The Personal is Political: The Relationship between Marital Status and Immigrant Political Incorporation

Catherine Simpson Bueker, PhD

Presented at the 2003 annual meeting of the American Sociological Association. Please direct all correspondence to: [email protected]

Abstract

The relationship between marital status and immigrant political incorporation has received surprisingly scant attention from scholars, particularly given the effect household factors are known to have on a variety of individual-level outcomes for both the native-born and the foreign-born. Using four years of the Voter and Registration Supplement of the Current Population Survey, I have run multivariate logistic regression models to examine how marital structure is linked to naturalizing and voting among immigrants. The study finds that immigrants who are married and residing with their spouses are the most likely to both naturalize and vote. Migrants who are married but living apart from their partners are the least likely to participate in either type of political activity. Single migrants are found in the middle, with rates of citizenship acquisition and voter turnout falling between those who are married and co-residing and those who are married, but estranged. The relationship between marital arrangement and political integration differs for male and female migrants, respectively. Men who are married and co-residing with their spouses are more likely to be politically incorporated than are their single or estranged male counterparts. In contrast, women look more similar to one another, with little differentiation by marital status. Further, married but separated female migrants are more likely to naturalize than are their male counterparts, while married and single migrants show no difference by gender. This gendered relationship is even more pronounced among Mexican migrants than it is among immigrants, generally. The differential effect of marital status by gender may be the result of men and women migrating for different reasons and with different motivations. Married men may be migrating to earn money to send back home to their families, with the intention of eventually returning to the country of origin. Married women who migrate independently may be doing so to permanently leave husbands in the country of origin, essentially dissolving their marriages and starting anew.

Introduction

Immigrant families, like all families, act as a buffer between the individual and the larger society, influencing a host of individual-level outcomes. Studies of the native-born have found that marital status affects everything from health and life expectancy to an individual’s social circles and economic status (Waite & Gallagher, 2000). Among immigrants, family structure has been identified as influencing who migrates, to where, and for how long (Foner, 1999; Pessar, 1999; Boyd, 1989). Migrating as part of a family both encourages and signifies greater incorporation and longer-term settlement plans.

Previous research exploring the relationship between family structure and individual-level political behavior has consistently found married people to be more politically active than their unmarried counterparts (Stoker, 1995; Wolfinger & Rosenstone, 1980; Milbraith & Goel, 1982; Pattie & Johnston, 2002; Burns et al., 2001). Scholars, again and again, cite the role of spouses as “built-in” political discussants, as well as witting and unwitting promoters of greater stability, larger social networks, and more communal integration.

Yet few studies have explored the impact of marital status on political incorporation among immigrants. Does marital status influence immigrants as it does the native-born? Are immigrant men and women similarly affected by household structure, or should one expect differences by gender? And how does culture, as measured by country of origin, mediate the relationship between marital status and political activity?

Although little is known about the relationship between marital structure and political incorporation among the foreign-born, research focusing on the native-born provides a basic roadmap for the potential mechanisms at work and the resultant outcomes. Studies of the native-born find that the presence of a spouse encourages political incorporation by extending an individual’s social contacts and increasing stability and entrenchment (Wolfinger & Rosenstone, 1980; Milbraith & Goel, 1982; Pattie & Johnston, 2002; Burns et al., 2001). Greater numbers of social ties mean more peer pressure and a higher likelihood of participation (Putnam, 2000). Research suggests that increased social networks similarly work to encourage political mobilization among immigrants (Togeby, 1999; Fennema & Tillie, 1999; Forment, 1989; Cho, 1999).

Marriage also translates into greater stability and community integration. Among the native-born, marriage decreases the likelihood of internal migration due to the increased costs of moving. International migration is significantly more costly and difficult than internal migration, and even more so for a couple than for a single migrant. As a result, marriage likely encourages “rootedness” and stability for the foreign-born, promoting civic incorporation.

Although studies of marital structure and political participation among the native-born can suggest some ways the institution of marriage may impact immigrants in the political realm, such studies have limited applicability. They have consistently measured marital status as a dichotomous variable: one is either married or unmarried. They have also assumed the effects of marriage (and its dissolution) to be the same for men and women, and for all ethnic and racial groups.

In reality, marital status is much more complicated, particularly for immigrant men and women, respectively. For example, many men participate in international migration for the purpose of earning money to send back to families in the country of origin. Such marital separation is, in fact, quite common for international economic migrants who enter the United States for a limited period of time, with the full expectation of returning home to the family that awaits (Lindstrom, 1996). In contrast, many immigrant women coming to the United States may be doing so for the express purpose of leaving a husband, with no intention of returning to the country of origin and the spouse left behind. Although women have historically been viewed as passive migrants, crossing borders for the sake of accompanying their husbands, recent research argues that this role needs to be rethought (Foner, 1999; Pessar, 1999).

The goal of this study is to better understand how marital status influences the political incorporation of immigrants. To systematically assess the effect of family structure on political incorporation, as measured by naturalizing and voting, I use the Voter and Registration Supplements from the 1994, 1996, 1998, and 2000 Current Population Surveys. Through this analysis, I attempt to examine how marital structure affects political incorporation directly and indirectly for both male and female immigrants.

Previous Research

Marriage & Naturalizing

Within the naturalization process, studies have found that married people are more likely to become citizens than are their single counterparts (Yang, 1994; Liang, 1994). Marriage is thought to increase stability, social networks and embeddedness, all factors associated with greater levels of economic, linguistic, and social incorporation (Lopez, 1999; Portes, 1987). These forms of immigrant integration may very well lead to greater civic incorporation (Gordon, 1964; Lien, 1994).

Beyond the increased social connections thought to result from marriage, immigrant couples in the United States suggest greater levels of stability and commitment to remaining in this country than do their single counterparts. Sojourners tend to migrate sans family, with the plan of eventually returning to the country of origin after a specific economic goal has been met (Lindstrom, 1996). In contrast, migrating as a couple suggests a longer-term vision, a pledge to permanent settlement. The process of finding a job, acquiring a visa, and locating a place to live is difficult enough for one person. For both members of a couple to migrate requires an even greater level of work and dedication. Not surprisingly, the benefits of remaining in the United States and the costs of return migration are also greater for a couple than for an unmarried or separated migrant. The increased emotional and financial expenses for married migrants, over a single individual, may further explain the greater propensity of married individuals to naturalize (Yang, 1994).

