Migration and Sexuality: A Comparison of Mexicans in Sending and Receiving Communities

Abstract

The reconstruction of sexuality after migration is a central dimension of immigrant health and an integral part of the process of adaptation and incorporation. Despite its significance there is little quantitative information measuring the changes in sexual behavior accompanying migration. This paper contributes to the literature connecting immigrant adaptation and health risks by comparing sexual practices and attitudes among Mexicans in Durham, NC and Mexican sending communities. Consistent with a social constructivist approach to sexuality we show that compared to non-migrants, Mexicans residing in the U.S. exhibit heightened exposure to risk, including casual and, among men, commercial partners. The enhanced risks associated with migration vary systematically by gender and marital status and are accompanied by variation in attitudes towards sexuality, with the U.S. context associated with higher tolerance for infidelity and biological explanations of sexuality. We discuss the implications for immigrant adaptation and health policies in the U.S. and abroad.

In an era of mass migration, understanding the forces that shape immigrant adaptation has become a pressing concern, particularly in the area of public health. One aspect of immigrant health that has become particularly relevant in the context of the AIDS epidemic is the way in which sexuality is shaped by the experience of migration (Herdt, 1997). Recent studies among Latinos have documented a link between international migration, the adoption of sexual risk behaviors, and the spread of sexually transmitted infections, including HIV/AIDS (Mishra, Conner & Magana, 1996; Parrado, Flippen, & McQuiston, 2004). Moreover, widespread cyclical and return migration implies that these risks also affect migrants’ communities of origin. With the number of Latinos in the U.S. infected with HIV continuing to rise (CDC, 2004) and the epidemic spreading to rural areas throughout Mexico and Central America (Magis Rodriguez et al., 2004; Sanchez et al., 2004), these issues are all the more timely and important.

Moreover, while relatively understudied in the literature on immigrant adaptation, the reconstruction of sexual relationships after migration has significant implications for long-term personal development and emotional well-being. This is especially true for those for whom migration entails family separation. In these cases, concerns about infidelity, union dissolution, and even formation of dual families on both sides of the border often mar the migration experience, affecting the overall functioning of migrant families.

Despite its significance, however, there is little quantitative information on the changes in sexual behavior accompanying migration. While a number of studies examine the social and structural context shaping migrants’ sexual behavior (for a review see Organista et al. 2004), research that systematically compares the practices and attitudes of migrants with those of their peers in their communities of origin remains rare. Lack of comparable data from sending and receiving societies clouds our understanding of the connection between migration and sexuality because changes in behavior must be inferred from retrospective accounts or multiple data sources that are not always comparable. There is a particular dearth of information on the sexuality of Mexican men because until recently research on sexuality in the less developed context focused on fertility, which was almost exclusively studied from the female perspective.

Accordingly, this paper contributes to the literature on immigrant adaptation by comparing sexual practices and attitudes among Mexican-born men and women in Mexico and the U.S. Drawing from original quantitative and qualitative data collected in Durham, NC, a rapidly growing immigrant receiving city in the Southeastern U.S., and four sending communities in Mexico (Parrado, McQuiston, & Flippen, 2004), we are able to disentangle sexual practices prevalent in communities of origin from those that arise in conjunction with migration. We concentrate on two dimensions of sexual behavior, sexual initiation and current sexual partners. In addition, we examine attitudinal changes towards sexuality connected with the migration experience. Results illustrate profound changes in sexuality accompanying migration with marked differences by gender and marital status, and highlight the health risks posed by family separation and the male-dominated context in Durham.

Migration and Sexuality in Theoretical Perspective

Rather than driven by natural and biological forces, the very nature of sexuality, including prescriptions regarding when, with whom, and how people may engage in sexual activity, is socially defined and varies over time and across space. Locally prescribed norms regarding with whom, when, where, and how individuals can engage in sexual activity establish boundaries between “good” and “bad,” legitimate and illicit sexualities, and classify certain desires, acts and identities as normal, healthy, and moral while casting others as abnormal, unhealthy, and sinful (Gagnon, 1990; Laumann et al., 1994; Seidman, 2003).

Sexual norms are not universal or uniform within local contexts but rather vary considerably across social and demographic groups. Our approach is to compare sexual behaviors across three main socio-demographic dimensions directly associated with sexuality: gender, marital status, and migration. We argue that it is in the intersection between the three dimensions that the changes in sexual behavior associated with migration can be understood.

