigrant hails from is a significant predictor of academic performance. Similarly, length of exposure or the amount of time spent in the United States is also a predictor of the level of educational incorporation, as measured by school enrollment and grades. Studies of educational incorporation have also examined how immigrant children of limited language ability and socioeconomic means come to have such success in school. These studies have frequently identified both family and community factors as predictors of success (Gibson & Ogbu, 1991; White & Glick, 2000; Zhou, 1999). Immigrant children embedded in more closely knit families and ethnic communities and those with greater amounts of social capital tend to have higher levels of educational performance, more consistent with the cultural expectations of their families and communities than children who are more “Americanized.” Social Incorporation Still others studies have examined “social” incorporation, specifically patterns of language acquisition and intermarriage (Lopez, 1999). What characteristics are associated with higher rates of English-

12


language attainment? Who is most likely to intermarry? How do immigrants become a part of the social fabric of American society? A study of destination language acquisition by male immigrants to Canada has found that age at migration is negatively correlated with linguistic attainment, while education, exposure, distance of country of origin, and linguistic proximity of native tongue to destination tongue, are all positively correlated with language acquisition. Once again, the factors of country of origin and length of exposure play key roles in this form of incorporation, supporting in part both straight-line and segmented assimilation theories (Chiswick & Miller, 2001). Another study examining immigrant social incorporation has focused on rates of English language acquisition and intermarriage among Asians and Latinos in California (Lopez, 1999). According to the study, Asians are more likely to learn English and lose the mother tongue than are Latinos, even with comparable levels of exposure. In keeping with the language findings, Asians also have higher rates of intermarriage than do Latinos, suggesting different rates of incorporation across a number of dimensions. These findings imply that factors other than length of exposure predict levels of social incorporation, supporting segmented assimilation theory. At the same time, the difference may be due to greater residential integration among Asians, suggesting that greater incorporation in one area leads to incorporation in other areas, as Gordon had argued. Furthermore, these findings reinforce the role that multilevel factors play in various types of immigrant integration. Political Incorporation Fewer studies have addressed questions of political incorporation. This is not to suggest that the field is entirely barren, but rather that it is less well-developed than are the areas of economic, educational, and social incorporation. By examining these studies in light of the research conducted in other more developed areas, it becomes clear that the political incorporation literature has fallen short. Ideally, this book can fill some of the gaps. Studies of citizenship acquisition, the most common measurement of political incorporation, have repeatedly focused on individual-level characteristics (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996; Smith & Edmonston, 1997). Demographic factors, such as age and


13

country of origin, and economic factors, such as education and income, are recurrently the variables of focus. Although individual-level characteristics are frequently found to be significant predictors of political incorporation, just as they have been identified as important indicators of others types of integration, it is essential to move beyond this individual-level focus. Research exploring other types of immigrant incorporation has identified multilevel factors as affecting levels of integration. For example, school performance has been found to be influenced by family-level factors, namely how tightly-knit the family unit is (Zhou, 1999). Economic incorporation is impacted by the community in which an individual settles, specifically whether or not an immigrant resides in an ethnic enclave (Chiswick & Miller, 2005; Portes, 1987; Sanders & Nee, 1987). Social incorporation, specifically language acquisition and intermarriage, is affected by levels of residential integration (Lopez, 1999). Some scholars have begun to identify the importance of multi-level factors in the political incorporation process, examining family structure and social contact in their research (Liang, 1994; Yang, 1994). Their findings, which will be discussed in detail in ensuing chapters, have shown these characteristics to be significant in the political incorporation process. Unfortunately, more detailed studies of the effect of family and community on political incorporation have not been conducted. On the most fundamental level, an immigrant’s family structure should have implications for the relative permanence of a migration, and thus the likelihood of naturalizing. Migrating as a family, rather than as an individual, suggests longer-term settlement plans, with citizenship acquisition acting as both a reflection and indicator of more permanent settlement. Furthermore, public policy is consistently argued to affect an individual immigrant’s reception into a new society (Portes, 1995b). This is particularly true in U.S. immigration policy, which has implications far beyond the individual. As discussed earlier, the 1965 immigration legislation gives preference to immediate relatives of American citizens over any other group of immigrants trying to gain legal access to the United States, thus encouraging citizenship acquisition for individual immigrants who may wish to help spouses or children migrate. The more recent 1996 Personal

14


Responsibility and Work Opportunity Reconciliation Act, known more commonly as PRWORA or the Welfare Reform Act, may similarly have implications for individual political incorporation within the context of the family (Lister, 1998). The presence of children and/ or the absence of a spouse may have more of an impact on political decisions in the post-welfare policy period than in the pre-policy era. For example, a single-mother who was eligible for food stamps in the pre-policy period may decide she needs to naturalize to protect her and her children’s benefits, even though she has no desire to vote or become involved in the electoral process. In short, studies of immigrant behavior too frequently focus on the individual when they should also focus on the family unit. Community factors, such as population size and composition, may also have implications for an immigrant’s level of incorporation. As scholars have uncovered in other areas of research, friends, neighbors, and acquaintances resulting from one’s residential neighborhood affect social networks, ties to mainstream society, information on jobs outside of the neighborhood, and general life chances (Fischer, 1984; Massey & Denton, 1993; Wilson, 1987). These factors may affect immigrant incorporation. Given that segmented assimilation theorists contend that the three potential paths of assimilation (upward, downward, and middle-of-the-road) are influenced by settlement area (Alba & Nee, 1999; Portes, 1995a; Zhou, 1999), and that empirical research in economic, educational, and social incorporation has supported this argument, community-level characteristics should most certainly be examined in the political incorporation process. Another limitation of many previous studies is their failure to take into account country of origin characteristics in the political integration process. Studies outlined above have generally included country of origin dummy variables in their models to explore how naturalization rates vary by nativity, but these studies have done little to tell us what it is about the country of origin that leads to such variation. As Chiswick and Miller (2001) argue in their study of destination language acquisition, “dichotomous variables for country of birth may be regarded as a measure of what we do not know: that is, of the relevant unmeasured behavioral variables that underlie country-of-birth effects” (pp.395-396). If, for example, Koreans consistently have the highest naturalization rates among men, but Filipinos consistently have the


15

highest naturalization rates among women (Smith & Edmonston, 1997), then we need to better understand whether there are factors in the country of origin that lead to such consistent differences. Although a few studies have taken the step to examine the relationship between home society characteristics and political incorporation (Jasso & Rosenzweig, 1990; Liang, 1994; Yang; 1994), country of origin factors have been frequently overlooked. The further weakness of political incorporation studies lies in their singular focus on citizenship acquisition (Jones-Correa, 2001; Rumbaut, 1999). Naturalizing is certainly one dimension of political incorporation, and an important one at that, but it is far from the only form of political integration. Immigrants can become incorporated politically through a wide range of formal and informal activities. Just as studies of economic, educational, and social incorporation use multiple measurements to assess the particular field being explored, so too should studies of political integration. Political incorporation studies that have examined voter turnout are even less common and more limited than those examining citizenship acquisition. The absence of attention to voting in the political incorporation literature is largely due to the dearth of quantitative data available (Bass & Casper, 1999, 2001). Research that has examined immigrant voting has many of the same limitations as citizenship studies, but to an even greater extent. These studies, which will be discussed in greater detail in upcoming chapters, have focused on the same individual-level factors examined in the citizenship literature, namely education, income, and place of origin (Bass & Casper, 1999, 2001). Unfortunately, some researchers have examined voting patterns by regions of the world (Bass & Casper, 1999, 2001), conflating the divergent migration flows from different countries of origin and ignoring the migration selectivity that may be taking place (Cain, Kiewiet & Uhlaner, 1991; Cho, 1999; White & Sassler, 2000). As one group of scholars writes about their own work on Asian and Latino immigrants “The divergence of our findings concerning Asians and Latinos thus underscores…[the] admonition against conflating the political experiences of different ethnic groups” (Cain et al., 1991, p.416). Others have examined a particular immigrant group, or a particular part of the country (de la Garza & DeSipio, 1992, 1996; DeSipio & de la Garza, 1998; Sierra, Carillo &

16


DeSipio, 2000). The limitation of these studies has been an inability to compare the voting trends of different immigrant groups around the United States Although individual-level characteristics are known to play an important role in various types of immigrant incorporation, the research conducted on economic, educational, and social integration also suggests that factors at the family, community, and country of origin level matter. One study of immigrant voter turnout has essentially come to this conclusion, finding that Latino voter turnout varies substantially from state to state (de la Garza & DeSipio, 1996). Some states, such as California and Arizona, experienced increases in Latino voter turnout by more than 30% from one election cycle to the next, while states such as Colorado and New York had declines in the rate of Latinos going to the polls. These findings suggest that contextual factors matter, but research has failed to identify what it is about settlement communities that differentially influence immigrant voter turnout. Additionally, immigrants enter the United States with very different political experiences, based upon the political system in their respective country of origin. Individuals raised in democracies are taught, either implicitly or explicitly, the accepted behaviors and values associated with democratic governance. Those raised in nondemocratic regimes similarly learn the conventional political behavior in those societies. If people have not been raised in countries with democratic traditions, they will not likely know how to participate in a democratic society, “Having come from other places and having been socialized to alien political norms, it is unlikely that that the majority will possess the necessary ideological credentials” (Harles, 1993, p.2). Such a statement is backed up by empirical research conducted on immigrant voting in other countries (Black et al., 1987; Finifter & Finifter, 1989). Research finds variation across immigrant groups, with those migrating from more similar political regimes able to translate their previous knowledge and experiences. Specifically, someone migrating from a democracy to a democracy should be more likely to vote than someone migrating from a communist country to a democracy. These findings argue strongly for an examination of the relationship between country of origin characteristics and individual political incorporation.


17

THE ORGANIZATION OF THIS VOLUME Scholars stress “the need for comparative cross-national studies of citizenship, membership, and modes of political inclusion and exclusion” (Rumbaut, 1999, p.191), as so little is understood about the political incorporation process. Previous work has been too focused on identifying individual-level factors that affect political incorporation, ignoring influences at other levels that selectively affect who migrates and who becomes incorporated. Qualitative work that has examined influences at multiple levels has failed to do so comparatively, instead focusing on one particular immigrant group (Forment, 1989; JonesCorrea, 1998; Karpathakis, 1999). The goal of this volume is to add to the political incorporation literature by empirically testing the influence of individual, family, host community, and country of origin factors on two types of formal political incorporation—naturalizing and voting—across multiple immigrant groups in the United States. Specifically, I will address the following questions: How are factors at different levels associated with political incorporation, net of other factors? How do these multi-level characteristics interact with one another to encourage or discourage political integration? Do patterns vary across types of political incorporation, i.e. are citizenship and voting differentially motivated? Do patterns vary across immigrant groups? The following chapter examines the data sources and methodology used to systematically answer the questions posed here. I start in Chapter 3 to examine the relationship between individual-level factors and their relationship with political incorporation. By beginning with individual-level factors, I am able to establish a baseline of political incorporation, as well as to compare the findings from this data source with research conducted using other data. Chapter 4 explores the effect of family structure, specifically marital status, both net of individual level factors, and in interaction with them. In Chapter 5, I introduce community characteristics into the model, to explore how factors such as population size, density, and segregation influence political incorporation. In Chapter 6, my final empirical chapter, I will examine the relationship between country of origin characteristics and individual political integration, again both net of individual, family, and community factors and in interaction with them. By including these many factors in one study, I can systematically demonstrate what

18


affects two different types of political incorporation, whether the factors vary across groups, and whether support is found for straightline or segmented assimilation, or something in between.