At the same time, many benefits resulting from citizenship accrue to the entire family, rather than solely to the individual who naturalizes. Although the political benefits of naturalizing, namely the opportunity to vote and run for elected office, are available exclusively to the individual holding U.S. citizenship, many other benefits of citizenship advantage the entire family. American citizenship provides greater access to state and federal loans, eligibility for social welfare benefits, and the opportunity for family reunification, with the benefits of citizenship increasing in recent years with the passage of legislation such as Proposition 187 in California and the 1996 federal welfare reform act. With either spouse holding citizenship, the entire family gains access to many of these benefits. As a result, an immigrant who is married to a native-born American or to a fellow immigrant who has acquired citizenship may actually be less likely to naturalize, as he or she already benefits indirectly from his or her spouse’s citizenship status (Jasso & Rosnzweig, 1990). Less is known about the effect of endogamous versus exogamous marriage partners on political participation. Unfortunately, the data used in this research do not contain information on the ethnicity or citizenship status of the spouse.

Marriage & Voting

Marital status also affects voter turnout. Married immigrants are about 20% more likely to vote than their single counterparts (Bass & Casper, 2001). Numerous voter studies conducted on the native-born have found parallel results (Leighley & Vedlitz, 1999; Tate, 1991; Putnam, 2000; Wolfinger, 1994; Kingston & Finkel, 1987), suggesting that the relationship between marriage and voting cuts across diverse populations.

One of the primary mechanisms mediating between family structure and political activity appears to be stability. A study with a more nuanced approach to marital status finds that individuals who have been in the state of marriage for an extended period of time are indeed more likely to vote than are their single counterparts. However, changes in marital status, be it getting married or getting divorced, temporarily depresses political participation rates to below those of their single counterparts (Stoker, 1995).

On the surface, these findings suggest a divergence from the many previous studies linking marriage with voter turnout. In actuality, the findings are not at all in contradiction with other work, but are the result of more accurately measuring the mechanism of stability thought to mediate between marital status and political activity. It is not marriage, per se, that encourages political involvement, but the constancy of the spouse and household setup.

Similarly, greater levels of socialization may result from marriage, with increased socialization leading to higher levels of voter turnout (Cho, 1999). Although the increased socialization and more extensive networks resulting from marriage may not include native-born or naturalized citizens, and instead be larger co-ethnic networks, co-ethnic networks can work to increase political mobilization in their own ways (Portes, 1987). As ethnic group awareness increases, so too, may political activity (Lien, 1994).

Greater levels of stability and larger social networks are largely a function of the ongoing presence of a spouse. Being married but physically separated from one’s partner likely translates into lower levels of stability and smaller social networks as compared with those who are married and residing together. Marital transitions, whether positive or negative, represent a period of instability. In short, “rootedness,” as opposed to the title of marriage, is what encourages political incorporation and participation.

Gender, Family Structure, and Country of Origin

The “gender gap” in political participation has existed for many years (Wolfinger & Rosenstone, 1980; Milbraith & Goel, 1982; Burns et al., 2000). Unfortunately, much of the knowledge about the political gender gap is based on studies of the 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 (Tate, 1991; Lien, 1998).

Similarly, the role of marriage, and its implications for men and women, respectively, varies by ethnic group. Marriage has historically been one of the most gendered institutions, having different implications for men and women (Katz Rothman, 1998). In most industrialized societies, marriage has positioned women inside the home to care for the domestic aspects of life, while placing men in the public sphere. The expectations for and roles given to married women have traditionally been distinct from those of married men, resulting in different health, economic, and social outcomes (Gove, 1972; Pienta, Hayward, & Jenkins, 2000; Waite & Gallagher, 2001).

Although these patterns have been reproduced across time and space, American society has developed to give women of all marital statuses greater freedoms and opportunities. The opportunities afforded married women in U.S. society likely become more salient when experienced by those from less egalitarian countries. As a result, many scholars argue that gender should play a particularly significant role in the incorporation process, with immigrant women embracing new found freedoms in the United States and immigrant men struggling with downward status readjustment (Pessar, 1999, Grasmuck & Pessar, 1991; Foner, 1999; Jones-Correa, 1998).

Case studies of Dominican immigrants have identified the differential impact of family structure on the incorporation of immigrant men and women, respectively (Jones-Correa, 1998; Grasmuck & Pessar, 1991). Dominican women, in caring for the household and family, develop increased contact with mainstream institutions, such as schools and social service agencies. This contact with mainstream institutions, combined with new found freedoms, leads to increased political activity. At the same time, their husbands are struggling with the indignities and difficulties of immigration. The men respond by staying apart from American society, participating in ethnic organizations, and focusing on return to the country of origin (Jones-Correa, 1998).

Other studies of Dominican and Mexican immigrants have found similar patterns (Hongdagneu-Sotelo, 1994; Massey, 1986; Bass & Casper, 1999; 2001). Dominican wives have been found sabotaging their husbands’ attempts to return to their country of origin by using up savings kept for reverse migration. In this way, stays in the United States become indefinite, with the husband constantly trying to rebuild the reserve spent by the wife (Grasmuck & Pessar, 1991). Mexican migrants similarly show variation by gender, “…the urge to return to Mexico is especially pronounced among men” (Myers & Lee, 1996, p.53). Quantitative work has likewise found Latino women more likely to vote than Latino men (Bass & Casper, 1999; 2001).

The differing foci, strategies, and outcomes of husbands and wives suggest that marital status has very different implications across genders and countries of origin. The effect of family on political incorporation must be examined both by sex and within particular cultural contexts. As one scholar argues “…it should always be remembered that families or households are defined by different economic, political, cultural, demographic, and ecological settings and are not social units with universal behavior” (Faist, 1997, p. 205). The impact of family structure on the political incorporation of immigrants is clearly an area that needs closer examination and must be done with a greater appreciation for the complexity of marital status, country of origin, and gender.