Gender is arguably one of the most important organizing dimensions of sexuality, as gender roles and beliefs about the nature of masculinity and femininity translate in different notions of appropriate behaviors for men and women (Villanueva & Buriel, in press). In Mexico the prevailing gender regime is characterized by tolerance of a high degree of male sexual activity, including infidelity, and the restriction of women’s sexuality to the sphere of fertility and reproduction (Gomez & Marin, 1996; Marin & Gomez, 1997; Sadalgo de Snyder et al., 1996, 2000). This also interacts importantly with marital status, which is another key factor that organizes sexual choices. Sexual experimentation is tolerated and encouraged among single men, who tend to experience their first sexual encounter at a relatively early age. Once married, men are thought to commonly engage in sex outside of marriage, both with casual partners and secondary stable relationships, referred to as “casas chicas.” Though the actual extent of these practices is debated (Gutmann, 1996), wives are expected to tolerate these “indiscretions” as part of men’s nature. Single women, on the contrary, are expected to control their sexual desires and arrive virgin to marriage, and sex outside of marriage for married women is outside the boundaries of acceptable behavior and a serious violation of social norms (Hirsch et al., 2002).

And finally, because sexual norms are transmitted and absorbed by local culture and social networks, sexuality is likely to be profoundly influenced by migration. Migration can alter sexuality at the cultural, personal, and structural levels. At the cultural level, the contrast between the more “traditional” culture of migrant sending communities and the more “liberal” sexual ethos in the U.S. could result in more liberal notions about sexual behaviors. This could be especially so for women if migration is associated with greater autonomy and interpersonal power (Hondagneu-Sotelo, 1992). On the other hand, other byproducts of migration may operate in favor of tradition. The marginal position occupied by many migrants can be alienating (Ramakrishnan & Viramontes, in press), encouraging migrants to turn inward and reinforce aspects of their home cultures in an effort to maintain stability and protect their identity from negative perceptions in the host society (Parrado and Flippen, 2005; Parrado, Flippen, and McQuiston, 2005). In this environment, cultural traits such as traditional gender roles, particularly those pertaining to sexuality, could be reinforced as women’s bodies become the site for struggle over disorienting cultural change (Espin, 1999).

At the personal level, migration is a disruptive event that relocates individuals across borders in an unfamiliar environment, dislocating social networks and structures of support. Migration removes individuals from the watchful eye of extended family and community members and weakens social control accordingly. The accompanying sense of anonymity together with the perceived temporary nature of migration may encourage migrants to engage in activity they might otherwise avoid (Organista & Organista, 1997; Viadro & Earp, 2000).

Finally, migration can also affect sexuality via aggregate level structural factors, particularly with respect to the sex ratio. Temporary labor migration from Mexico to the United States has historically been male-centered. While the development of transnational communities and fortification of migrant networks encourages the migration of women, both married and unmarried, the dangers and expense associated with border crossing often perpetuates an uneven gender composition. In new areas of destination such as Durham, the sex ratio is particularly uneven (Suro and Singer, 2000). The implications for sexuality are obvious and multi-faceted, as finding opposite sex partners becomes very difficult for men but relatively easy for women. At the same time, migration also has an impact on sexuality in sending communities, where the sex ratio can be equally unbalanced in favor of women (Hirsch, 2003).

While migration holds the potential to have a significant impact on sexuality, with important implications for public health, research on the topic remains incomplete. Recent national surveys on sexuality lack an adequate sample of migrants, and most studies of migrants tend to focus on condom use, rather than the wider spectrum of sexual behavior, or on migrant farm workers and border regions, which is problematic in light of the tremendous expansion of migration to non-traditional areas of settlement and occupations. Moreover, lack of comparable information on both sides of the border limits the capacity to understand the connection between migration and sexuality. In most cases, cultural values and traditions are inferred from recollection or generalizations drawn by subjects, without actually assessing their presence in countries of origin. Comparable information for migrants and non-migrants is a prerequisite to separate the sexual practices that migrants bring with them from their communities of origin from those that arise in connection with migration.

Our design addresses some of these limitations. We take a bi-national approach that draws on data collected in both sending and receiving immigrant communities. We focus on sexuality more broadly defined rather than condom use and compare sexual behaviors across gender, marital status, and migration status.