CHAPTER 2

Assessing Immigrant Political Activity The lack of studies on political incorporation is certainly not a reflection of the level of interest (or lack thereof), but is instead a reflection of quantitative data availability. Quantitative studies of immigrant political incorporation, by definition, require data on both immigration status and measures of political incorporation. Some of the more common sources of data, such as the U.S. Census, simply fail to include questions related to political participation, other than citizenship status. Other data sources that focus explicitly on political participation, such as the National Election Study (NES), sample only citizens, making it impossible to measure the foreign-born who select citizenship versus those who remain non-citizens. Further, the NES fails to ask for country of origin, likely due to the presumption that American citizens are overwhelmingly native-born. As a result, it is impossible to determine whether naturalized citizens from different countries of origin behave similarly to or differently from one another, or to or from the native-born. The lack of quantitative data has led to a variety of responses from immigration scholars. Most commonly, researchers interested in immigrant political incorporation have used qualitative methodology to examine the process. Case studies (Jones-Correa, 1998; Karpathakis, 1999; Rivera-Salgado, 1999) have provided us with insights into the mechanisms encouraging or discouraging political incorporation, but their general applicability is limited without further empirical testing. Theoretical works (Harles, 1993; Plotke, 1999) similarly contribute to the discussion, but without rigorous empirical testing, they remain hypotheses to be explored. Others have “made due” with the quantitative data available, or have even collected their own survey data (de la Garza & DeSipio, 1992, 1996). Unfortunately, these quantitative studies are of limited usefulness as similar issues emerge again and again: the lumping together of all immigrants into one “foreign-born” group (Black et al., 1987); the equally problematic grouping and study of immigrants by broad region of origin, rather than 19

20


country of origin (Bass & Casper, 1999, 2001), or the focus on immigrants from a singular area of origin (Cho, 1999; de la Garza & DeSipio, 1992, 1996). Other quantitative studies that have moved beyond some of these limitations have tended to explore immigrant political incorporation solely through the avenue of citizenship (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996). The limitations of many of these studies have grown out of the common problem of insufficient data by which to study more precisely various questions in the area of political incorporation. Fortunately, recent changes in one data source, the Current Population Survey (CPS), have made the study of immigrant political incorporation more feasible. THE CURRENT POPULATION SURVEY Background This project relies primarily on the CPS, a joint venture between the U.S. Census Bureau and the Bureau of Labor Statistics. The primary goal of this month-by-month, national survey is to gain information on workforce participation. A certain body of questions on labor force involvement is asked every month, providing comprehensive data on movements in and out of the workforce. The CPS has been the primary source of information on labor force participation in the United States for over fifty years. As well as serving as the primary source of American labor statistics, the CPS also asks a series of demographic questions and, in 1994, began including a question on citizenship status. The addition of this citizenship question has meant that social scientists no longer have to infer citizenship by length of time in the United States, as they were forced to do prior to 1994. The specific citizenship question on which I will be relying asks “In what citizenship group do you belong?” with the possible responses including: 1) Native, born in US; 2) Native, born in Puerto Rico or US outlying area; 3) Native, born abroad of US parent(s); 4) Foreign Born, US citizen by naturalization; 5) Foreign Born, not a US citizen.

Assessing Immigrant Political Activity

21

My citizenship analysis subdivides responses to this citizenship question into a dichotomous variable, looking exclusively at individuals who identify themselves as being in categories 4 or 5. Furthermore, some months of the CPS have special supplements, focusing on a particular aspect of American society. Every election year, the November supplement to the CPS, entitled the “Voter and Registration Supplement,” includes questions on whether or not a respondent participated in the most recent set of elections. The specific voting question that I use to examine voter participation is asked as follows: “In any election some people are not able to vote because they are sick or busy or have some other reason, and others do not want to vote. Did (you/name) vote in the election held on Tuesday, November __?” The format of this question is intended to limit over-reporting by removing the stigma of non-voting. The question is only asked of those 18 years of age or older, who reported U.S. citizenship by birth or naturalization. The CPS’s inclusion of citizenship and voting questions makes it possible to accurately explore two types of formal political participation across a large number of immigrants—an improvement over what other data sets allow. Additionally, the survey includes many demographic characteristics associated with immigrant political participation. The respondent’s country of origin is asked as an open-ended question, and reads “In what country were you born?” Gender is dichotomous: male or female. The education variable is categorical, giving respondents a series of possible responses. I have recoded this variable to create two new variables, one that measures lower levels of education and one that measures higher levels. Country specific migration flows tend to be characterized by higher or lower levels of skill or education, due to selectivity. Creating the education-level variable in this manner does not appear to have country-specific effects, however, as each sub-group included in the study has enough educational variability. Income is similarly reported as a categorical variable. I have created a new economic variable, one that measures whether an individual is above or below the poverty line, by combining information on total family income and total family size. Individuals are assigned a poverty status based on whether their combination of family income and size place them above or below the national poverty

22


threshold, as dictated by the Department of Health and Human Services, in the year in which they participated in the CPS. I have chosen to measure the number of years that an immigrant is eligible for citizenship, rather than his or her length of residence in the United States, as documented immigrants are not able to participate formally in the political sphere for the first five years. 1 I have calculated length of eligibility in the United States by using the “year of entry” variable in the survey. The “year of entry” variable is categorical, asking in what period of time an individual entered the United States. I have calculated the mid-point of each of these categories, and subtracted that year from the year in which the respondent participated in the survey. I have then subtracted five additional years off of the total, as this is the usual length of time required before an immigrant has the opportunity to naturalize. The result is a “quasi” continuous variable, measuring the amount of time one has been eligible to formally participate.2 Table 2.1 illustrates the operationalization of these individual characteristics. The construction of this length of eligibility variable is not ideal for multiple reasons. First, the immigrant’s time of entry is collected as a categorical variable, with categories of varying lengths. As a result, the final variable is not truly continuous. I could simply have used the initial categories to measure length of time in the United States, or collapsed them into a handful of time periods. Doing so would make it 1

In a few instances (marriage to an American citizen, military service), the residency requirement is cut to three years. Because these cases represent such a small percentage of all immigrants in the U.S., I have chosen to use the standard five year residency requirement. 2

Correspondingly, age at migration could be calculated in a similar manner: age at migration=current age-(survey year-year of entry). The age at which an individual migrates to the United States is predictive of immigrant assimilation, with those arriving at younger ages more likely to become incorporated on a host of fronts. Immigration scholars even refer to immigrants who migrate as children as the “.5” generation. Unfortunately, it is not possible to include both current age and age at migration in the same model, due to issues of multicollinearity. As a result, I include only age at the time of the survey and length of eligibility. This is not to suggest that age at migration is unimportant in the political incorporation process.

TABLE 2.1: Operationalization of Individual Characteristics Name

Meaning

Measurement

Coding

Source

AGE

R’s age at time of survey participation

Self-reported age

Continuous

CPS

AGE

R’s age at time of survey participation, squared

Self-reported age* Self-reported age

Continuous

CPS

FEMALE

R’s sex

Self-reported sex

CPS

LESSHS

Whether R has less than a high school degree or GED

Calculated from multiple categories from self-reported citing of years of education completed

1=Female 0=Male 1=Less than high school degree/ GED 0=High school degree/ GED

MOREHS

Whether R has more than a high school degree or GED

Calculated from multiple categories from self-reported citing of years of education completed

1=More than high school degree/ GED 0=High school degree/ GED

CPS

POVERTY

Whether R is living below the poverty line, as measured in survey year

Calculated from self-reported size of family and self-reported family income, compared with poverty line for family of that size and income in survey year

1=At or below poverty line 0=Above poverty line

CPS, Department of Health and Human Services

WORK

Whether R reports working for pay outside of the home

Self-reported workforce participation

1=Works outside of home 0=Does not work outside of home

CPS

ELIG

Length of time R has been eligible to naturalize and vote in the US

Survey year-midpoint of entry category-5

Quasi continuous

CPS

YEAR96

Whether survey year is 1996

Year of survey participation

CPS

YEAR98



YEAR00



1=1996 0=1994 1=1998 0=1994 1=2000 0=1994

2

CPS

CPS CPS

24


difficult to determine the effect of cohort versus period versus length of eligibility in the United States. The length of eligibility variable is also limited by lack of information on entry status. Many, particularly Mexican immigrants, may have entered the United States as undocumented. By subtracting 5 years from the date of entry, I assume that everyone has initially entered as a documented migrant and is therefore eligible for citizenship five years later. This is certainly not the case. As a result, we could find that the length of eligibility variable does not appear to have a relationship with naturalizing or voting, or has a more limited relationship for certain groups, resulting from “starting the clock” on political eligibility too soon. I have included three control variables of age, age2, and workforce participation. The age variables are both continuous. I have included both age and age2 in the models, as previous studies have found a curvilinear relationship between age and political participation (Converse, 1969; Niemi, Stanley, & Evans, 1984). The workforce participation variable is categorical and measures whether or not the respondent is in the workforce. I have also included control variables for the year in which the respondent participated in the survey. Additionally, the CPS provides some limited information on a respondent’s family and household structure. In some years, respondents are asked how many, if any, of their own minor children are living in the same household. Questions on one’s marital status are also included and provide data on whether a respondent is single or married, and if married, whether they reside with or apart from the spouse. Table 2.2 outlines the operationalization of these family characteristics. Sampling The CPS is a multistage stratified probability sample of approximately 56,000 housing units from 792 sample areas. The first stage of the sample takes place by dividing the United States into primary sampling units, or PSUs, comprised of a single metropolitan area, a county, or a group of smaller counties. A sampling of housing units, updated from the most recent U.S. Census, is then drawn from each PSU. Individuals within each household are selected through a simple, random sample to represent the civilian non-institutional population, 16 years of

TABLE 2.2: Operationalization of Family Characteristics Name

Meaning

CHILDREN

Measurement

Coding

Source

Number of R’s own minor children in Self-reported number of R’s own the household minor children in the household

Continuous

CPS

UNMARRIED

Whether or not R is married

Self-reported measure of R’s marital status

1=Unmarried CPS 0=Married and residing with spouse

SPOUSE ABSENT

Whether or not R is living with spouse

Self-reported measure of R’s marital status and household structure

1=Married, R and spouse CPS residing apart 0=Married and residing with spouse

26


age and older. Each household is surveyed for four months in a row, cycled out, and then brought back into the survey sample eight months later to be interviewed for another four months in a row, before being discarded. The data were designed to be used as cross-sectional, rather than panel or longitudinal data. 3,4 Weaknesses of the CPS I have chosen to use the CPS to systematically examine the political incorporation process as it contains information on citizenship status and voter turnout, the two most formal measurements of political integration. Furthermore, the CPS allows for the study of these two processes across multiple immigrant groups, providing data on country

3

For a more detailed discussion of the CPS, please refer to Technical Paper 63, entitled “Design and Methodology,” published by the Bureau of Labor Statistics and the U.S. Census Bureau. It can be found at the following address: http://www.census.gov/prod/2000pubs/tp63.pdf

4

The CPS weights are largely designed to account for state-level sampling differences, as well as to correct for other types of over and under representation. The CPS weight “PWCMPWGT,” is an individual person’s composite final weight and is intended to produce population level statistics for the adult civilian population ages 16 and above in the United States, based upon the CPS sample. This weight takes into account 1) the basic weight, 2) adjustments for special weighting, such as might be done for a particular racial group, 3) non-interview bias, 4) first stage ratio adjustment factor, and 5) second stage ratio adjustment factor. This is the weight used by the Bureau of Labor Statistics to tabulate the official published labor force numbers. Initial descriptive statistics were run using this individual-level composite final weight. Descriptive statistics have also been run without weights and with an individual second stage weight provided by the CPS. This second stage weight, “PWSSWGT,” only weights for the second stage sampling, as opposed to the multiple factors that are used to comprise the composite weight. Though the total population sizes associated with each of these three different methods of weighting (none, PWCMPWGT, and PWSSWGT) vary, the percentages remain very similar to one another. Because the multivariate regression models presented in later chapters are unweighted, the final descriptive statistics presented here are the unweighted statistics.