Limitations of Previous Studies

Although previous research has provided insight into how marital structure impacts individual political activity, many questions remain unanswered. Much of the previous research conducted in this area focuses on the native-born (Stoker, 1995; Wolfinger & Rosenstone, 1980; Milbraith & Goel, 1982; Pattie & Johnston, 2002; Burns et al., 2001). Given that gaps in political activity exist by ethnic group (Lien, 1998; Junn, 2000), the applicability of earlier studies largely conducted on the white native-born is questionable.

Firstly, the native-born are automatically citizens and therefore never entertain the notion of citizenship acquisition. The mechanisms that encourage the native-born to go to the polls may be quite different from those that encourage immigrants to become citizens, given the more extensive and immediate benefits that accrue from naturalizing versus voting. Citizenship, though necessary for voting and holding many elected offices, is not just about access to the political realm. It also provides social welfare benefits, legal protections, greater access to jobs and loans, and family reunification (Jasso & Rosenzweig, 1990). Given the family-wide benefits resulting from one member of the family naturalizing, it seems particularly important to better understand how marital structure influences political decisions.

Focusing exclusively on the native-born also means a lack of attention paid to the selectivity of migration flows and the influence that such selectivity has on mediating between marital structure and political incorporation. Migration flows from certain countries of origin, such as Mexico, tend to be dominated by the traditional male sojourner. Mexican men may arrive without spouses, but with the intention of returning to the home society. In this instance, the migration may be viewed as temporary, with the express purpose of earning money for the family at home. Flows from other countries, such as India, tend to be dominated more by married couples migrating together. In this instance, the migration may be viewed as more permanent, with both spouses acquiring citizenship.

Further, the relationships between marital structure and political participation in studies of the native-born vary for men and women (Burns et al., 2001; Stoker, 1995). If this is the case for the native-born, variations likely also exist for men and women from different countries of origin. Unfortunately, there is again a dearth of research examining how culture interacts with marital structure to influence political incorporation.

The little research that has examined the effect of marital structure on the political activity of different immigrant groups has tended to be qualitative (Grasmuck & Pessar, 1991; Hongdagneu-Sotelo, 1994; Jones-Correa, 1998). Although this research has helped us to better understand the relationship between marital structure and political activity, it has focused on Mexicans and Dominicans, and in very specific regions of the United States. One cannot assume greater generalizability.

Larger-scale quantitative studies that have examined the impact of marital structure on political activity have failed to control for some very important variables, namely country of origin (Jasso & Rosenweig, 1990; Bass & Casper, 1999; 2001). Including variables that measure country of origin characteristics, such as geographic distance, GDP, or native-language, certainly tap into some important home society characteristics, but could potentially place very diverse immigrant groups together. Given the gender and marital selectivity of migration flows by country of origin, one must explicitly control for this variable. Although some studies have begun to examine such gender/culture interactions (Bass & Casper, 1999; 2001), more work must be done in this area.

Contributions and Limitations of This Study

Very little has been done to gain insight into how marital structure influences the political activity of immigrants. The goal of this study is to provide a systematic analysis of whether marital structure influences citizenship acquisition and voter turnout among immigrants, and whether the effect varies by male and female immigrants, respectively.

This research attempts to further contribute to the field of family and immigration studies by looking at marital status in a somewhat more multifaceted manner. Previous research that has examined the role of marital structure on political activity among the native-born has commonly used a “married” or “unmarried” dichotomy. Such broad categorization is particularly limiting when examining immigrants, many of whom are married but migrate individually.

Further, this study will specifically examine how marital structure impacts the political incorporation of Mexican-origin men and women, respectively, to see how culture mediates the effect of marital structure. Ideally, this study would be able to examine whether marital structure differentially affects men and women from a variety of countries of origin, in an effort to compare and contrast the role of culture. Unfortunately, the Current Population Survey does not over sample immigrants and as a result, there are simply too few men and women of each marital status for the majority of countries studied in this project. Although the inability to do this larger comparison is certainly a limitation, examining the effect of marital structure on male and female immigrants, generally, and within one country of origin group provides an additional window onto the patterns at work.

Hypotheses

Marriage is thought to encourage political participation by providing a conversation partner, more stability, and greater social networks. Although increased social networks resulting from marriage may or may not be U.S. oriented and could be dominated by co-ethnics, increased social networks of either variety may encourage political activity through ethnic mobilization and/or through increasing contact with mainstream American society.¹

The mechanisms that mediate between marriage and political incorporation for immigrants- the conversation partner, the larger social networks, and the greater stability- likely only work with the presence of a spouse. It is likely not the title of marriage that impacts political activity, but something about the structure. Previous studies have tended to view marital status as a dichotomous variable, failing to examine the more multifaceted nature of marriage. Hypotheses one and two attempt to move past the limitations of previous studies by more precisely defining different marital statuses.

The Marriage Hypothesis:Married immigrants should be more likely to both naturalize and vote than their single counterparts, given the greater levels of stability, social contact and embeddedness thought to result from marriage.

The Separation Hypothesis: Married immigrants who are separated from their spouses should show lower levels of naturalizing and voting than their married or single counterparts.

Previous research has suggested that marital status, in and of itself, should effect political incorporation and participation. However, gender theory suggests that marital status should differentially affect men and women, with women more likely to become politically incorporated than their male counterparts, across all levels of marital status. Research in other areas has shown men to be more reliant on their partners for social contact than are women (Gove, 1972; Waite & Gallagher, 2000). Immigration research has also found that women are more likely to become politically incorporated than are their male counterparts (Jones-Correa, 1998; Bass & Casper, 1999; 2001). The third hypothesis is a test of this interaction between marital status and gender, examining whether the gender patterns found in other areas of research reproduce themselves in the realm of political incorporation.