Data and Methods

Our analysis integrates qualitative and quantitative methods with a bi-national data collection design. The quantitative analysis draws from 464 face-to-face interviews with a randomly selected sample of Mexican migrants residing in areas of high Hispanic concentration across Durham, NC (333 men and 161 women) conducted between March 2002 and July 2003, and 800 random surveys (400 men and 400 women) conducted in each of four migrant-sending communities in Mexico between December 2002 and April 2003.

Durham, NC is an interesting setting to examine migration and sexuality. Latino migration to Durham is situated within a larger trend of increasing diversity in migrant destinations in recent decades, particularly to metropolitan and rural areas throughout the U.S. Southeast. Growth of the high-tech sector during the 1990s spurred a boom in business and residential construction, heightening demand for semi-skilled laborers, as well as for domestic work and other service employment for the growing class of professionals in the area.

The result of these forces was dramatic: between 1990 and 2000 the Latino population grew from 2,054 to 17,039, or 8% of the total population of Durham. According to the 2000 Census, almost 75% of the Latino population (primarily of Mexican and Central American origins) is foreign born and of recent arrival, with more than 85% migrating to the U.S. between 1990 and 2000. While a large share (44%) of migrants came to Durham via another U.S. location, the majority moved to Durham directly from their countries of origin. Like most areas of new migrant destination, the gender composition of the Latino population is uneven, with 2.3 men aged 20 to 29 for every like-aged woman (Suro & Singer, 2002).

The recent emergence of the Durham Latino community has important implications for the social context of sexual behavior. Latino advocacy organizations are present, but remain small compared to their counterparts in more established immigrant communities. A number of local churches cater to Latinos, and the size of the migrant population is clearly evident in the proliferation of tiendas, taquerías, and mercados throughout the area. Numerous soccer leagues have been formed by migrants, often centered on particular communities of origin, and several bars and clubs cater mostly to migrant clientele. In spite of these organizations and venues Durham migrants frequently complain of a lack of recreational opportunities, particularly those that offer a safe, non-threatening environment for meeting members of the opposite sex.

Another aspect of the local Latino community is the presence of commercial sex workers (CSWs). An extensive literature describes the social environment of migrant farm workers, where CSWs actively solicit at labor camps, bars, and other locations where male migrants congregate, often in accordance with paydays (Mishra, Conner & Magana, 1996; Magana & Carrier, 1991; Bronfman & Minello, 1995). CSWs are also common in areas of Latino concentration in Durham. Street-walking CSWs solicit in areas where migrants congregate, and a number of brothels (or “casas de cita”) operate in and around apartment complexes with large numbers of single male migrants. In addition, groups of women also frequent the apartment complexes, soliciting men gathered in common areas or going door to door in search of clients.

The structural context of Latino migration in Durham thus includes both an uneven sex ratio and the ready availability of CSWs. The implications of these patterns for sexuality are profound, and their importance to public health is underscored by the rapid increase in Latino representation in HIV cases in the area (NC Department of Health and Human Services, 2002).

Surveying the nascent Durham Latino communities raised several methodological challenges. To enhance access to the community and facilitate the collection of sensitive information we relied heavily on Community-Based Participatory Research (CBPR). CBPR is a collaborative approach to research that involves community members in all stages of the research process. In our case 14 community members (6 men and 8 women) were integral to all aspects of the research process, including the conceptualization of the project, data collection, and interpretation of research findings, and helped achieve response rates of 89 and 93% for men and women, respectively (see Parrado, McQuiston, & Flippen, 2005 for more detail).

While growing rapidly, the Durham Latino community remains a small fraction of the total population of Durham, rendering a simple random sample difficult. We therefore employed targeted random sampling of areas of Latino concentration. In collaboration with the CBPR group, we identified 13 apartment complexes and blocks that house large numbers of migrant Latinos, and used the more than 2,100 housing units in them as our sampling frame.

Information was obtained from four Mexican communities representing different areas of out-migration of Durham migrants, and includes two towns in the state of Michoacán and one each in the states of Guerrero and Veracruz. The communities were selected to represent different population sizes and economic conditions. Respondents in each community were selected using random sampling techniques and information from the 2000 Mexican Census.

Respondents in both contexts were administered a semi-structured questionnaire that collected detailed information on demographic, social, and economic characteristics, as well as data on migration experience and family arrangements, including partners’ characteristics and place of residence. In addition, extensive information on sexual practices was collected in both Mexico and Durham. Respondents were asked to report the timing, place, and relationship of their first sexual partner as well as the sexual partners during the previous year. Our analysis centers on how sexual initiation, current sexual partner, and attitudes regarding sexuality differ by gender, migration, and marital status.