27

of origin, time period of entry, and a host of other important demographic characteristics. However, like most secondary data sources, the CPS is not without limitations. The data on which I rely are self-reported measurements of citizenship status and voter turnout. Because both questions touch on sensitive issues, albeit for different reasons, both are susceptible to misreporting. In the instance of citizenship status, individuals who are in the United States illegally may be hesitant to report that they are not citizens. Although the survey does not touch on questions of legality, the question of citizenship status may keep some people from answering the survey honestly and others from answering the survey entirely. A preliminary examination of the data suggests that the citizenship status reported by the respondents seems to be generally accurate. I have checked this variable by constructing cross-tabs with other variables that give hints at citizenship status, specifically questions on the respondent’s country of origin and year of entry to the United States. For instance, by finding that virtually everyone who reports that they are native-born also report the United States as their birth place suggests accuracy in the data. Finding that those who cite their status as naturalized citizens and who also report that they have been in the country for at least five years (the minimum amount of time needed to gain citizenship) also points to consistent self-reporting. The lack of information on visa type or legal status is a significant weakness of these data, particularly for this study, as the variation in legal migration flows by country of origin could be appearing as country of origin effects. For instance, lower naturalization rates among Mexican immigrants could speak more to the selectivity of that particular migration flow, comprised largely of undocumented workers, than to country of origin effects. Mexican immigrants might be as likely to naturalize as any other group, given the opportunity. Unfortunately, these data do not allow for the identification of an immigrant’s legal status. The best that can be done is to interpret the results with caution. A perennial concern of survey data is non-response bias. In this study, the concern lies with individuals choosing not to participate due to their immigration or citizenship status. Some comfort can be found in the fact that previous scholars who tracked the naturalization trajectories of Mexican immigrants found little difference between

28


those who “went missing” and those followed throughout the process (Portes & Bach, 1985; Portes & Curtis, 1987).5 Still, non-response bias remains a concern. Finally, over-reporting on questions of voter turnout is common, as respondents want to appear to be “doing their civic duty.” A study of Latino voter turnout in the 1996 election found significant overreporting when aggregate self-reports were examined against aggregate validated rates (Shaw, de la Garza, & Lee, 2000). The self-reported voter turnout rate in the CPS is lower than what has been recorded in other surveys, such as the National Election Study, viewed as the “gold standard” of electoral data. In comparing self-reported rates of voter turnout of all respondents in the CPS with the Federal Election Commission’s actual tally for the years 1994, 1996, 1998, and 2000, I have found an over-reporting rate of approximately 10%. For instance, in 1994, the CPS has a self-reported turnout of 85,700,000 voters across the country versus the Federal Election Commission’s tally of 75,105,860. In 1996, the CPS reports 105,000,000 Americans going to the polls versus 96,456,345 actual votes cast. In 1998, 73,117,022 Americans actually cast ballots versus 83,100,000 self-reported votes. Finally in the 2000 election, 110,800,000 Americans self-reported voting, versus an actual turnout of 105,586,274. The primary concern with over-reporting is that it may not be randomly distributed, with certain groups, such as the better educated, over-reporting at higher rates than their less educated counterparts. Though this study is concerned with actual political participation as a measurement of political incorporation, over-reporting tells us something in its own right. That an immigrant incorrectly reports having voted suggests, in its own way, a form of political incorporation, by recognizing the normative pressure to vote in American society. The collection of certain demographic information is also imperfect. Characteristics such as education, income, and year of entry were collected as categorical variables, rather than as continuous 5

Portes and Bach (1985) were able to determine similarities between the tracked and “lost” Mexican immigrants by examining initial records and demographic characteristics collected on the immigrants upon their entry to the United States.


29

measurements. Other desirable information, specifically visa type and level of English-language proficiency, was simply not collected. Family and community-level data are also limited, or missing entirely. Previous research suggests that family income 6 has a consistent and significant impact on political incorporation, with higher levels of income leading to higher rates of incorporation. The CPS records family income,7 but the lifestyle afforded by the family income is very dependent upon the family size. Receipt of social welfare benefits, such as food stamps and cash transfers, are determined by both family income and family size. The 1996 Welfare Reform Act limited many means-tested benefits to citizens, increasing the incentive to naturalize among those at the lower levels of the economic spectrum. Because the CPS does not include information on welfare receipt or poverty status, potentially an important factor in political incorporation, I have constructed a variable that will place an individual’s family above or below 100% of the poverty line. Although I am able to correct for the lack of information on welfare dependency, other absent information on the family, household, and community is not so easily corrected. For instance, relationships and social networks outside of the household likely affect political incorporation, as do characteristics in the country of origin. Unfortunately, this information is absent in the CPS. I supplement the CPS data with other data sources to correct for the limitations found in the primary data source.

6 Ideally, one would be able to examine both the impact of individual and household income on political incorporation. Unfortunately, the individuallevel income variable has a large percentage of missing cases, making the household income variable the better choice. 7

The variable “HUFAMINC,” described as the total household-family income in past 12 months, reads as follows: “I am going to read a list of income categories. Which category represents the total combined income of all members of this Family during the past 12 months. This includes money from jobs, net income from business, farm or rent, pensions, dividends, interest, social security payments and any other money income received by members of this Family who are 15 years of age or older?” The possible responses are categorical.

30


A final weakness of the CPS, as it relates to this study, is the lack of over-sampling among immigrants. In order to achieve more reliable estimates, I have pooled four years of data. Furthermore, I have made the decision to include two supra-national groups in my analysis: former Soviets and Southeast Asians, respectively. This is not to suggest these two grouped categories are monolithic religiously, culturally, politically, or educationally. I have formed these supranational groups in an attempt to gain some insight into the political incorporation of these immigrants, whom I would otherwise be unable to include due to small sample size. Immigrants from both the former Soviet countries, as well as those from the Southeast Asian countries of Vietnam, Laos, and Cambodia, are too significant in American immigration flows to disregard entirely. Even with these shortcomings, these data continue to be the best available for a large-scale examination of immigrant political integration in the United States. Missing Data As with all data sets, some of the cases in the CPS have missing data. With respect to the variables of age, education, and income, I have filled in an individual’s missing data with the median value from his or her respective immigrant group. This is a common method of dealing with missing information. I have chosen to drop cases in which the citizenship status or voting information is missing. As these are the dependent variables on which this study focuses, I thought it imprudent to infer the missing information.8 The Sample I have drawn my individual cases from the November 1994, 1996, 1998, and 2000 CPS (“Voter and Registration Supplement”). Because the sample sizes of some of the immigrant groups are small, I have merged all four years of data to gain greater statistical reliability. Each 8

I have compared the cases with missing information to those with complete information. Though there are statistically significant differences between those who answer the citizenship or voting questions versus those who do not, this study is limited to cases with information on the respective dependent variables.


31

wave of the survey contributes approximately ¼ of the cases to my pooled sub-samples. Those included in the sub-samples must be at least 18 years old, as this is the age at which someone is eligible to acquire citizenship and vote. Among immigrants, individuals must have been in the country for at least five years, as this is generally the minimum required length of residency in order to be eligible for citizenship. The individuals must also have answered the question on citizenship status. To be included in the voting sub-samples (see Samples B and C), the individuals must be citizens of the United States and have answered the voting question. Table 2.3 illustrates each of the samples used in this study. In addition to the qualifications cited above, the overall sub-sample from the CPS is further limited to Native-Born Whites, Blacks, Asians, and Latinos, and to immigrants from Britain, Italy, the former Soviet Union, Cuba, Canada, Mexico, the Philippines, China, India, and Southeast Asia (Vietnam, Laos, Cambodia). 9 I have chosen these immigrant groups as they have varied immigrant histories in the United States, reasons for immigrating, and political and economic situations in their countries of origin, allowing us to see how variations in different characteristics affect the political incorporation process. Sample A: “The Naturalization Sample” Sub-sample A is limited to eligible immigrants (age 18+ and residing in the United States for at least five years) from the ten chosen immigrant groups who answered the citizenship question. Native-born Americans are not included in this sub-sample, as those born in the United States are citizens by law. As Table 2.3 shows, Sub-sample A is comprised of 17,019 cases, consisting of 7,373 Mexicans, 1,157 Cubans, 1,060 Canadians, 854 Britons, 967 Italians, 638 Former Soviets, 1,768 Filipinos, 900 Indians, 1,255 Southeast Asians, and 1,047 Mainland Chinese. Models predicting citizenship are run on Sub-sample A.

9

I refer to the “overall” sample because sub-samples will be drawn from this larger sample as I examine particular groups throughout this project. The subsamples are defined in this section. Please refer to Table 2.3 for a summary of the three different samples drawn from the CPS.

TABLE 2.3: Definition of Sub-Samples Sub-Sample A: Citizenship Eligible

Sub-Sample B: Voting Eligible Foreign & Native

Sub-Sample C: Voting Eligible Foreign O nly

M EXICANS, N=7,373

M EXICANS, N=1,492

M EXICANS, N=1,492

CUBANS, N=1,157

CUBANS, N=693

CUBANS, N=693

CANADIANS, N=1,060

CANADIANS, N=551

CANADIANS, N=551

BRITONS, N=854

BRITONS, N=430

BRITONS, N=430

ITALIANS, N=967

ITALIANS, N=622

ITALIANS, N=622

FORM ER SOVIETS, N=638



FILIPINOS, N=1,768

FILIPINOS, N=1,078

FILIPINOS, N=1,078

INDIANS, N=900

INDIANS, N=404

INDIANS, N=404

SE ASIANS, N=1,255

SE ASIANS, N=560

SE ASIANS, N=560

CHINESE, N=1,047

CHINESE, N=463

CHINESE, N=463

NATIVE-BORN W HITES, N=254,097* NATIVE- BORN BLACKS, N=29,111* NATIVE-BORN ASIANS, N=3,102 NATIVE-BORN LATINOS, N=10,692** N=17,019

N=303,643

N=6,641

* Non-Hispanic. **The Latino category is comprised of those who identified racially as Black or White, and ethnically as Latino. Source: The November 1994, 1996, 1998, 2000 Current Population Surveys.


33

Sub-Sample B: “The Voting Sample” Sub-sample B, the larger voting sample, includes White, Black, Asian and Latino native-born Americans who are eligible to vote (18+) and naturalized citizens, all of whom must have answered the voting question. The overall voting sample consists of 297,002 native-born Americans and 6,641 naturalized immigrants, 10 comprised of 1,492 Mexicans, 693 Cubans, 551 Canadians, 430 Britons, 622 Italians, 348 former Soviets, 1,078 Filipinos, 404 Indians, 560 Southeast Asians, and 463 Mainland Chinese, for a total of 303,643 cases. Sub-Sample C: “The Immigrant Voting Sample” Sub-sample C is culled from Sub-sample B, but includes only the naturalized citizens who answered the voting question. This sample is comprised of 1,492 Mexicans, 693 Cubans, 551 Canadians, 430 Britons, 622 Italians, 348 former Soviets, 1,078 Filipinos, 404 Indians, 560 Southeast Asians, and 463 Mainland Chinese, for a total of 6,641 cases. SUPPLEMENTAL DATA SOURCES 1990 U.S. Census The CPS provides the individual cases, and most of the necessary information regarding individual and family characteristics, but these data lack some important indicators needed for a more complete examination of immigrant political incorporation. Although the CPS provides some geographic information for each respondent, including state of residence and MSA, it fails to provide characteristics of the respective community. In order to gain some insight into the effect of community-level factors, I use measurements of residential segregation to act as a proxy for social and community networks. Fortunately, by providing an immigrant’s MSA, I am able to merge metropolitan level characteristics from the 1990 U.S. Census with individual cases. There are a total of 262 MSAs represented in the CPS.