The Gender Hypothesis: Marital status should differentially affect men and women. Further, women should be more likely to naturalize and vote than their male counterparts at all levels of marital status.

As discussed earlier, the interaction between marital status, gender, and culture has not been fully explored within large-scale quantitative studies. One brief examination of the interaction between gender and culture, regardless of marital status, has found that Latino women are more politically incorporated than Latino men (Bass & Casper, 1999; 2001). Similarly, a series of community studies exploring incorporation among Latino immigrants have found marital status to have differential effects on men and women, with women more likely to become incorporated (Jones-Correa, 1998; Grasmuck & Pessar, 1991). Ideally this study would systematically examine the interaction between gender and marital status among immigrants from multiple countries of origin. Unfortunately the small sample sizes of many of the immigrant groups in the CPS limit our ability to conduct such a statistical analysis. The following hypothesis is designed to quantitatively test the differential impact of marital status on political incorporation among Mexican men and women, respectively, in an effort to gain some greater insight into the relationship between gender, marital structure, and culture on political activity.

The Gender/Culture Hypothesis: Mexican women will be more likely to naturalize and vote than Mexican men at all levels of marital status. Additionally, marital status should have a smaller impact on Mexican women’s political incorporation.

Data and Methods

I have pooled the November 1994, 1996, 1998, and 2000 Current Population Surveys, entitled The Voter and Registration Supplements, and run a series of multivariate logistic regression models to examine the relationship between marital structure and political incorporation among immigrants.

The specific citizenship question on which I will be relying asks, “In what citizenship group do you belong?” with the possible responses including:

Native, born in US;
Native, born in PR or US outlying area;
Native, born abroad of US parent(s);
Foreign Born, US citizen by naturalization;
Foreign Born, not a US citizen.

I subdivide this citizenship question into a dichotomous variable by using responses 4 and 5 of the original question.

The specific voting question that I will use in this study 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 _?” with the question asked only of those 18 years of age or older who cited U.S. citizenship by birth or naturalization. The format of this question is intended to limit over-reporting by removing the stigma of non-voting. This question is specifically asking about voter participation at the federal level.

The marital status variables included in the analysis are constructed from the question “Are you now married, widowed, divorced, separated or never married?” with the possible responses being:

Married – Spouse Present
Married-Spouse Absent
Widowed
Divorced
Separated
Never Married

I have created two new dichotomous variables from this question, “nomar” and “marab.” Respondents who reported themselves as widowed, divorced, separated, or never married are coded as “1” for “nomar.” Those who report being married but having an absent spouse are coded as “1” for “marab.” All other responses are recoded as “married” and serve as the reference category. Ideally, one could compare the effect of cohabitation versus marriage on immigrant political activity, but this is not possible, given the data.

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. 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. I have used the national poverty thresholds, released annually by the Census Bureau, to calculate whether an individual is above or below the poverty line. Individuals are assigned this status based on the poverty thresholds in the year in which they participated in the CPS. 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 naturalize.²

I have included three control variables of age, age ² , and workforce participation. The age variables are both continuous. I have included both age and age ² 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 working full-time.

I have drawn 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 wave of the survey contributes approximately one fourth of the cases to my pooled samples. I have included control variables for the year in which the respondent participated in the survey. The years 1994 and 1998 are congressional election years, versus 1996 and 2000, which are presidential election years. Those who vote in congressional elections may be quite different from those who vote in presidential elections. By including survey year in the regression model, I am able to control for any variability resulting from the year or type of election.

The sample consists of the following ten immigrant groups: Mexicans, Cubans, Canadians, Britons, Italians, Former Soviets, Filipinos, Indians, Southeast Asians, and Mainland Chinese. These ten groups provide diversity in terms of geography, history of migration, and political experience.

For the citizenship sample, I selected those age 18 or older, who had lived in the country for at least five years, and who had answered the citizenship question. This sample is comprised of 17,019 cases.

The voting sample is culled from the larger citizenship sample. The voting sample consists of 6,641 naturalized immigrants who met the age, residency, and citizenship requirements, and answered the voting question. The total number of naturalized cases from the citizenship sample does not equal the total number in the voting sample, as some naturalized immigrants were dropped for failing to answer the voting question.

None of the cases have missing data on citizenship status. About 10% or 751 of the 7,392 individuals who have naturalized have missing voting information. I have made the decision to drop these cases from the second stage voting model. Though there may be some selectivity among cases with missing voting data, the “missings” look quite similar to the cases with complete information. Approximately 22% of those with missing voting information have less than a high school degree, as compared with 22% of voters and 34% of non-voters. Similarly, 50% of those with missing voting information have more than a high school diploma, as compared with 39% of non-voters and 55% of voters. In keeping with this educational distribution, approximately 14% of respondents with missing voting data live below the poverty line, versus 13% of voters and 21% of non-voters. Women comprise 49% of those with missing data, versus 54% of both voters and non-voters. Missing cases are fairly evenly distributed across country of origin. Between 11% and 14% of immigrants from Southeast Asia, China, the Soviet Union, Italy, the Philippines, and India lack data. About 9% of Mexicans have missing information and between 5% and 7% of Canadians, Cubans, and Britons lack voting data. Based on the relative similarity of those with missing voting information versus those with complete voting data, I have chosen to take the more cautious route and restrict the two samples to immigrants with complete information on the respective form of political participation under study. Therefore, all of the data included in this study, including descriptive statistics, are based on cases with complete information on the dependent variables.

Approximately 10% of all of the cases have missing income data. I have calculated the median income for the respective immigrant group and filled in the missing information with these values.

Findings

Table 1 presents the descriptive statistics of the citizenship sample. The sample is evenly split between men and women. The majority of both male and female immigrants report being married and living with their spouses. Interestingly, over one third (37%) of women report not being married, versus 29% of men. Clearly, the days of the single male sojourner are over. At the same time, male migrants are slightly more likely to be married but living apart from their spouses than are female migrants, in keeping with the notion of the sole male economic migrant, earning money to return home.