Quantitative survey data is buttressed with qualitative data drawn from 8 years of group discussions with the CBPR group as well as 30 in-depth interviews conducted with a convenience sample of 15 men and 15 women representing different migration experiences.

Demographic Composition

The connection between migration and sexuality is systematically mediated by marital status. Mexican migration to the U.S. has long been male-centered and the connection between migration and marital status varies considerably by gender. Mexican men in Durham are significantly less likely than their peers in Mexico to be married, with 38.7% single, divorced, or widowed compared to 30.5% in Mexico. Mexican men in Durham are also dramatically more likely to be married but living apart from their wives. While virtually no married men in Mexico live apart from their wives (less than 1%), a quarter (25.2%) of all men in Durham are married to women who continue to reside in Mexico. Thus, while the majority of adult men in Mexico are married and living with their wives (69%), only 36% of men in Durham are doing so. Moreover, consensual unions, as opposed to formal legal marriage, are more common in Durham than they are in Mexico.

A very different pattern is evinced by women. Women are dramatically less likely to be single in the U.S. than in Mexico, with only 9.2% single, divorced, or widowed in Durham compared to 36.1% in Mexico. The vast majority of Mexican women are married in Durham, 90.8% compared to 55% in Mexico. It is important to note though, that this is not because unattached Mexican women do not migrate to the U.S. In fact, 31% of the women in our sample migrated while single and an additional 14% migrated separated or divorced. Among these women, half subsequently formed a union in the U.S. In Mexico the phenomenon of married men migrating without their wives translates into 8.9% of all women in sending communities being unaccompanied married women. As was the case for men, consensual unions are significantly more common in Durham than in Mexico.

This intersection of gender, marital, and migration status provides the background for understanding sexual behavior. The significant differences in marital status by place of residence have profound implications for sexual behavior and health risks, especially for men. The next set of analyses documents these differences focusing on sexual initiation and current partners.

Sexual Initiation

One of the ways that migration can affect sexuality is through sexual initiation. First intercourse is a formative experience central to reproduction and transition to adulthood in most cultures. Moreover, at least since Kinsey’s landmark study of sexuality it has been widely appreciated that the system of sexual socialization under which one grows up and attains maturity is of great importance to understanding the structure of their sexuality throughout the life-course (Laumann et al., 1994). Factors such as timing and partner at first intercourse can have a long lasting impact on socio-emotional functioning and may structure intimate relationships later in life (Udry & Campbell, 1994).

Using retrospective information we separate patterns of sexual initiation according to whether men and women had ever migrated to the U.S. Results, reported in Table 1, show that for both men and women ever-migrants average younger ages of sexual initiation than their non-migrating peers. The median age at sexual initiation is 3 years younger (17) among ever-migrant men compared to non-migrants (20). A smaller difference is found among women, among whom the median age at sexual initiation is 18 and 20 among ever-migrant and non-migrant women, respectively. As could be expected, by age 35 the proportion sexually initiated across groups tends to converge, though women remain more likely than men to report never having had sex.

Table 1.

Sexual Initiation by Gender and Migrantion Status

	MEN			WOMEN
	Non-migrant	Ever Migrant		Non-migrant	Ever Migrant
Age at sexual initiation
25%	18	16		18	17
50%	20	17		20	18
75%	22	19		24	21
Percent not having had sex by age 35	0.8	1.1		2.9	2.3
Proportion sexually inititated in the U.S.a		15.7			18.5
N	307	426		376	176
	Initiated in Mexico		Initiated in U.S.	Initiated in Mexico		Initiated in U.S.
	Non-migrant	Ever Migrant		Non-migrant	Ever Migrant
	(1) (ref)	(2)	(3)	(4) (ref)	(5)	(6)
Partner at sexual initiationa (%)
Wife/Husband	44.8	22.5 **	24.6 **	77.7	71.8	75.8
Girl/Boyfriend	38.1	46.8 **	33.9	20.6	24.7	18.1
Casual partner	13.2	20.8 **	20.0	1.7	3.5	6.1
Sex Worker	3.9	9.9 **	21.5 **
Used contraception at initiation(%)	36.2	32.7	56.9	13.7	13.1	33.3
Used condom at initiation (%)	93.6	83.5	100.0	56.1	76.2	83.3
N	257	347	65	293	145	33

^**Difference with never-migrants (columns 1 and 4) statistically significant p<.05

^aAmong those ever having sex

While some of these differences could relate to the selectivity of the migrant flow (migrants differ somewhat from non-migrants with respect to factors such as rural origin, educational attainment, and employment background),the size of the disparities strongly suggests the role of migration and context in structuring behavior. For most men and women, the timing of migration tends to coincide with a myriad of other life-course transitions, such as school completion, first employment, and also sexual initiation. In fact, results reported in Table 1 show that 15.7 and 18.5% of ever-migrant men and women, respectively, report being sexually initiated in the U.S. These figures are noteworthy because place of initiation could be associated with other important aspects of initiation, particularly in light of the unbalanced sex ratio among migrants.