10 The total number of naturalized cases from the first sample does not equal the total number of cases in the voting sample, as some naturalized immigrants were dropped for failing to answer the voting question, as was discussed in the Missing Data section of this chapter.

34


I merge indices of residential dissimilarity, or residential segregation, by metropolitan area. The indices of dissimilarity are calculated as follows: N D=1/2Σ Pi1-Pi2 i=1 P1 P2 where D is the level of dissimilarity, P1 and P2 are the populations of the two groups of interest in a given geographic area, such as a state or metropolitan area, while the I values are the populations of those groups in the smaller geographic area of focus, i.e. an MSA or census tract (White & Glick, 1999). The residential segregation variable, calculated from the 1990 U.S. Census, will be attached to the individual cases to determine how and to what extent community, as measured by residential segregation, impacts individual political incorporation. In addition to a measurement of residential segregation within a metropolitan area, I will also include a measure of immigrant concentration, also calculated from the 1990 U.S. Census. By including the total percentage of foreign-born in a given metropolitan area, I have another proxy for social ties. Larger percentages of foreign-born might suggest a higher likelihood of pan-ethnic identity formation, leading to greater political incorporation. Alternatively, smaller percentages of foreign-born in an area might lead to greater assimilation by immigrants. Other community-level characteristics are also borrowed from the 1990 U.S. Census and merged with the CPS data, including overall population size of an MSA and whether it is defined as metropolitan or rural. Table 2.4 illustrates these community characteristics and how they are measured. U.S. Census Bureau Just as factors in the host community may influence an individual migrant’s level of political incorporation, so too may characteristics of the country of origin. The distance of the home society to the United States is one such factor that likely influences both an individual’s initial decision to migrate to the United States and how incorporated he or she becomes once here.

TABLE 2.4: Operationalization of Community Characteristics Name

Meaning

Measurement

Coding

Source

METROPOLITAN

Urban/suburban v. rural location

Whether R lives in urban/suburban v. rural area

1=Urban/suburban 0=Rural

1990 U.S. Census

POPULATION

Population in R’s reported MSA

Log of overall population in R’s reported MSA

Continuous

1990 U.S. Census

PERFOR

Percentage of overall population in (Raw number of foreign-born in R’s reported MSA/ Continuous R’s reported MSA that is foreign- divided by the total population in R’s reported born MSA)*100

1990 U.S. Census

W-B DIS

Level of White-Black dissimilarity D=1/2Σ [(Black pop in MSA/Black pop in state)in R’s reported MSA (White pop in MSA/White pop in state)]

Continuous

1990 U.S. Census

W-L DIS

Level of White-Latino dissimilarity D=1/2Σ [(Latino pop in MSA/Latino pop in state)- Continuous in R’s reported MSA (White pop in MSA/White pop in state)]

1990 U.S. Census

W-A DIS

Level of White-Asian dissimilarity D=1/2Σ[(Asian pop in MSA/Asian pop in state)in R’s reported MSA (White pop in MSA/White pop in state)]

1990 U.S. Census

Continuous

36


The Census Bureau provides geographic mileage between most major cities in the United States and most major cities around the world. Distance between home and host society is measured by the mileage between the country of origin’s capital city and the closest port of entry to the United States. Although the Census Bureau reports these distances as continuous, I have chosen to categorize this variable by breaking it into three categories of less than 1,000 miles, 1,000 to 5,000 miles, and more than 5,000 miles.11 Once again, these state-level factors are attached to the individual immigrant cases. Table 2.5 illustrates the country of origin factors included in the analysis and how they are measured. The Freedom House Index (FHI) Additional country of origin characteristics, such as the level of political and civic freedom in the home society, may affect an individual immigrant’s propensity to stay in the United States or return to his or her country of origin. Similarly, it may impact an individual’s predisposition to naturalize or vote, if longer-term settlement is planned. Freedom House, an independent, not-for-profit, pro-democracy organization founded more than 60 years ago, has developed an index that includes two measurements, one gauging the level of civil rights in a country and one measuring the level of the country’s political freedoms. Both the level of civil rights and political rights are measured on a scale of 1 through 7, with scores increasing as levels of freedom decrease. The two measurements are combined to create an overall assessment of the country’s level of civic and political freedom. The overall assessment is constructed as a 3-point scale, with higher numbers suggesting lower levels of freedom. A score of “1” means the country has high levels of civic and political freedom, a “2” suggests partial freedom, and a “3” means little if any freedom. The FHI has been calculated for countries around the world every two years since 1972. The country of origin FHI score is attached to individual respondents to make this state-level factor an individual-level 11

I have run the models with the distance variable as both continuous and categorical. The significance and direction is similar in both models, but the results are more easily interpreted with the categorical version.


37

characteristic. I have allowed the level of freedom in the country of origin to vary by year of immigrant entry to the United States to more precisely measure the civic and political conditions in the country of origin when the individual migrated to the United States. In actuality, however, the level of freedom, or lack thereof, has remained very stable for most countries over the 30-year period this information has been collected. Immigrants who entered the United States prior to 1972 receive the earliest FHI assessment. The FHI is considered reliable as a measurement of the political freedom of a country—the purpose of the index. The most common criticism of the index is that such a measurement comprises much more than simply the political state of the country. More democratic countries tend to be more developed countries, with higher levels of health care, income, and education. I attempt to remedy part of this problem by including other state-level characteristics. Unfortunately, it is not possible to include every variable that is possibly subsumed by a measurement of political freedom. Human Development Indicators (HDI) The World Bank, as part of its overall mission, collects annual statistics on the social and economic development of virtually every country around the world. I utilize the HDIs, collected annually by the World Bank, to gain greater insight into the countries of origin. Like the level of democracy in the country of origin, other statelevel factors may impact the likelihood of an individual becoming politically incorporated in the United States. The level of economic development, or GDP, suggests the extent to which an individual has economic opportunities in the country of origin. Individuals from countries with greater levels of economic development will likely have lower incentives to become politically incorporated through naturalizing and voting, while those from less economically developed countries likely have greater incentives to remain in the United States

TABLE 2.5: Operationalization of Country of Origin Characteristics Name

Meaning

OPPRESSION

Measurement

Coding

Source

3-point scale of civil and political Level of civil and political oppression oppression

Continuous

FHI

Continuous

HDI

1=English 0=Not English

CIA Factbook

GDP

Per person gross domestic product

Log of per person gross domestic product

ENGLISH

Language in country of origin

Whether English is an official language in country of origin

CLOSE GEO

Distance from capital city in country 1=less than 1,000 miles of origin to closest U..S. port of 0=more than 5,000 miles U.S. Census Bureau How close country of origin is to U.S. entry

MED GEO

Distance from capital city in country 1=1,000-5,000 miles How close country of origin is to U.S. of origin to closest U.S. port of entry 0=more than 5,000 miles U.S. Census Bureau

EDUC RATIO

Ratio of boys to girls in primary school

Percentage of boys enrolled in primary school/ percentage of girls enrolled in primary school

Continuous

HDI


39

and acquire citizenship. I attach the 1997-1998 reported GDP in the country of origin to each case to again make this an individual-level characteristic. Individual opportunities may also vary by gender in many countries of origin. Life opportunities, though somewhat difficult to operationalize, can be measured in part by educational opportunities, as education dictates many other life chances. The World Bank Indicators include information on the percentages of school age boys and girls who are enrolled in primary school. From these two variables measuring school enrollment by gender, I have constructed a ratio of percentage of school age boys to girls enrolled in school as a proxy for gender equity in the country of origin. Again I attach this state-level indicator to the individual cases. The CIA Factbook The language spoken in the country of origin may also have implications for an immigrant’s level of political incorporation. Specifically, immigrants who migrate from countries where English is an official language may be more likely to become politically integrated as a result of linguistic congruence. The CIA Factbook, a data source compiled annually by the Central Intelligence Agency in Washington, D.C., provides economic, political, and cultural information on every country around the world, including the most commonly used languages. The CIA Factbook cites the countries of Britain, Canada, and the Philippines as using English as an official language. The country of India uses it as an “associate” language, a status that suggests less universality. I examine the English-language variable in two ways: once including India and once excluding India. I have chosen the more conservative construction, limiting the English language variable to countries where it is used as an official language. This linguistic measurement is again attached to the individual cases to make it an individual-level characteristic of each immigrant. THE MODELS Because the two outcome variables, naturalizing and voting, are both dichotomous, multivariate logistic regression models, using SAS’s “proc logistic” procedure, are used. Figure 2.1 outlines the variables

40


that are included in each of the four levels of analysis (individual, family, community, and country of origin), and the posited direction of the coefficients. Factors that indicate greater levels of social or economic incorporation into American society, such as education, income, workforce participation, length of residence, family ties, or English-language ability, likely have positive relationships with both naturalizing and voting. Other characteristics, such as migrating from a democracy, should have a negative association with citizenship acquisition, but a positive relationship with voting. The posited relationships outlined in Figure 2.1 will be discussed in greater detail in subsequent chapters. Each model builds on the previous model in a cumulative manner, so as to control for lower level characteristics. In a few models, I have had to remove certain variables in order to include others. For instance, I am unable to include both country of origin dummy variables and country of origin characteristics in the same model, due to multicollinearity. These specific issues are discussed in the respective empirical chapters. Each of the four models are used to predict both citizenship and voting. I begin by examining the effect of individual-level characteristics, particularly country of origin, on the political incorporation process (see Chapter 3). Models one through four when predicting naturalizing, will be limited to Sub-sample A, as naturalizing is only a possibility for immigrants. Models one through four will also be used to predict voting. In the models predicting this second stage of political incorporation, I will rely on sub-samples B and C, the sub-samples limited to native-born and naturalized citizens, all of whom must have answered the voting question. Sub-sample B will allow for a comparison in voter turnouts between the native-born and naturalized immigrants. Using Subsample C, I will examine the likelihood of different naturalized groups voting, in comparison to one another. The results from Sub-samples B and C illustrate whether the same factors affect all of the native-born and naturalized groups in the same way, or whether the strength or the direction of a variable differs in its affects on voter turnout by immigrant group.


41

A Few Caveats When examining the foreign-born versus the native-born, it is not possible to include both age and length of voting eligibility in the same equation. For immigrants, length of voting eligibility is calculated by subtracting the year of entry from the current year. This leaves the number of years the individual has been in the country. An additional five years are subtracted from this number, as that is the residency requirement to become a citizen, and citizenship is a pre-requisite for voting. The length of voting eligibility among the native-born is calculated by subtracting 18 (the earliest age at which one can vote) from the current year. This leaves the number of years a native-born citizen has been eligible to vote. Because I am calculating length of voting eligibility among the native-born in this way, it is not possible to include both age and length of voting eligibility in the same model, due to problems of multicollinearity. Some researchers have argued that age is the more important factor (Black, et al., 1987; Converse, 1969; Niemi, et al., 1984), while others argue that length of voting eligibility is critical in predicting likelihood of voting (Bass & Casper, 1999, 2001). Therefore, I run every model on Sub-sample B in two ways, one with age and age2, and one with length of voting eligibility. When the models are run on Sub-sample C (the immigrant voting subsample), I am able to include both age and length of voting eligibility in the same model, certainly the preferable option, as it gets at both political experience in the United States and age at the same time.

FIGURE 2.1: Estimated Direction of Coefficients_____________________________________________________ Model #1 COUNTRY OF ORIGIN

V

AGE

+

AGE2

-

FEMALE

+

LESS THAN HIGH-SCHOOL (v. HS)

-

MORE THAN HIGH-SCHOOL (v. HS)

+

POVERTY

-

N A T U R A L I Z E

V + + + -

LENGTH OF EXPOSURE

+

WORK FORCE PARTICIPATION

+

+

YEAR 96, 98, 00 (v. 1994)

V

COUNTRY OF ORIGIN*INDIV. FACTORS

V

Model #2 SINGLE (V. MAR., RES. W/ SPOUSE)

-

SPOUSE AB (V. MAR., RES. W/ SPOUSE)

-

CHILDREN (V. NO CHILDREN)

+

+

MARRIED*FEMALE

?