Table 2 highlights the relationship between marital status and naturalizing.³ In Model 1 (the main-effects model), both being single and being married but separated are statistically significant in predicting citizenship. Single individuals are about 92% as likely to naturalize as are their counterparts who are married and living with spouses. Individuals who are married but separated from their spouses are only about 65% as likely to hold U.S. citizenship as those who are married and living with their spouses.

These findings support the Marriage Hypothesis that states that married immigrants residing with their mates should be more likely to naturalize than their unmarried counterparts. The mechanisms at work are likely the greater levels of stability and larger social networks resulting from co-residential marriage.

The data also show evidence for the Separation Hypothesis that states married but separated individuals should have the lowest propensities to hold citizenship of the three marital status categories. Stability and social networks are again the probable influences mediating between marital status and political incorporation, only now one sees what happens in the absence of such stability and connections.

The relationship between marital status and political incorporation becomes more pronounced in Model 2, which includes interaction terms between gender and marital status. Because interactions are included, the interpretation of the main-effects must be newly interpreted. The main-effect of being unmarried is now substantively the effect of being unmarried and male, as the interactions between gender and marital status are included in the model. Single men are about 89% as likely to be U.S. citizens as married men. Men who are married but living apart from their spouses are about 53% as likely to naturalize as their married male counterparts, living with their wives.⁴ Figure 1 illustrates the effect of marital status on men and women, respectively.

As Figure 1 shows, the patterns are a bit different among women, suggesting marital status does vary in its effect by gender. Women who are married but living apart from their spouses are about 84% as likely to naturalize as married female migrants who are living with their husbands. Unmarried women do not show statistically significant differences in their propensity to naturalize from women living with their spouses.

The impact of marital status on political incorporation clearly affects men and women differently. The differences across marital statuses among women are smaller and less statistically significant than what is found among men. This is illustrated in Figure 1 where the bars representing single and separated women are closer to 1 than are the bars representing single and separated men, thus showing smaller differences in the odds of naturalizing from the reference group of immigrants who are married and residing with spouses. Whereas single women show no difference in their propensity to become citizens from married women living with husbands, single men are statistically less likely to naturalize than are married men living with wives. Further, the difference in the propensity to naturalize between those who are separated from spouses and those who are married and living together is greater for men than for women. Men tend to be more substantially impacted by marital status in the naturalization process than are their female counterparts, in keeping with previous health and income studies (Gove, 1972). These findings, supporting the Gender Hypothesis, are likely the result of men’s greater dependency on their spouses for social contact and conversation partners. Women tend to have more social options and larger networks outside of the home, resulting in lower levels of spousal dependency.

But how do men and women within the same marital statuses compare with each other? The findings, illustrated in Figure 2, show that married men and women who are living with their spouses do not have statistically significant differences from one another in their propensities to naturalize. The same pattern is found among single men and women. Only among the married but separated are there statistically significant differences between males and females. Women in this category are about 55% more likely to naturalize than are men.

Table 3 highlights the relationship between marital status variables and voting. Both forms of marital status show statistically significant differences from the reference group of married and co-residing immigrants. Those who are married but separated are about 63% as likely to vote as those living with their spouses. Unmarried migrants are about 79% as likely to vote as the reference group.

These findings are in keeping with those from Table 2 and lend additional support for the Marriage and Separation Hypotheses. Those who are married and living with their spouses have the strongest positive relationships with voting, followed by those who are single. Married but separated individuals show the lowest odds of going to the polls.

Though Tables 2 and 3 show similar findings, it should be noted that marital status has a stronger association with voting than it does with naturalizing. A married but separated individual is about 65% as likely to naturalize as someone who is married and living with a spouse. The odds of the same separated individual voting are 63%, compared with someone who is married and living with his or her spouse. Similarly, an unmarried individual in Table 1 is about 92% as likely to naturalize as someone married and living with his or her partner. In Table 2, the comparable odds are 79%.

The Mexican Sample

Table 4 highlights the relationship between marital status and political incorporation among Mexican-origin immigrants. In Model 1, the status of being separated from one’s spouse is statistically significant. This group is less than half (48%) as likely to naturalize as those who are married and co-residing. Unmarried Mexican migrants are no more likely to naturalize than are their married counterparts living with spouses.

These data suggest that the association between marital status and political incorporation varies in its effect across immigrant groups. In Table 2, both unmarried and married but separated individuals show statistically significant differences from the reference group (married immigrants residing with spouses). In Model 1 of Table 4, one sees that the difference between the unmarried and the reference group has disappeared. Mexican-origin persons of these two marital statuses are not distinct from one another in their propensities to naturalize. Though the Married Hypothesis has strong support from Tables 2 and 3, the Mexican sub-sample does not reinforce this hypothesis. However, the effect of marital separation among Mexican-origin immigrants does provide further support for the Separation Hypothesis.

It should be noted that Model 1 shows a difference in the naturalization propensities between Mexican men and women. Women are about 12% more likely to naturalize than are their male counterparts. Though this study is not focused on individual-level factors, but rather on family factors and how their effects vary across levels of gender, this main-effect finding is of interest. It suggests that gender matters for the political incorporation of Mexican-origin immigrants, in keeping with previous studies of Latinos (Jones-Correa, 1998; Grasmuck & Pessar, 1991; Bass & Casper, 1999; 2001). These main-effect gender findings may also translate into interaction effects with marital status.

Model 2 in Table 4 includes interaction terms between gender and marital status, finding that the effect of marital status does indeed vary by gender for Mexican-origin immigrants. In this model, it is again the absence of a spouse that seems to matter. Men who are married but separated from their spouses are only 34% as likely to naturalize as their married counterparts who are living with spouses, as is shown in Figure 1. The relationship is substantially less for women who are separated from their husbands, with these women about 82% as likely to naturalize as married Mexican women who are living with their husbands. There is no difference between unmarried Mexican immigrants of either sex and their married counterparts who are living with spouses. Overall, these findings suggest that the relationship between marital status and political incorporation varies for Mexican men and women, with married but separated men substantially less likely to naturalize than Mexican women in similar family structures.