To investigate these issues, Table 1 also presents differences in type of partner and contraceptive use at initiation by migration and marital status. The analysis first distinguishes between place of initiation (U.S. or in Mexico) and among those initiated in Mexico between ever-migrants and non-migrants. The rationale is that non-migrants (column 1 for men and 4 for women) constitute a reference group against which we can assess the connection between migration and partner and contraceptive use at initiation.

Overall, results show that migration correlates significantly with initiation partner among men but not among women. Column 1 in Table 1 shows that sexual initiation with a spouse is the most prevalent form of sexual initiation among non-migrant men in Mexico (44.8%). The figure is significantly lower among return migrants and those initiated in the U.S., 22.5 and 24.6%, respectively. The share of men reporting initiation with a girlfriend is not significantly different among non-migrant men and those initiated in the U.S.; 38.1 and 33.9%, respectively. However, the percentage is significantly higher among return migrants sexually initiated in Mexico, 46.8%. The same applies to initiation with casual partners, which is significantly higher among migrants than among those who remain in Mexico.

These results reflect the disruptive effect of the migration experience. For single men in Mexico, migration to the U.S. can be a viable strategy for capital accumulation and formation of an independent household and studies have shown that remittances and savings do in fact facilitate union formation among men in Mexico (Parrado, 2004). At the same time, though, migration often results in the breakup of early relationships that might otherwise have resulted in marriage. The duration of a trip to the U.S. is always uncertain and plans and relationships change over time. One of our in-depth interview subjects recounted that he had a serious girlfriend in Mexico but saw no economic opportunities that would allow him to establish an independent household. He decided to come to the U.S. for a while to accumulate capital so that they could marry, but after 2 years in the country he suspected that his girlfriend was dating somebody else and terminated the relationship. The greater incidence of initiation with a girlfriend, rather than a spouse, associated with migration highlights one of the many ways through which a trip to the U.S. positions men back into the dating market.

Particularly relevant for HIV risk is the dramatic association between migration and sexual initiation with a commercial sex worker (CSW). While only 3.9% of non-migrant men reported that their first partner was a CSW, the figure was 9.9% among return migrants initiated in Mexico and a full 21.5% among men initiated in the U.S. The combination of an unbalanced sex ratio with the sexual geographies of the communities in which men live, including weakened informal social controls, loneliness and a lack of other avenues for warmth and emotional attachment, contributes to a strong association between place of initiation and CSW use. Moreover, the more common use of CSWs in the U.S. appears as an important factor accounting for the younger sexual initiation among migrant men discussed above.

As is always the case with retrospective information, caution should be taken when drawing conclusions from these results because individuals could potentially reconstruct their sexual histories according to their current status. For instance, married men may underreport sexual initiation with a CSW so that their histories conform more closely to the idealized image of marriage. To further investigate the validity of these differences we tabulated partner at initiation using data only for men in the U.S. Results (not reported) show that 12 and 27% of those initiated in Mexico and the U.S., respectively, reported doing so with a CSW. If we further limit our sample to single men at time of the survey, i.e., men with the lowest pressure to reconstruct their histories, results show that 16 and a dramatic 40% of those initiated in Mexico and the U.S., respectively, reported doing so with a CSW.

Together, these findings support the higher likelihood of sexual initiation with a CSW in the U.S. compared to Mexico. Thus, the health implications between migration and sexual initiation are numerous, especially since the efficiency of condom use tends to be compromised during first sexual encounters and early CSW use may establish a predisposition for the practice that endures, presenting an ongoing risk to migrants and their future partners.

The experience of one of our interviewees provides a more nuanced understanding of this connection. Joaquín migrated to the U.S. with his parents when he was 15 and attended middle school for a short time before leaving school to work in gardening with his father. Even though he came to the U.S. at a relatively young age and had a girlfriend in school, when asked about whether it was easy to form relationships in the U.S. he replied,

“Really, I don’t think so. Because, one way or another it’s easier in Mexico. Because there you know more people, more women. You relate with more people and so I think it could be easier to get involved with people, because there’re more people you could meet. And here, it’s just not the same.”