?

+

V O

V V

-

T E

Model #3 POPULATION SIZE

?

METROPOLITAN (V. RURAL)

?

% FOREIGN BORN

?

LEVEL OF RESIDENTIAL SEG

?

Model #4 GDP

-

FHI

+

SCHOOL SEX RATIO

?

DISTANCE

+

DISTANCE*FHI

+

SCHOOL SEX RATIO*FEMALE

?

SCHOOL SEX RATIO* MARSTAT

?

GDP*POVERTY

?

GDP*EDUCATION

?

?

N A T U R A L I Z E

? ? ?

+ ? -

V O T

? ? ?

E

?

+=positive relationship -=negative relationship V=varied relationship ?=direction of relationship not posited

44


PRELIMINARY FINDINGS Table 2.6 provides some basic demographic information on Sub-sample A, immigrants who are eligible to naturalize. TABLE 2.6: Descriptive Statistics of Independent Variables in Sub-Sample A Demographic Characteristics AVERAGE AGE % FEMALE % MARRIED % MARRIED BUT APART % SINGLE

45.0 50.1% 63.7% 3.0% 33.3%

Socioeconomic Characteristics % WITH LESS THAN HS DEGREE % WITH MORE THAN HS DEGREE % WORKING FULL TIME % LIVING IN POVERTY

42.7% 34.6% 63.3% 30.3%

Contextual Characteristics % IN METROPOLITAN AREAS

93.4%

Migration Characteristics AVERAGE LENGTH OF RESIDENCE IN US (in years)

15.8

Source: November 1994, 1996, 1998, 2000 Current Population Surveys.

The sample is evenly divided between men and women. These immigrants average 45 years of age and have an average length of residency in the United States of nearly 16 years. Approximately twothirds, or 63.7% of the sample, are married and residing with their spouses, with most of the remaining migrants reporting that they are unmarried. A small percentage, 3%, report being married but living apart from their husbands or wives. A plurality of these immigrants, 42.7%, have less than a high school degree or general equivalency diploma (GED) and nearly one third, or 30.3%, of this sample lives at or below the poverty line. Although 63.3%, or nearly two-thirds of these immigrants, work full time, the low levels of education and high levels of poverty suggest they are likely working in unskilled, low-paying jobs. At the same


45

time, more than a third of these immigrants, 34.6%, have education beyond high-school, suggesting an hourglass distribution in terms of schooling. Table 2.7 presents similar descriptive statistics for Sub-sample C, immigrants who have naturalized and are eligible to vote. On average, this latter group is slightly older at 50.5 years. A slightly larger percentage of this group is female. A full two-thirds of these naturalized immigrants report being married and residing with their spouses. Slightly smaller percentages of these immigrants fall into the other two marital status categories of single, or married but living apart. TABLE 2.7: Descriptive Statistics of Independent Variables in Sub-Sample C Demographic Characteristics AVERAGE AGE % FEMALE % MARRIED % MARRIED BUT APART % SINGLE

50.5 52.6% 66.5% 2.0% 31.6%

Socioeconomic Characteristics % WITH LESS THAN HS DEGREE % WITH MORE THAN HS DEGREE % WORKING FULL TIME % LIVING IN POVERTY

29.4% 45.5% 60.5% 19.6%

Contextual Characteristics % IN METROPOLITAN AREAS

93.3%

Migration Characteristics AVERAGE LENGTH OF RESIDENCE IN US (in years) Source: November 1994, 1996, 1998, 2000 Current Population Surveys.

21.6

100%

FIGURE 2.2: Levels of Political Incorporation among Immigrants with Varying Levels of Education and Income

80% 54.7%

60%

45.5%

42.7%

40%

34.6%

29.4%

30.3%

21.8%

19.6% 12.6%

20% 0% % with Less than a HS Degree

% with More than a HS Degree

Eligible to Naturalize

Naturalized

% Living in Poverty Voted



47

In comparing Tables 2.6 and 2.7, one should note significant variation between the immigrants in Sub-sample A, those who are eligible to naturalize, and the immigrants in Sub-sample C, those who have naturalized and are eligible to vote, particularly in the areas of education, income, and length of residency in the United States. The latter group has, on average, higher levels of education, lower levels of poverty, and longer residencies in the United States. Figure 2.2 illustrates these distinctions in education and income between those who can naturalize, those who have naturalized and can vote, and those who have voted. The bar chart in Figure 2.2 reveals a bivariate relationship between socioeconomic characteristics and the level of political incorporation. Immigrants who are progressively more incorporated, via naturalizing and voting, appear to have greater amounts of education and lower levels of poverty. For example, 34.6% of all immigrants who are eligible for citizenship have more than a high-school degree, while 45.5% of those who have naturalized have greater levels of education. Among those who have voted the rate of more than a high-school education is 54.7%. The other notable distinction between the samples concerns length of residency in the United States. As Table 2.6 highlights, the citizenship-eligible sample shows an average length of residency of 15.8 years. The average length of stay among those who have naturalized and are able to vote is 21.6 years. Those who have voted have an average residency of 24.6 years. Table 2.8 examines the relationship between country of origin, rates of citizenship acquisition, and other demographic characteristics among those eligible to naturalize. Of immediate note is the huge range in rates of naturalizing. At the high end, nearly ¾ of Italians, 74.5%, have naturalized. This is followed by Filipinos, Cubans, and former Soviets, groups with rates of citizenship acquisition well over 60%. Canadians, Britons, Chinese, Southeast Asians, and Indians all follow, with naturalization rates ranging from about 50-55%. At the low end, only 22.2% of Mexican immigrants have become U.S. citizens, a dramatic drop-off from the other immigrant groups under examination in this study. Also of note is the economic and educational diversity among groups. For example, two thirds of Mexicans have less than a highschool diploma, while less than one tenth of Britons and Indians fall

48


into this category. Indians and Filipinos dominate at the higher educational levels. Eighty percent of Indians and 68.5% of Filipinos have more than a high-school degree. Educational differences translate into economic differences. Mexicans have the highest rate of poverty at 45.9%, while Indians have the lowest at 5.8%. Although the initial relationship between citizenship and country of origin gives reason enough to move from a bivariate to a multivariate analysis, these educational and economic variations make an in-depth analysis even more important to determine whether country of origin is a proxy for other characteristics. Table 2.9 examines the distribution of characteristics among all those who have naturalized and are eligible to vote. At the high end of voter turnout are Canadians, Britons, and Cubans, with 67.0%, 66.1%, and 64.4%, respectively. Italians, Indians, and Filipinos have voter turnout rates in the low 50% range. Southeast Asians and Chinese have the lowest turnout rates among all of those qualified to vote, with rates of 34.1% and 39.3%, respectively. Clearly, voter turnout varies by country of origin. Further, the relationship between country of origin and political incorporation appears to fluctuate by type of political activity, given the significantly different patterns identified in Tables 2.8 and 2.9. Certain immigrant groups with rates of citizenship acquisition on the higher end of the spectrum, such as former Soviets and Chinese, have voter turnout rates on the lower end. Table 2.9 also shows variation by country of origin on certain demographic characteristics. Educational attainment is wide-ranging, with 21.0% of Mexicans having more than a high-school degree versus 86.9% of Indians. The average length of eligibility also varies significantly, reflecting diverse migration histories. Canadians have the longest average length of political eligibility at 33.3 years versus Southeast Asians who have the shortest average length of political eligibility at 11.8 years. These are the same two groups that define the upper and lower bounds of voting rates, suggesting a correlation between voter turnout and length of eligibility.

TABLE 2.8: Summary Statistics for Those Eligible to Naturalize, by Immigrant Group

Immigrant Group

N in Citizenship Models

% Naturalized

% Less HS Degree

% More HS Degree

% Below Poverty

Mean Eligibility (in years)

MEXICANS

7,373

22.2%

66.4%

13.3%

45.9%

12.8

CUBANS

1,157

63.6%

30.6%

37.5%

18.8%

21.0

CANADIANS

1,060

55.4%

17.7%

55.9%

7.9%

27.1

BRITONS

854

54.1%

9.0%

56.7%

9.4%

25.0

ITALIANS

967

74.5%

47.7%

21.0%

11.1%

30.0

FORMER SOVIETS

638

61.6%

17.2%

54.1%

26.5%

15.2

1,768

69.0%

13.4%

68.5%

8.1%

13.4

900

50.4%

9.8%

80.0%

5.8%

11.3

SE ASIANS

1,255

51.9%

31.5%

39.2%

26.0%

9.9

CHINESE

1,047

50.8%

29.0%

49.5%

16.5%

12.3

FILIPINOS INDIANS


TABLE 2.9: Summary Statistics for Those Eligible to Vote, by Immigrant Group

Immigrant Group

N in Voting Models

% Voted

% Less HS Degree

% More HS Degree

% Below Poverty

Mean Eligibility (in years)

1,492

40.0%

55.4%

21.0%

34.7%

19.6

CUBANS

693

64.4%

24.2%

44.4%

16.3%

24.4

CANADIANS

551

67.0%

19.4%

52.6%

8.7%

33.3

BRITONS

430

66.1%

9.1%

50.7%

10.0%

31.7

ITALIANS

622

56.4%

46.8%

23.6%

10.0%

32.8

FORMER SOVIETS

348

48.6%

16.4%

56.6%

25.9%

21.4

CHINESE

463

39.3%

23.5%

52.5%

13.8%

17.8

SE ASIANS

560

34.1%

18.0%

53.6%

15.9%

11.8

FILIPINOS

1,078

53.0%

11.3%

71.2%

7.0%

15.8

404

54.2%

4.5%

86.9%

4.7%

15.3

MEXICANS

INDIANS



51

CONCLUSION This preliminary examination of the data suggests variability in political incorporation by country of origin, education, income, and length of eligibility. However, some important differences emerge in the relationships of these characteristics with naturalizing and voting, respectively. Figure 2.2 illustrated increasing rates of education and decreasing rates of poverty with rising levels of political incorporation. Those eligible to naturalize show lower levels of education and income, on average, than those who have naturalized. Those eligible to vote, i.e. the naturalized, show lower levels of education and income than those who have voted. In short, political incorporation is not random, but is selectively sought out by those with greater schooling and/or economic means. Further breakdown of the samples by country of origin reveals a more complex picture. While education and income appear to be correlated with both forms of political incorporation, the relationship is not perfectly consistent within or across immigrant groups. Britons and Canadians who have naturalized exhibit lower levels of income and education, on average, than among the larger sample of their eligible (but not necessarily naturalized) compatriots. Interestingly, naturalized Britons and Canadians have higher levels of voter turnout than any other eligible group. This preliminary finding suggests that “key” predictors of incorporation may work differently for different groups. Finally, this initial analysis of naturalizing and voting by immigrant group suggests differential motivations for these two political processes. Immigrant groups with high levels of citizenship acquisition do not necessarily have high levels of voter turnout. Conversely, low levels of naturalizing do not necessarily mean low levels of voting. This finding suggests that voting is not necessarily a natural outgrowth of naturalizing. These two processes may be quite distinct, with varied motivations. The prima facia evidence of a relationship between country of origin and political incorporation, the variation in other types of predictors by country of origin, and the seemingly distinct nature of naturalizing and voting, respectively, require a more in-depth analysis. The next four chapters explore these initial findings further to see how individual, family, community, and country of origin factors more fully explain these two formal modes of political incorporation, or lack thereof, among immigrants in the United States.