In comparing Mexican men and women within the same marital structure, the findings show no statistically significant difference between those who are married and residing with a spouse, and those who are single. The big difference between men and women is among those who are married but separated. In this category, women are at significantly greater odds of naturalizing, as is illustrated in Figure 2. Females who are married but estranged from their spouses are 2.6 times more likely to naturalize than are their male counterparts.

The critical findings here are both the differential effect of marital status by gender and the influence of gender on marital status. The first set of findings show that Mexican men and women are differentially influenced by their marital status. These findings may, in part, be the result of selective migration flows, with Mexican men more likely to enter the United States as undocumented laborers than are their female counterparts. If this is indeed the case, Mexican men may simply be naturalizing at lower rates than Mexican women because fewer are able to do so.

Another possible explanation for the difference in naturalization rates between Mexican men and women may have to do with their respective adaptation to American society. Earlier research on Dominican-origin immigrants has found Dominican men to have a harder time adjusting to American society. As a result, Dominican men stay tied to ethnic organizations and remain focused on returning to the country of origin. In contrast, their female counterparts embrace American society and do what they can to remain in the United States (Jones-Correa, 1998; Grasmuck & Pessar, 1991). Mexican-origin women may similarly find the opportunities in American society more appealing than do their male counterparts who often struggle with a downward status readjustment and as a result, keep their focus firmly on returning to the homeland.

The second set of results show that separated women are far more likely to naturalize than are separated men. These data strongly support the Gender/ Culture Hypothesis. Again, the difference may be in male and female reactions to American society- new opportunities and freedom for women, but downward status readjustment and indignity for men. The difference may also result from why separated men and women migrate to the United States. As discussed earlier, married men who migrate without their wives may come to the United States as temporary economic migrants, with the intention of returning to Mexico after having earned enough money. In contrast, married Mexican women who migrate without their husbands may be doing so to escape an unhappy marriage, with the intention of staying in the United States permanently.

The data also show that among Mexicans, the only marital status that shows a difference from the reference group are those who are married but separated from their spouses. Mexican immigrants who are married and residing with their spouses and those who are unmarried show similar levels of political incorporation. Individuals residing with their spouses may have put down roots and made the decision to stay permanently in the U.S. Those who are unmarried may have created their own networks and plans. It is those Mexican immigrants who are estranged from their spouses who are the least likely to commit politically to the United States.

Within the voting model (Table 5), the relationship with marital status among Mexicans is more extreme. In Model 1, those who are married but separated are only 21% as likely to go to the polls as those who are married and residing with spouses. Unmarried Mexican immigrants are 69% as likely to vote as the reference group. Clearly, the absence of a spouse has a more depressing effect on voting than on naturalizing.

Model 2 includes interactions between marital status and gender. Though the interactions are not significant, there is a change in both the interpretation and the results of the main-effects. Married but separated men show no difference in their propensity to vote from married men living with spouses. Unmarried men are 66% as likely to vote as their counterparts who are married and living together. Women show no differences from one another in their propensity to naturalize across all levels of marital status. Though the actual interactions are not significant, there is still some difference by gender and marital status, as is evidenced by the main-effects. This suggests some additional support for the Gender/ Culture Hypothesis.

Discussion

The results from this study reflect the relationship between family factors, specifically marital status, and political incorporation. Generally speaking, married immigrants who are living with their spouses appear to be more politically integrated, via naturalizing and voting, than are their unmarried or separated counterparts, supporting the Marriage Hypothesis. This general finding is in keeping with previous work on political incorporation among the foreign-born (Liang, 1994; Yang, 1994) and among the native-born (Leighley & Vedlitz, 1999; Tate, 1991; Putnam, 2000; Wolfinger & Rosenstone, 1980; Milbraith & Goel, 1982). Marriage appears to provide the stability, larger social networks, and greater community integration that encourage both naturalizing and voting among immigrants.

The non-dichotomous nature of marital status is also revealed in these findings. It is not simply that those who are married and living together are more likely to be politically incorporated than everyone else, but rather that there is a type of hierarchy. Those who are married but separated from their spouses have the lowest rates of political integration, below those of unmarried migrants. This is true for both naturalizing and voting. These results are in keeping with the posited mechanisms of stability and embeddedness. Married immigrants who are living with their spouses likely have the greatest levels of stability, ties, and commitment to the United States. Single migrants, though possibly less bound to the host-community, may be considering permanent settlement in the U.S. Those who are married but separated have the lowest levels of permanence and commitment. The estranged spouses may be in the country of origin or simply in another American city. In either case, the respondents who live apart from their partners appear less likely to become politically incorporated and participatory, supporting the Separation Hypothesis.

It is not enough to say that marital status has a relationship with political incorporation. Matrimony has different implications for men and women, providing them with diverse opportunities and expectations. Examining how gender interacts with marital status across both types of political incorporation has revealed that men and women are differentially impacted by marital status. Men who are married but separated from their spouses are only half as likely to have naturalized as are men living with their wives. Though women in the same marital status are less likely to naturalize than are their married counterparts who are residing with spouses, the difference between the two groups of women is substantially smaller. At the same time, unmarried men are less likely to naturalize than are married men living with spouses, while women in these two marital statuses show no statistically significant difference from one another.

The test of the Gender/ Culture hypothesis, run on the Mexican sub-sample, further elucidates these findings. The interaction effect between gender and spousal separation is significant on the larger immigrant sample. The data show even more significance on the smaller sub-sample, an interesting result given that significance is harder to achieve as samples become smaller. Though unmarried Mexican migrants, both male and female, show no difference in their propensities to naturalize from their married counterparts living with spouses, there is a substantial difference among those who are married but separated. This difference is further clarified through interactions. Men in this category are only about a third as likely to have naturalized as married men residing with spouses. Married but separated women are about 82% as likely to be U.S. citizens as Mexican women living with husbands. Clearly, gender, culture, and marital status all influence political incorporation, with Mexican men more impacted by separation than either Mexican women or than immigrant men, generally.