At around age 18 he and a friend were both initiated with sex workers. As he described it:

“One night, we were out, you know, hanging around. And my friend knew where to go. I guess he heard it from some other friends, or I don’t know. These things are just known, between friends, one passes it along to the other. …And, once I met a guy who was out passing out cards, too. They don’t put on (the card) that it’s a prostitute or anything like that. They put it another way – like with a phone number and it says massage services or photography.”

Among women, a different pattern is evident. Almost 75% of Mexican women report being sexually initiated with their husbands (including women who were initiated with a boyfriend who they married) and this holds for those initiated in Mexico and the U.S. This figure is similar to that seen in national samples, such as the 2003 National Survey of Reproductive Health (Encuesta Nacional de Salud Reproductiva). This figure could be biased upwards if married women feel pressure to conform to perceived societal expectations of virginity at marriage. To minimize this bias, we restricted our sample to separated women who may feel less pressure to report their husbands as their first partner since they are no longer together. Results show that even among this group 76% report being sexually initiated with their first spouse. So, the pattern of sexual initiation with a spouse, either before or at the time of marriage, appears as a feature of Mexican female sexuality that holds regardless of place of initiation.

Table 1 also shows that a smaller share of women initiated in the U.S. report doing so with a boyfriend. At the same time, female migrants are more than twice as likely to be initiated with a casual partner as their peers in Mexico, although at 6% the overall proportion remains low. Thus, for a small proportion of women the U.S. context of less constraining social control and favorable sex balance is associated with an increased prevalence of casual partnering.

Current Partner

While sexual initiation can have long lasting effects, it is current sexual partners that relate more immediately to health risks and have numerous implications for union formation and stability, fertility, and public health. As documented in Table 2, differences in current partner associated with migration are stark, with considerable variation by marital status and gender.

Table 2.

Sexual Partners in Previous Year by Gender, Migration and Marital Status

	In Mexico					In Durham
	Marital Status
	Single/ Sep/Div	in Union		Partner in U.S.		Single/ Sep/Div	in Union		Partner in Mexico
Men
Sexual relationship type
No partner	51.6	---		---		29.7 **	---		41.7
Stable Girlfriend    /Casa Chica	31.5	2.5		---		8.6 **	0.0		7.1 **
Casual Partner	22.6	8.0		---		29.7	5.8		17.9 **
Sex Worker	4.8	6.2		---		48.4 **	7.4		39.3 **
N	124	276		0		128	121		84
Women
Sexual relationship type
No partner	81.9	---		97.1		48.0	---		100.0
Stable Boyfriend	13.4	0.0		0.0		32.0 **	0.0		0.0
Casual Partner	5.4	0.0		2.9		28.0 **	5.9		0.0
N	145	220		35		14	137		1

Note: The categories are not mutually exclusive, so the sum of percentages could exceed 100. As all respondents in co-residing unions also reported having had sex with their partners, this practice is not reported.

Among single men, results show that those in the U.S. are less likely than their peers in Mexico to report no partner in the previous year, 51.6 relative to 39.7%, respectively. This pattern of greater sexual activity among single men in Durham is consistent with the higher incidence of sexual initiation among ever migrant than non-migrant men. This higher likelihood of sexual activity is accompanied by variation in partner type across contexts. First, fewer single men report having a stable partner during the previous year in Durham (8.6%) than in Mexico (31.5%). Second, single men in Durham are ten times more likely than their Mexican counterparts to report a CSW partner in the last year, 48.4 and 4.8%, respectively.

Similar patterns are seen among unaccompanied married migrant men in Durham. Compared to their single counterparts, a higher proportion of unaccompanied married men reported not having a non-spouse sexual partner during the last year, 41.7% to 29.7%. However, 39.3% reported sex with a CSW and 17.9% reported a casual sexual partner. In qualitative interviews, men often asserted that for unaccompanied married men, sexual relations with CSWs were part of men’s biological need for sex and thus did not necessarily constitute cheating. However, casual encounters were viewed as having the potential to lead to more permanent and romantic relationships that would violate the expectations of fidelity associated with marriage, and thus were viewed as socially more taboo than CSW use.