CHAPTER 3

The Effect of Individual Characteristics on Political Incorporation The differing benefits attached to naturalizing and voting, respectively, beg the question of whether these two formal acts of political incorporation are attractive to the same groups of people for the same reasons. Naturalizing may be undertaken for political reasons, but there are countless economic, legal, and social incentives to naturalize, as well. Voting, although frequently viewed as the natural outgrowth of citizenship, provides far fewer immediate and tangible benefits, with its primary benefit being participation in the polis. Thus, voting suggests to an even greater extent than does citizenship the individual incorporation of a democratic society’s political values. Citizens, both foreign-born and native-born equally, voice their political concerns, affect the outcome of elections, and influence public policy through voting, with little individual benefit. Perhaps an individual will help influence the outcome of an election. Perhaps his or her candidate will gain political office. However, the individual benefits that an immigrant will accrue from voting pale in comparison to what citizenship provides. As a result of these differential benefits, one should not be surprised that more immigrants naturalize than vote. Of greater interest is that both citizenship and voting, respectively, are differentially distributed throughout the foreign-born population. Not every immigrant is as likely to naturalize as every other immigrant, nor is every eligible citizen as likely to vote as every other. Previous studies have repeatedly found that individual-level characteristics, such as education, income, country of origin, and length of exposure to the United States, significantly affect the odds of various types of immigrant incorporation. But how do these individual-level characteristics impact political integration? Further, are the sources of variation similar in magnitude, or even direction, for both naturalizing and voting? 53

54


Individual-level characteristics, specifically education, income, and length of exposure to the United States, are likely linked to political incorporation through the mechanisms of knowledge and contact. As immigrants come to possess higher levels of education, income, or experience in the United States, their knowledge of the benefits of naturalizing likely increases. Greater education and experience should also make the process of political incorporation easier, with the sometimes overwhelming litany of tests and procedures required for citizenship made easier for those with more knowledge of the system. These same factors should influence voting, as well, with higher levels of education, income, or exposure increasing familiarity with the political system, thus encouraging and facilitating voter turnout. Greater economic, educational, and social incorporation should also encourage immigrant political incorporation through increased contact with the native-born. Immigrants with higher levels of education, income, or length of exposure to the United States are more likely to interact with native-born Americans in professional, educational, residential, and social settings. As immigrants become more entrenched in mainstream society, they should be more likely to view their migration as relatively more permanent, encouraging both naturalizing and voting. Just as individual immigrants enter the United States with different amounts of human capital, so too do they enter with diverse political histories, cultural experiences, and reasons for migrating. Therefore it is not surprising to find that country of origin affects political incorporation, with some immigrant groups, such as Filipinos or Chinese, naturalizing at higher rates than other groups, such as Mexicans or Britons (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996; Smith & Edmonston, 1997). In previous research, country of origin variables have been included in models as summary measures of home society characteristics (Portes & Rumbaut, 1996). A country of origin variable might stand for the ease of reverse migration, i.e., how costly or difficult it is to return to one’s country of origin. An individual migrating from a geographically proximate democracy would likely have an easier time returning to his or her home society than an individual migrating from a greater distance or from a more hostile political regime. The ease with which one can reverse his or her

The Effect of Individual Characteristics on Political Incorporation 55 migratory course and return home is known as the level of “reversibility” (Portes & Rumbaut, 1996). An immigrant’s previous political experiences may also influence his or her likelihood of participating in the electoral process. Individuals with previous democratic experience may be able to “translate” this political knowledge and apply it to new political experiences in the United States. The ability to apply prior political knowledge to a new political environment is known as “translation” (Black et al., 1987; Finifter & Finifter, 1989). Greater distances or more hostile regimes may mean greater propensities to naturalize, while previous democratic experience likely translates into a higher likelihood of voter turnout, once naturalized. Both the reversibility and translation theories suggest that country of origin really constitute a proxy for something else, such as geographic distance between countries, previous political experience, or linguistic congruity, with the effect of country of origin likely varying by political process. However, apart from these factors, country of origin likely has its own influence on political incorporation. Immigrants encounter very different receptions upon entering the United States, based largely upon their country of origin (Guranizo, Portes, & Haller, 2003; Koopmans, 1999; Portes, 1995b). Immigrants from certain countries of origin, such as Cuba, are more warmly received and encounter greater help navigating through the naturalization process than immigrants from other countries, such as Haiti, who meet greater discrimination (Alvarez, 1987; Garcia, 1987; Pedraza-Bailey, 1987). An immigrant’s country of origin also provides access to ethnic social networks and organizations; the best example is that of Cuban immigrants who have access to an extensive and highly entrenched set of formal and informal groups (Fennema & Tillie, 1999; Forment, 1989; Togeby, 1999). In this example, it is likely the country of origin as such that affects political incorporation, rather than some other characteristic subsumed in the country of origin variable. Immigrants from different countries of origin, even those sharing similar geographic, economic, or political characteristics, enter into different environments in the United States, which in turn influence political integration. Country of origin, while likely important in its own right, may also mediate the effect of other individual-level characteristics, such as

56


income and education, on political incorporation. The selectivity of migration flows means that countries tend to send a particular type of immigrant. Immigrants from some countries, such as India, tend to come from the highest echelons of their respective societies. In the case of other countries, such as Mexico, the lower economic and educational tiers of the society dominate immigration to the United States. Furthermore, the reasons for migrating to the U.S. and the longer-term settlement plans may reflect not just an individual’s personal situation, but also the larger country of origin context. A Canadian immigrant with a high-level of education may be substantially less likely to naturalize than a Soviet immigrant with a comparable level of education, due to differences in the home society. The two personal characteristics of education and country of origin likely interact to result in different propensities to become politically incorporated. Does an individual migrant with a high level of education have professional opportunities in his or her country of origin? Is the goal of migration to the United States part of a larger plan to raise capital and return to opportunities in the home society, or are job opportunities for the highly educated so limited in the country of origin that the plan is for more permanent settlement in the United States? As a result, the political incorporation of immigrants with shared economic or educational characteristics, but from different countries of origin, might differ dramatically, as individual characteristics interact with country of origin to influence political behavior. It is essential to examine both the main effects of country of origin, as well as its interactions with other factors, to understand fully the interplay of all of these influences on political incorporation. The goals of this chapter are (1) to create a baseline of the individual level characteristics that predict political incorporation among the foreign-born, looking closely at the influences of education, income, exposure, and country of origin; (2) to determine whether the two processes of naturalizing and voting are predicted by the same or different individual-traits; (3) to examine how individual-characteristics might interact with each other to affect political incorporation; and (4) to compare the foreign-born with the native-born to explore the extent to which individual-level characteristics predict voter turnout for both groups.

The Effect of Individual Characteristics on Political Incorporation 57 CHANGING POLICIES, CHANGING PATTERNS The passage of the 1965 Immigration Act led to a dramatic change in the composition of migration flows to the United States. Whereas immigration was dominated by Europeans during the first half of the century, the change in legislation has since led to immigrant streams comprised primarily of Asians and Latinos. Such a dramatic shift in country of origin has highlighted the need to answer the question of whether today’s immigrants assimilate into American society as their European predecessors did. At the foundation of this debate is the question of how and to what extent individual-level characteristics determine incorporation. Straight-line assimilation theorists (Gordon, 1964) argue that with the passage of enough time, immigrants eventually become indistinguishable from the native-born. Through language-acquisition, residential integration, and intermarriage, the two groups meld into one, with time being the major difference in the level of incorporation of different groups. In this theory, individual traits, other than length of exposure to American society, have less to do with incorporation. More recent arguments suggest that time alone will not erase the differences between the native-born and the foreign-born (Alba & Nee, 1999; Portes, 1995a). Contemporary immigrants are simply too different from the native stock and no amount of time will make up for these differences. Today’s migrants are of “lesser quality” than previous waves (Borjas, 1998). Their lower levels of education result in lower levels of income, and the development of a two-tiered society (Frenette & Morissette, 2005). The questions and goals outlined in the introduction of this chapter– what predicts political incorporation and how does it vary by process or group? – are fundamentally part of this larger debate about incorporation. To begin to answer these questions, it is essential to understand what characteristics have been identified as crucial components of incorporation and how such characteristics encourage or discourage political integration. PREDICTORS OF POLITICAL INCORPORATION Socio-Economic Factors Historically, models developed to predict citizenship have been of the socio-economic variety, exploring income, education, and occupational status to examine their impacts on the naturalization process. Though a

58


considered and conscious cost-benefit analysis is unlikely to be performed by individuals in either the naturalizing or voting realm, given the many unquantifiable emotional factors involved, consideration of some of the costs and benefits of participation is justified. Studies of citizenship acquisition have specifically focused on the influence of income and education, repeatedly finding that increased income and education lead to higher odds of naturalizing (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996; Smith & Edmonston, 1997; Yang, 1994). One study finds that approximately 66% of immigrants with college degrees, household incomes above $100,000, and white-collar jobs have naturalized. At the other end of the educational and economic spectrum, the naturalization rate of those who have not completed high school stands at 44% and the naturalization rate of those in households earning less than $15,000 annually is 52% (Mogelonsky, 1997). The posited mechanisms behind these relationships vary. Income and education may encourage naturalizing by increasing the benefits, or perceived benefits, of U.S. citizenship, with the better educated more fully aware of the social, political, and economic advantages of naturalizing (Yang, 1994). Greater levels of income or schooling may also encourage naturalizing by easing the daunting citizenship process. The more educated are better able to navigate through the complicated naturalization process that includes tests of civics, history, and Englishlanguage (Portes & Rumbaut, 1996). Those with more income can better afford a lawyer to guide them through the maze of laws, tests, and procedures associated with naturalizing. As with naturalizing, resource or socio-economic based theories dominate the literature on voting, and identify education and income as the biggest predictors of turnout (Bass & Casper, 2001; de la Garza & DeSipio, 1992; Rosenstone & Hanson, 1993; Sierra et al., 2000; Verba, Schlozman, & Brady, 1995). More educated individuals are more likely to participate in politics than those with less education, likely due to a greater awareness of and interest in the political system and stronger feelings of political efficacy. Higher levels of income translate into individuals feeling that they have a greater stake in society and more to protect. Individuals compare the time and effort of voting with what may be won or lost by going to the polls (Verba et al., 1995).

The Effect of Individual Characteristics on Political Incorporation 59 Studies of voter turnout among the native-born suggest that education and income work in the same way for those born in the United States as they do for the foreign born (Bass & Casper, 1999, 2001; Berelson, Lazarsfeld, & McPhee, 1954; Campbell, Converse, Miller & Stokes, 1960; Rosenstone & Hansen, 1993; Tate, 1991; Verba & Nie, 1972; Verba et al., 1995; Wolfinger & Rosenstone, 1980). As education and income rise, so does the propensity of the native-born to vote. The mechanisms- interest, efficacy, and resources- are thought to be similar for both groups. Increasing levels of income and education also likely encourage immigrant political incorporation through greater contact with mainstream society. Immigrants with lower levels of skill, income, or education are more likely to be segregated into jobs, social circles, and neighborhoods that are dominated by other immigrants. In contrast, immigrants who have higher levels of income or education, such as those from India, may find themselves working with or living near the native-born. This contact likely encourages immigrant political incorporation, with the native-born acting as role models, applying peer pressure, or simply helping to further entrench immigrants into American society. Exposure Exposure, understood as the length of time an immigrant lives in the United States and is exposed to American society, is one of the most consistent predictors of immigrant incorporation (Gordon, 1964). Longer residencies expose individuals to the language, habits, culture, people, and institutions found in the United States. This greater contact with both people and institutions provides immigrants with more “local” knowledge and increasingly strong ties to American society. As an immigrant’s exposure increases, he or she becomes more entrenched in mainstream culture, with social, economic, educational, and linguistic incorporation all increasing (Lopez, 1999). The propensity of immigrants to become politically incorporated also increases with extended stays, reducing many of the educational, economic, and country of origin differences (Bass & Casper, 1999, 2001). More time in the United States translates into greater experience with American institutions, increased knowledge about the potential benefits of naturalizing and how to do so, greater contacts with the native-born, a more permanent view of migration, and higher