The differing propensities of Mexican men and women to naturalize and vote suggest the same patterns found by previous qualitative research (Jones-Correa, 1998; Grasmuck & Pessar, 1991; Pessar, 1999). Prior studies have argued that Latino men have a harder time acclimating to American society, often experiencing painful status readjustment. Their method of handling this difficult environment is to focus on return to the country of origin. At the same time, the new prospects and liberty immigrant women experience in the United States may lead to permanent settlement and incorporation, made more concrete by citizenship acquisition.

At the same time, the results from the Mexican sub-sample may say as much about Mexican migration patterns, generally, as they do about the role of gender or marital status. Mexican immigrants, male and female, married and unmarried, are less likely to become politically incorporated than are many of their foreign-born counterparts from other parts of the world (Portes & Rumbaut, 1996). These depressed levels of citizenship acquisition may be the result of an overrepresentation of undocumented immigrants within the Mexican migration stream to the United States. The United States has a long history of relying on Mexican men to work in agriculture, a practice that began with the Bracero Agreement between the U.S. and Mexican governments (Massey et. al, 1999). Though this agreement has long since expired, many Mexican immigrants, particularly younger men, continue to cross their northern border to find work in the United States. The fact that Mexican men show a lower likelihood of holding U.S. citizenship than immigrant men of similar marital statuses and than their female counterparts from Mexico may be further evidence of the selectivity of the Mexican migration flow.

Overall, it is important to note the differing relationships between marital status and naturalizing and voting, respectively. Though the general trends are similar, with married migrants living with spouses having the highest naturalization and voting rates and the estranged having the lowest propensities to naturalize and vote, the magnitude varies by process. On both the larger immigrant sample and on the Mexican sub-sample, the differences in participation by marital status are more profound in predicting voter turnout. This suggests that the association between citizenship and marital status is weaker, and perhaps, more determined by individual-level characteristics. In contrast, the relationship between marital structure and voting behavior appears to be stronger and may be more affected by those immediately surrounding the individual.

Conclusion

The findings from this research suggest that family factors, in conjunction with individual-level characteristics, help to tell the story of immigrant political incorporation. Naturalizing and voting are determined by more than socio-economic or singular demographic characteristics. Those around us affect the decisions we make and how we behave.

Although these findings have helped to uncover more of the political incorporation story, they are only a beginning. Ideally, future researchers will explore how other family and household factors influence civic integration. Due to data limitations, I have only examined the influence of formal unions on political activity. Are those who co-habitate similar to those who are married and residing together, or does the legal institution of marriage have some additional influence on immigrant political integration? What influence do others in the household, such as children, have on civic incorporation? Further, a systematic analysis of immigrants from other countries of origin would allow for a fuller understanding of how gender and marital relations vary by culture to influence political incorporation. Finally, how do various immigration and social welfare policies differentially influence immigrants of different marital statuses? Though naturalizing and voting are undertaken by the individual, the factors that encourage or discourage such behavior exist at multiple levels, not least of which is the household.

References

Bass, L. and Casper, L. (1999). “Are There Differences in Registration and Voting Behavior Between Naturalized and Native-Born Americans?” Population Division, U.S. Census Bureau, Washington, D.C.

—. (2001). “Differences in Registering and Voting Between Native-Born and Naturalized Americans.” Population Research and Policy Review 20:483-511.

Boyd, M. (1989). “Family and Personal Networks in Migration.” International Migration Review:638-665.

Burns, N., Schlozman, K., and Verba, S. (2001). The Private Roots of Public Action: Gender, Equality, and Political Participation. Cambridge: Harvard University Press.

Cho. W. (1999). “Naturalization, Socialization, Participation: Immigrants and (Non-) Voting. The Journal of Politics 61(4):1140-1155.

Converse, P. (1969). “Of Time and Partisan Stability.” Comparative Political Studies 2:139-171.

Faist, T. (1997). “The Crucial Meso Level.” pp. 187-218 in International Migration, Immobility and Development: Multidisciplinary Perspectives, edited by T. Hammar. Oxford, UK.

Fennema, M. and Tillie, J. (1999). “Political Participation and Political Trust in Amsterdam: Civic Communities and Ethnic Networks.” Journal of Ethnic and Migration Studies 25, no.4:703-726.

Foner, N. (1999). “The Immigrant Family: Cultural Legacies and Cultural Changes.” Pp. 257-274 in The Handbook of International Migration: The American Experience, edited by C. Hirschman, P. Kasinitz and J. DeWind. New York: Russell Sage Foundation.

Forment, C. (1989). “Political Practice and the Rise of an Ethnic Enclave: The Cuban American Case, 1959-1979.” Theory and Society 18:47-81.

Goldscheider, C. (1995). Population, Ethnicity, and Nation-Building. Boulder, Colo.: Westview Press.

Gordon, M. (1964). Assimilation in American life: the Role of Race, Religion, and National Origins. New York,: Oxford University Press.

Gove, W. (1972). “The Relationship between Sex Roles, Marital Status and Mental Illness.” Social Forces 51:34-44.

Grasmuck, S. and Pessar, P. (1991). Between Two Islands: Dominican International Migration. Berkeley: University of California Press.

Hongdagneu-Sotelo, P. (1994). Gendered Transitions: Mexican Experiences of Immigration. Berkeley: University of California Press.

Jasso, G. and Rosenzweig, M. (1990). The New Chosen People : Immigrants in the United States. New York: Russell Sage Foundation.

Jones-Correa, M. (1998). “Different Paths: Gender, Immigration, and Political Incorporation.” International Migration Review XXXII:326-349.

Junn, J. (2000). “Participation in Liberal Democracy: The Political Assimilation of Immigrants and Ethnic Minorities in the United States.” pp. 187-214 in Immigration Research for a New Century, edited by N. Foner, R. Rumbaut, S. Gold. New York: Russell Sage Foundation.

Katz Rothman, B. (1998). “Motherhood Under Patriarchy.” pp. 21-32 in Families in the U.S., edited by K. Healey and A. Garey. Philadelphia: Temple University Press.