In any event, the implications of these trends for HIV risks are ominous. The structural forces framing Mexican migration to Durham, including family separation, the unbalanced sex ratio, reduced social and familial controls, and impediments to connecting with the broader dating market are among the main sources of these differentials.

Table 2 also shows that the sexual behavior of married men living with their spouses in Mexico and Durham is not that different. In addition to their spouse, 8.0 and 5.8% of married men reported having sex with a casual partner, and 6.2 and 7.4% reported encounters with a CSW in Mexico and the U.S., respectively. Thus the 10% of married men with extramarital partners should be considered when evaluating HIV risks.

Of particular interest are reports on the prevalence of secondary households or “casa chicas” among married men in Mexico and Durham. Results show that 2.5% of married men in Mexico reported having a casa chica. In Durham, none of the married men residing with their partner reported an additional stable relationship. However, among unaccompanied married men a full 7.1% reported a second stable partner. It is important to note that the possibility of migrant husbands forming a second union in the U.S. and abandoning their families in Mexico is a common concern among wives remaining in their communities of origin, and in in-depth interviews was a frequently cited impetus for the migration of married women. Again while we do not find the practice to be common, women’s fears are not unwarranted and formation of second households among unaccompanied men appears as an important by-product of migration.

The experience of Eliseo illustrates one of the avenues through which the process of forming a second household in the U.S. evolves. Eliseo married young in Mexico and migrated to the U.S. to support his family after his second child was born. He’d come every year around April and return in December. Over several years of this cycle he had two more kids. Several years ago, though, his migration strategy changed. On one of his trips to the U.S. he started dating a Mexican woman from a neighboring town who had come to Durham with her sister.

“I grew up one hour away from her home town…so friendship, going out to dinners, going here and there…and well it happened whatever happened. So everything was alright. I never lied to her. I told her, ‘I’m married, I have my kids and my wife.’ But she thought that if she was pregnant I would be more under pressure, more tied up. But I told her, when she got pregnant, ’I’m going to help you. I’m going to be here for you as much as you need me. But I’m not going to marry you. I can’t marry you. I’m already married.’”

When we spoke to Eliseo he was still contemplating returning to his family in Mexico but did not want to leave because his daughter in Durham was still too young. He told his wife in Mexico that he had a daughter in the U.S. but his wife did not know that he was living with another woman. When asked about his Mexican wife’s reaction to the situation he said:

“Well, she has to put up with it. What is she going to do? We are doing the best we can, we accommodate things the best we can. (The important thing is that) I’m still sending her money.”

He also considered bringing his U.S. partner back with him and supporting both families in Mexico. He said he could not see a better solution; even if he separated from one wife he would still have to support her, so he did not see the merit in doing so. When asked whether he was worried that one of his partners would see somebody else he said that he told both women:

“Everything will be the same as long as I don’t know that you have someone else. Not because I’m going to make problems for you or anything like that, it’s just that I will stop giving you money for clothes, rent, and other things and just give you money for the children. If you have someone else you should never let me know, you should not tell me, and I will keep helping you. I’m giving you the chance to do what you want, I just don’t want to think of myself as a sucker (verme la cara de pendejo).”

The attitude of not wanting to know about infidelity is a recurrent theme permeating descriptions of the migration-sexuality connection among both men and women. The marginal position of migrants in the U.S. contributes to a sense of fatalism and the attitude that many important things in life are out of one’s hands. Our CBPR group indicated that in the context of migration, the old adage “ojos que no ven, corazón que no siente” (out of sight out of mind) might be a defense mechanism against the pervasiveness of infidelity.

Mexican women report diverse patterns of current partners by migrant status, although differences are most pronounced among single women. Results in Table 2 show that similar to men, single women are less likely to report no partner in the past year in Durham relative to Mexico, 48 compared to 81.9%. Results show that the likelihood of sexual encounters with both stable and casual partners is higher among single women in Durham relative to Mexico. The relatively low level of reported sexual experimentation among single women in Mexico is consistent with the high prevalence of sexual initiation with spouses. At the same time, results show that in Durham the small fraction of Mexican women who remain single might be at an enhanced risk of HIV transmission, as over a quarter (28%) report a casual partner during the previous year, relative to a mere 5.4% among single women residing in Mexico. Either because they are a highly selective group or because the unbalanced sex ratio and social context in Durham facilitates women’s sexual experimentation, this group should also be taken into consideration when understanding the connection between migration and HIV transmission.