60


levels of social capital (Liang, 1994). These factors, in turn, encourage citizenship acquisition. In the voting realm, increased time in the United States may translate into greater internalization of American political norms, more familiarity with the American democratic system, and greater feelings of peer pressure to become “American” and participate in the political life of the community (Bass & Casper, 1999, 2001; Gordon, 1964). Age The impact of certain demographic characteristics on the naturalization and voting processes has also been explored. An individual’s age has repeatedly appeared as a significant predictor of voting for the native-born and, for naturalizing and voting for the foreign-born. In the naturalization process, the relationship with age is curvilinear. Immigrants who are younger when they arrive are less likely to naturalize than are their more senior counterparts. Progressively older immigrants are ever more likely to acquire citizenship until approximately middle age. At that point, the relationship changes direction, with each passing year decreasing the likelihood of naturalizing (Yang, 1994). A similar trend has been identified in the voting realm. Younger people, both native-born and foreign-born, are less likely to be politically active than are their older counterparts. Age increases interest and involvement. This positive trend continues on until individuals reach old age. At the oldest ages, involvement drops off, with this pattern identified among both native-born (Converse, 1969; Niemi et al., 1984; Rosenstone & Hanson, 1993; Verba et al., 1995) and foreign-born (Black et al., 1987). There are multiple mechanisms mediating between age and political incorporation. In the naturalization process, younger migrants may not yet have decided whether they will remain in the United States. As they get older, they view their host country as a more permanent home. Furthermore, younger migrants may not see the benefits of naturalizing, and only with increased age do the social, economic, and political benefits of citizenship fully become apparent. At too advanced an age, citizenship becomes less appealing. Older immigrants may have a harder time emotionally giving up formal ties to their country of origin to acquire American citizenship. From a

The Effect of Individual Characteristics on Political Incorporation 61 practical perspective, language acquisition becomes more difficult for people as they age. Since English-language ability is a component of the citizenship process, elderly immigrants may feel they are unable to pass the naturalization exams. Finally, elderly immigrants may simply conduct the same cost-benefit analysis that their younger counterparts conduct. Because they are older, the length of time they will benefit from U.S. citizenship is shorter, making the somewhat difficult and time-consuming process seem less appealing. Though the relationship between age and voting is similar to that of age and citizenship, the mechanisms at work are different. Younger people, both native-born and foreign-born, are simply less engaged in politics (Converse, 1969; Niemi et al., 1984; Rosenstone & Hanson, 1993; Verba et al., 1995). Increased age boosts interest in the political system in a variety of ways. Increased age may translate into greater wealth and a larger stake in society. In an effort to protect what one has earned, individuals are more likely to go to the polls. Older people may also have more free time to become involved in politics, or simply develop greater interest. Among the elderly, the time, interest, and desire to be involved in politics may still exist. Unfortunately, older people have a harder time getting to the polls to vote. Sickness, isolation, and immobility make voting a more difficult process for both the foreign-born and the nativeborn, leading to a negative relationship between turnout and age at the upper reaches of the age spectrum. Age may also stand as a proxy for where an individual is in the life-course. As one advances through young adulthood into the twenties and thirties, they are more likely to be married and have children. As will be discussed in Chapter 4, these familial relationships may encourage both the foreign-born and the native-born to become politically engaged through a variety of mechanisms, ranging from greater stability resulting from marriage to greater interest in the distribution of resources resulting from having children. Gender The gender gap in political participation has existed for years, with women frequently participating at lower rates than men (Burns, Lehman Schlozman & Verba, 2001; Milbraith & Goel, 1982; Wolfinger & Rosenstone, 1980). Unfortunately, much of the knowledge about the political gender gap is based upon studies of the

62


white native-born (Lien, 1998). Knowing that white men are more likely to vote than white women or that native-born women are more likely to be Democrats than native-born men does little to inform our knowledge of immigrant political behaviors, given the very different ways gender manifests itself across cultures even within the United States (Lien, 1998; Tate, 1991). One study finds women are less likely to participate politically than are men, due to their lower rates of participation in the workforce, an arena thought to encourage political discussion and mobilization (Burns et al., 2001). In a study focusing specifically on black Americans, women are found to vote more than their male counterparts, arguably resulting from women’s greater involvement in the black churches (Tate, 1991). Case studies of Latino immigrant incorporation find a greater propensity among female migrants to become integrated into American society (Hongdagneu-Sotelo, 1994; Jones-Correa, 1998). Women tend to seek permanent settlement in the host society, while men continue to look back towards the country of origin. Studies that have explicitly examined political behaviors have found mixed results. While some studies find women to be more likely to naturalize (Yang, 1994) and vote (Bass & Casper, 1999, 2001), another study finds lower naturalization rates among women (Jasso & Rosenzweig, 1990). Clearly, the role of gender varies from context to context and must be understood and interpreted for each group. Country of Origin Even after factors such as income, education, length of residence, and age have been taken into account, both naturalizing and voting vary by country of origin (Alba & Nee, 1999; de la Garza & DeSipio, 1996; Portes & Rumbaut, 1996; Sierra et al., 2000; Smith & Edmonston, 1997). One author argues “national origin is the overriding predictor of the tendency to naturalize, however, above and beyond socioeconomic characteristics. A Mexican college graduate is less likely than a highschool dropout from China to become a U.S. citizen…” (Mogelonsky, 1997, p.47). The numbers may support this argument, but we are left wondering what it is about country of origin that leads to such differences. Surely some immigrant groups are not innately more likely to naturalize than other immigrant groups.

The Effect of Individual Characteristics on Political Incorporation 63 Country of origin likely acts as a proxy for a host of other characteristics that define and distinguish that particular country of origin, such as geographic distance, political system, economic opportunities, or gender structure. Previous research examining the effect of country of origin on political incorporation has identified the cost/benefit structure as one of the mechanisms mediating the relationship (Portes & Rumbaut, 1996). Individual immigrants assess the costs and benefits of acquiring U.S. citizenship within their country of origin context. Although applying a formal cost-benefit analysis to something as emotionally driven as naturalizing or voting is questionable, using the general framework to think of both the positives and negatives for an individual immigrant can be useful. How difficult, expensive, or painful is it to return to the country of origin? Is reversibility dangerous or prohibitive in some other way? Immigrants migrating from greater distances, such as the Philippines, or from nondemocratic regimes, such as China, are far more likely to naturalize than are those from geographically closer or more democratic countries, such as Canada, where the financial or psychic costs of return are lower (Jasso & Rosenzweig, 1990; Liang, 1994; Portes & Rumbaut, 1996). However, country of origin likely influences the voting process differently than it does the naturalization process. The process of voting generally provides little tangible, immediate, or individual benefit. Individuals vote because they gain psychic satisfaction from “doing their civic duty.” Some scholars argue that this ingrained sense of civic responsibility results from having grown up in a democratic society (Black et al., 1987). According to the “translation” argument (Finifter & Finifter, 1989), previous experience with democratic systems is among the best predictors of future participation for immigrants. Other scholars make a slightly different argument regarding the influence of country of origin on voting, contending that refugees are less likely to participate politically, not because of a lack of experience, but rather due to their histories of state-sponsored oppression and lack of trust in government institutions (Fennema & Tillie, 1999; Harles, 1993; Portes & Rumbaut, 1996). In these instances, country of origin is influencing individual political incorporation, but through different mechanisms than those that impacted naturalizing. More recently, the effect of country of origin on voter participation has been viewed as a mechanism of mobilization. Ethnic social

64


networks and organizations encourage political participation by turning the powerlessness of one vote into the power of many (Barreto, 2005; Fennema & Tillie, 1999; Shaw et al., 2000; Togeby, 1999). An isolated individual may vote for psychic satisfaction, out of a sense of duty, or simply out of habit. An individual who is part of a larger ethnic network may turnout to vote in an effort to push a particular agenda or elect a certain official. As part of a mobilized group, an individual likely feels a sense of duty to others in the network, as well as the real possibility of electoral success. Being from a particular country of origin is what provides access to and involvement in such ethnic networks and organizations, and the resulting political mobilization (Gibson & Ogbu, 1991; Portes, 1995b; Portes & Zhou, 1999). Successful ethnic mobilization is best seen in the Cuban community, where extensive social networks and ethnic organizations have led to high levels of voter turnout (Forment, 1989; PedrazaBailey, 1987; Portes & Bach, 1985). Country of origin also influences political incorporation indirectly by mediating the effect of other characteristics. In research conducted by Liang (1994), predictors of political incorporation vary by immigrant group. For instance, homeownership is a bigger predictor of naturalizing among Chinese, Koreans, and Cubans than among Mexicans, while residential integration matters the most for naturalizing among the latter group. In this instance, having an economic “stake in society” seems to encourage political incorporation for certain groups, while having social ties seems to matter more for other groups. These findings suggest that mechanisms vary by country of origin, making source country important to study as much for the influence it exerts indirectly as for its main effects. Gender is another characteristic that is affected by country of origin and cannot simply be studied in a vacuum (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996). Irish women are more likely to acquire U.S. citizenship than are Irish men, while among Israelis, men are more likely to naturalize than are women. Though scholars have not determined the mechanisms at work, the mixed gender findings by country of origin suggests that the impact of gender on political incorporation varies by home society. Even characteristics found to affect political incorporation in the same way across immigrant groups must be examined more closely.

The Effect of Individual Characteristics on Political Incorporation 65 For instance, greater levels of education translate into greater odds of naturalizing, but the importance varies by country of origin (Liang, 1994; Yang, 1994). A college-educated Mexican immigrant is more likely to naturalize than a compatriot with lesser education, but a Chinese immigrant with the lowest level of education has the highest odds of naturalizing (Mogelonsky, 1997). Little work has been done to understand how different characteristics interact with country of origin to affect voting. Two studies that have begun to explore these relationships have focused on how gender varies in its effect across different immigrant groups. Female immigrants from Latin America seem to have higher propensities to become politically active (Jones-Correa, 1998) and vote (Bass & Casper, 1999, 2001), as compared with their male counterparts. Scholars argue that women from Latin American countries experience new freedoms and opportunities in the United States, while Latin American men find economic and social adjustment difficult (Grasmuck & Pessar, 1991). The reaction of the women is to become more politically incorporated, flexing their new power. Men from the same countries react by retreating into ethnic groups and organizations that reflect the more traditional patriarchal structures (Jones-Correa, 1998). Because so little work has been done to explore how gender differentially affects political incorporation across immigrant groups, it is not known whether this gender pattern exists across all migrant communities. What we do know is that gender roles vary significantly by place of origin. Such diversity in gender structures suggests that it likely does impact the political incorporation of different immigrant groups in a range of ways. HYPOTHESES Socio-economic factors have consistently been identified as some of the biggest predictors of incorporation (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996; Smith & Edmonston, 1997; Yang, 1994). An individual’s level of education and income impact the type of job one holds, with whom one comes into contact, and where one lives. As education and income increase, they encourage these multiple types of social and structural integration which in turn encourage political incorporation through increased contact with the native-born, stronger ties to American society, and greater knowledge about the benefits and processes of naturalizing and voting. This leads to the first hypothesis.

66

From Immigrant to Naturalized Citizen The Socio-Economic Hypothesis: Immigrants with greater levels of education and income should show higher rates of naturalizing and voting, net of other characteristics. However, the power of these predictors will vary by country of origin.