Kingston, P. and Finkel, S. (1987). “Is there a Marriage Gap in Politics?” Journal of Marriage and the Family 49:57-64.

Liang, Z. (1994). “Social Contact, Social Capital, and the Naturalization Process: Evidence from Six Immigrant Groups.” Social Science Research 23:407-437.

Lien, P. (1994). “Ethnicity and Political Participation: A Comparison between Asian and Mexican Americans.” Political Behavior 16(2):237-264.

Lien, P. (1998). “Does the Gender Gap in Political Attitudes and Behavior Vary across Racial Groups?” Political Research Quarterly 51(4):869-894.

Lindstrom, D. (1996). “Economic Opportunity in Mexico and Return Migration From the United States.” Demography 33(3):357-374.

Lopez, D. (1999). “Social and Linguistic Aspects of Assimilation Today.” pp. 212-222 in The Handbook of International Migration: The American Experience, edited by C. Hirschman, P. Kasinitz and J. DeWind. New York: Russell Sage Foundation.

Massey, D. (1986). “The Settlement Process Among Mexican Migrants to the United States.” American Sociological Review 51:670-684.

Milbrath, L. and Goel, M. (1982). Political Participation: How and Why do People Get Involved in Politics? Washington, D.C.: University Press of America.

Myers, D. and Lee, S. (1996). “Immigration Cohorts and Residential Overcrowding in Southern California.” Demography 33:51-65.

Niemi, R., Stanley, H and Evans, C. (1984). “Age and Turnout Among the Newly Enfranchised: Life Cycle versus Experience Effects.” European Journal of Political Research 12:371-386.

Pessar, P. (1999). “The Role of Gender, Households, and Social Networks in the Migration Process: A Review and Appraisal.” pp. 53-70 in The Handbook of International Migration: The American Experience, edited by C. Hirschman, P. Kasinitz and J. DeWind. New York: Russell Sage Foundation.

Pienta, A., Hayward, M. and Jenkins, K. (2000). “Health Consequences of Marriage for the Retirement Years.” Journal of Family Issues 21:559-587.

Portes, A. and Bach, R. (1985). Latin Journey: Cuban and Mexican Immigrants in the United States. Berkeley: University of California Press.

Portes, A. (1987). “The Social Origins of the Cuban Enclave Economy in Miami.” Sociological Perspectives 30:340-372.

Portes, A. and Rumbaut, R. (1996). Immigrant America : A Portrait. 2nd ed. Berkeley, CA: University of California Press.

Putnam, R. (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

Stoker, L. and Jennings, M. (1995). “Life-Cycle Transitions and Political Participation: The Case of Marriage.” The American Political Science Review 89:421-433.

Togeby, L. (1999). “Migrants at the Polls: An Analysis of Immigrant and Refugee Participation in Danish Local Elections.” Journal of Ethnic and Migration Studies 25, no. 4: 665-684.

Tate, K. (1991). “Black Political Participation in the 1984 and 1988 Presidential Elections.” The American Political Science Review 85:1159-1176.

Waite, L. and Gallagher, M. (2000). The Case for Marriage: Why Married People are Happier, Healthier, and Better Off Financially. New York: Doubleday.

Wolfinger, R. and Rosenstone, S. (1980). Who votes? New Haven: Yale University Press.

Wuthnow, R. (1998). Loose Connections: Joining Together in America’s Fragmented Communities. Cambridge, Mass.: Harvard University Press.

Yang, P. (1994). “Explaining Immigrant Naturalization.” International Migration Review 28:449-467.

Zhou, M. (1999). “Segmented Assimilation: Issues, Controversies, and Recent Research on the New Second Generation.” pp. 196-211 in The Handbook of International Migration: The American Experience, edited by C. Hirschman, P. Kasinitz and J. DeWind. New York: Russell Sage Foundation.

Appendix One

Appendix Two

Endnotes

1. The mechanisms thought to encourage political incorporation among married couples are likely even more pronounced among families with children residing with them in the United States. These children draw their parents into mainstream institutions, such as schools, hospitals, and social service agencies, increasing contact and social networks with mainstream society. Unfortunately, these data do not allow for an examination of the effect of children on political participation, as information on even the presence of children under age 16 is restricted in the CPS. Certainly a larger study examining the effect of children in the political incorporation process, along the lines of Bass and Casper’s 1999 work, is worthy of being conducted.

2. The construction of this length of eligibility variable is not ideal for a variety of reasons. Firstly, 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 a truly continuous, as the initial categories of entry are of varying lengths. I could simply have used the initial categorical variables to measure length of time in the United States, or collapsed them into a handful of time periods. Doing so would make it difficult to determine the effect of cohort versus period versus length of time in the United States. The length of eligibility variable is also limited by our lack of information on entry status. Many, particularly Mexicans, may have entered the United States as undocumented immigrants. By subtracting 5 years from the date of entry, I am assuming 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, as a result of “starting the clock” on political eligibility too soon.

3. The full main-effects models predicting citizenship and voting for the whole sample can be found in Appendix 1.

4. To calculate the true main and interaction effects, one must determine the variable of interest. I am interested in how the effect of a particular marital status varies across genders. To determine what the effect of being married but separated is for women, I multiply the odds-ratios of “spouse absent” and “female*spouse absent.” The respective odds ratios are .53 and 1.58. The result is .84. Substantively, this means that females who are married but separated from their spouses are 84% as likely to naturalize as women who are married and co-residing. To determine the effect of being male and married but separated, I simply refer to the main effect from the “spouse absent” variable. This is because men score a 0 on the “female” variable, causing the interaction effect of “female*spouse absent” to drop out. The remaining variable is “spouse absent.” So, the odds of naturalizing for a male who is married but separated from his spouse are .53. Substantively, this means that males who are married but separated from their spouses are 53% as likely to naturalize as their married and co-residing male counterparts. Clearly, the effect of being married but separated is larger for men than it is for women, more severely dampening their likelihood of acquiring citizenship.