Previous research has identified ease or difficulty of reversibility as a factor affecting an individual’s decision to naturalize (Liang, 1994; Portes & Rumbaut, 1996). Immigrants from greater distances, more hostile political regimes, or generally from countries where return is more difficult are more likely to acquire American citizenship than immigrants from societies with fewer barriers to return. The second hypothesis is intended to test this argument, examining immigrants from ten places of origin, each with distinct levels of economic, political, and geographic barriers. The Reversibility Hypothesis: Because of the geographic, economic, and/or political difficulties associated with reverse migration, immigrants from China, the former Soviet Union, Cuba, Southeast Asia, the Philippines, and India should show higher propensities to naturalize than immigrants from Mexico, Canada, Great Britain, and Italy. Other scholars have argued that similarities between the home country and the host country lead to greater levels of political incorporation (Black et al., 1987; Finifter & Finifter, 1989). Those that have experienced democratic elections in his or her home society will be more likely to participate in electoral politics in the United States, as previous political experience can be translated. Though earlier work (Black et al., 1987; Finifter & Finifter, 1989) has identified these trends, this hypothesis has not previously been tested on voter participation in the United States, leading to the third hypothesis. The Translation Hypothesis: The political environment of the country of origin impacts an immigrant’s level of political integration in the United States. Immigrants from countries with histories of democracy, specifically Great Britain, Italy, Canada, and India, should be more likely to vote, once naturalized, than immigrants from the former Soviet Union,

The Effect of Individual Characteristics on Political Incorporation 67 Southeast Asia, and China, societies that lack a democratic tradition. Immigrants from Mexico and the Philippines are likely found somewhere between the extremes, given their more mixed democratic histories. An immigrant group’s initial reception in the United States has implications for its longer-term incorporation (Portes, 1995b). Immigrants that are welcomed and targeted to receive financial aid and assistance will have a greater likelihood of integration, as has happened in the Cuban community (Pedraza-Bailey, 1987). Settlement patterns also play a role, with concentrations of immigrants both helping to integrate new waves from the home society into the political and economic systems in the United States (Forment, 1989), as well as to increase the interest of major political parties who identify a voting bloc worthy of courtship. These multiple factors, characteristic of the Cuban community, lead to the next hypothesis. The Mobilization Hypothesis: Cubans, though emigrating from a non-democratic society, will have high levels of voter turnout due to extensive mobilization efforts. This voter push emanates both from the ethnic social networks and organizations within the highly segregated Cuban community and externally, from the major political parties. Though an increasing amount of attention has recently focused on the gendered nature of migration, much of the work has focused on one or two immigrant groups (Bass & Casper, 1999, 2001; Grasmuck & Pessar, 1991; Jones-Correa, 1998). This hypothesis is applied to test the effect of gender across ten distinct countries of origin, building on the growing body of research on gender and migration. The Gender Diversity Hypothesis: The effect of gender on political incorporation should vary by country of origin in both the naturalization and voting processes. An immigrant’s length of exposure to the United States, net of other characteristics, should play a significant role in encouraging immigrant political incorporation. Immigrants, regardless of education, income, country of origin, or any of the other characteristics previously

68


discussed, will likely become more “American” with the passage of time, as Gordon (1964) suggested. The Straight-Line Assimilation Hypothesis: As length of time in the United States increases, immigrants should be more likely to become politically incorporated. Further, the differences between native-born and foreign-born voter turnout should decrease in significance when length of eligibility is included in the model. FINDINGS The Pooled Models Tables 3.1 and 3.2 give the odds-ratios of naturalizing and voting, respectively, among individuals with certain socio-economic and demographic characteristics. Taken together, the tables show a strong, positive relationship between socio-economic predictors, specifically income and education, and political incorporation. For instance, immigrants with less than a high-school degree are only about half as likely to naturalize as those with a high-school degree. Those with at least some higher education are 38% more likely to naturalize than someone who holds only a highschool degree. In terms of voting, those in the lowest educational category are about a third less likely to vote as those immigrants holding highschool diplomas, while those with higher levels of education are nearly twice as likely to vote as those in the reference group. Income, similarly, is a strong predictor of both naturalizing and voting, with those living below the poverty line only 78% as likely to naturalize as those living above the poverty line. Poverty has a similarly negative relationship with voting, with poorer immigrants only 76% as likely to vote as their wealthier counterparts. These findings suggest that socio-economic variables are similar in direction for both forms of political incorporation, in keeping with previous studies (Jasso & Rosenzweig, 1990; Portes & Rumbaut, 1996; Smith & Edmonston, 1997). The results in Tables 3.1 and 3.2 also lend strong support for the Socio-economic Hypothesis; as education and income increase, so does the propensity for political incorporation.

TABLE 3.1: The Odds of Citizenship Acquisition Among Immigrants, N=17,019 CONTROLS

NATIVITY

CUBA

MEXICO

CANADA

SE ASIA

INDIA

CHINA

PHIL

USSR

ITALY

BRIT

AGE 2 AGE

1.01

**

CUBA

NA

NA

2.91

***

3.43

***

0.62

***

1.04

0.83

(a)

0.47

***

0.60

***

1.11

1.00

(a)

MEXICO

0.34 ***

NA

NA

1.18

(a)

0.21

***

0.36

***

0.29

***

0.16

***

0.21

***

0.38

***

1.07

FEMALE

1.00

CANADA

0.29 ***

0.85

(a)

NA

NA

0.18

***

0.30

***

0.24

***

0.14

***

0.18

***

0.32

***

0.91

LESSHS

0.53 *** SE ASIA

1.61 ***

4.67

***

5.51

***

NA

NA

1.67

***

1.33

**

0.75

***

0.97

1.78

***

MOREHS

1.38 *** INDIA

0.96

2.80

***

3.30

***

0.60

***

NA

NA

0.78

*

0.45

***

0.58

***

1.07

POVERTY

0.78 *** CHINA

1.21

(a)

3.51

***

4.14

***

0.75

**

1.25

*

NA

NA

0.56

***

0.73

**

1.34

WORK

1.18 *** PHIL

2.14 ***

6.24

***

7.35

***

1.33

***

2.23

***

1.78

***

NA

NA

1.29

*

ELIGTIME

1.10 *** USSR

1.66 ***

4.84

***

5.70

***

1.04

1.73

***

1.37

**

0.78

*

NA

NA

YEAR96

1.15

0.90

2.63

***

3.10

***

0.56

***

0.94

0.75

*

0.42

***

0.54

***

YEAR98

1.43 *** BRIT

0.32 ***

0.94

0.20

***

0.34

0.27

***

0.15

***

0.19

***

YEAR00

1.37 ***

-2 LOG LIKELIHOOD PSEUDO R2

**

ITALY

17,048.76 .31

MAX. RESCALED PSEUDO R

2

.41

Numbers reported are odds-ratios. Source: November 1994, 1996, 1998, 2000 Current Population Surveys. ***P

From Immigrant to Naturalized Citizen: Political Incorporation in the United States (The New Americans: Recent Immigration and American Society)

Social Determinants of Immigrant Selection: The United States, Canada, And Australia (The New Americans: Recent Immigration and American Society)

Health and Behavior among Immigrant Youth (The New Americans: Recent Immigration and American Society)

Mexican Immigrant Parents Advocating School Reform (The New Americans: Recent Immigration and American Society)

Nativism and Immigration: Regulating the American Dream (New Americans: Recent Immigration and American Society)

Immigration, Acculturation, And Health: The Mexican Diaspora (The New Americans: Recent Immigration and American Society)

Iranian Immigrants in Los Angeles: The Role of Networks and Economic Integration (The New Americans: Recent Immigration and American Society) (The New ... Recent Immigration and American Society)

Unequal Origins: Immigrant Selection and the Education of the Second Generation (The New Americans: Recent Immigration and American Society)

Mental Health Among Taiwanese Americans: Gender, Immigration, And Transnational Struggles (The New Americans: Recent Immigration and American Society)

The Politics of Citizenship of Mexican Migrants (The New Americans: Recent Immigration and American Society)

Ethnic Americans: Immigration and American Society

Between Two Generations: Language Maintenance and Acculturation Among Chinese Immigrant Families (The New Americans: Recent Immigration and American Society)

German Immigrants (Immigration to the United States)

Jewish Immigrants (Immigration to the United States)

Italian Immigrants (Immigration to the United States)

Irish Immigrants (Immigration to the United States)

Chinese Immigrants (Immigration to the United States)

Mexican Immigration to the United States

Mexican Immigrants (Immigration to the United States)

Jamaican Immigrants in the United States and Canada: Race, Transnationalism, and Social Capital (The New Americans: Recent Immigration and American Society)

The History of Ethiopian Immigrants and Refugees in America, 1900-2000 (The New Americans: Recent Immigration and American Society)

The American Dream--For Men Only?: Gender, Immigration, and the Assimilation of Israelis in the United States (New Americans)

Americans in Waiting: The Lost Story of Immigration and Citizenship in the United States

Africans In America (Immigration to the United States)

Russian Immigrants in the United States: Adapting to American Culture (New Americans (Lfb Scholarly Publishing Llc).)

Skilled Immigrant and Native Workers in the United States: The Economic Competition Debate and Beyond (The New Americans) (Criminal Justice)

Mexican Immigrants in the Labor Market: The Strength of Strong Ties (The New Americans: Recent Immigration and American Society) (New Americans (Lfb Scholarly Publishing Llc).)

Intermarriage Across Race and Ethnicity Among Immigrants: E Pluribus Unions (The New Americans: Recent Immigration and American Society)

From Immigrant to Naturalized Citizen: Political Incorporation in the United States (The New Americans: Recent Immigration and American Society)

Social Determinants of Immigrant Selection: The United States, Canada, And Australia (The New Americans: Recent Immigration and American Society)

Health and Behavior among Immigrant Youth (The New Americans: Recent Immigration and American Society)

Mexican Immigrant Parents Advocating School Reform (The New Americans: Recent Immigration and American Society)

Nativism and Immigration: Regulating the American Dream (New Americans: Recent Immigration and American Society)

Immigration, Acculturation, And Health: The Mexican Diaspora (The New Americans: Recent Immigration and American Society)

Iranian Immigrants in Los Angeles: The Role of Networks and Economic Integration (The New Americans: Recent Immigration and American Society) (The New ... Recent Immigration and American Society)

Unequal Origins: Immigrant Selection and the Education of the Second Generation (The New Americans: Recent Immigration and American Society)

Mental Health Among Taiwanese Americans: Gender, Immigration, And Transnational Struggles (The New Americans: Recent Immigration and American Society)

The Politics of Citizenship of Mexican Migrants (The New Americans: Recent Immigration and American Society)

Ethnic Americans: Immigration and American Society

Between Two Generations: Language Maintenance and Acculturation Among Chinese Immigrant Families (The New Americans: Recent Immigration and American Society)

German Immigrants (Immigration to the United States)

Jewish Immigrants (Immigration to the United States)

Italian Immigrants (Immigration to the United States)

Irish Immigrants (Immigration to the United States)

Chinese Immigrants (Immigration to the United States)

Mexican Immigration to the United States

Mexican Immigrants (Immigration to the United States)

Jamaican Immigrants in the United States and Canada: Race, Transnationalism, and Social Capital (The New Americans: Recent Immigration and American Society)

The History of Ethiopian Immigrants and Refugees in America, 1900-2000 (The New Americans: Recent Immigration and American Society)

The American Dream--For Men Only?: Gender, Immigration, and the Assimilation of Israelis in the United States (New Americans)

Americans in Waiting: The Lost Story of Immigration and Citizenship in the United States

Africans In America (Immigration to the United States)

Russian Immigrants in the United States: Adapting to American Culture (New Americans (Lfb Scholarly Publishing Llc).)

Skilled Immigrant and Native Workers in the United States: The Economic Competition Debate and Beyond (The New Americans) (Criminal Justice)

Mexican Immigrants in the Labor Market: The Strength of Strong Ties (The New Americans: Recent Immigration and American Society) (New Americans (Lfb Scholarly Publishing Llc).)

Intermarriage Across Race and Ethnicity Among Immigrants: E Pluribus Unions (The New Americans: Recent Immigration and American Society)

Recommend Documents