The experience of Aurora documents this connection. Aurora’s husband had been in the U.S. for over 5 years and had for all practical purposes abandoned her and their two children. She came to Durham to live with her aunt because it was the only way she felt she could provide for her children in Mexico, who were now living with her mother. When asked about whether it was difficult to find a partner in the U.S. she said:

“No, it is very easy to find someone to be with. What is difficult is to find someone with the same interests, someone responsible. A lot of men only want to have fun, have a good time, and they don’t want any responsibilities. Right after arriving to Durham I met (and was involved with) a guy who had AIDS and after a few months he died. I did not know he had AIDS and I was worried because my idea was not to come to the U.S. to get sick.”

While Mexican women’s concerns about male infidelity associated with migration are documented, less attention has been paid to married men’s concerns regarding their spouses’ sexual behavior. As discussed in Eliseo’s case, many unaccompanied married men are concerned that their wives may be unfaithful within their communities of origin. Likewise, in the U.S. the highly unbalanced sex ratio coupled with complex household arrangements that in many cases involve married couples sharing apartments with other adult males also triggers concerns over women’s infidelity. In fact, 2.9% of unaccompanied married women in Mexico reported a casual partner during the previous year, as did 5.9% of women in stable unions in Durham.

Attitudes towards Sex and Sexuality

The importance of migration to sexuality is also pronounced when it comes to attitudes and beliefs, as seen in Table 3. According to the common portrayal of Mexican culture, sexual experience prior to marriage should be viewed favorably or at least neutrally for men but seriously frowned upon for women. Indeed, respondents in both Mexico and the U.S. are roughly twice as likely to report that it is a good idea for men to have a lot of sexual experience before marriage as they are to report the same for women. However, U.S. migrants are significantly more tolerant of premarital sexual experience than their peers in Mexico. Specifically, both men and women in Durham are significantly more likely to believe that men should have premarital sexual experience. Migrant women, but not migrant men, are also more likely than their peers in Mexico to believe that premarital sexual experience is good for women.

Table 3.

Attitudes Toward Sexuality by Gender and Current Migration Status

	Men		Women
Percent agree:	Mexico	Durham	Mexico	Durham
Gender attitudes
It’s a good idea for men to have a lot of sexual experience before they get married.	45.3	60.4 **	29.8	47.2 **
It’s a good idea for women to have a lot of sexual experience before they get married.	24.0	24.6	11.8	32.3 **
It is the woman’s responsibility to prevent pregnancy.	21.5	29.7 **	42.5	33.5 **
It is a woman's responsibility to prevent sexually transmitted diseases.	18.8	30.0 **	42.5	43.5
If a man has an extramarital affair, his wife should just keep quiet and accept it.	17.0	20.1	3.3	10.6 **
If a woman has an extramarital affair, her husband should just keep quiet and accept it.	2.3	12.6 **	2.0	9.3 **
Women don’t need to have sex as much as men do.	24.5	37.8 **	20.8	42.9 **
Condom use
If a condom is not handy, I would have sex anyway	51.5	37.8 **	61.0	80.8 **
If I asked my partner to use a condom, she would think I had a disease	43.3	55.3 **	28.5	38.5 **
Condoms are only for sex with prostitutes	44.3	29.7 **
It is difficult to buy condoms where I live	9.0	14.1 **	12.8	34.2 **
It is embarrassing to buy condoms	43.5	21.6 **	54.5	45.3 **
Condoms are expensive	30.5	20.1 **	37.5	48.5 **
You feel less pleasure when you use a condom	60.5	70.0 **	66.3	79.5 **
Condoms are tight and uncomfortable	57.3	69.1 **
If a woman carries condoms I would think she is loose sexually	58.8	54.7	71.2	72.7
N	400	333	400	152

^**Difference between Mexico and Durham significant at p<.05

Evidence of attitude change is also seen in responses to whether or not it is women’s sole responsibility to prevent pregnancy and STDs. First, Mexican men are less traditional in this respect than Mexican women, with the latter more likely to believe it is a woman’s responsibility to prevent pregnancy and STDs in both contexts. Migration has the opposite effect, rendering men more traditional and women less so with respect to pregnancy and STD prevention.

Migration, which often entails extended periods of family separation, is also strongly associated with attitudes towards marital infidelity. The reality of forced separation leads both men and women to become more tolerant of husband’s extramarital affairs in the Durham setting. This is not surprising given the portrayal of Mexican culture as permissive of male infidelity in general. However, results show that both men and women are also more tolerant of wives’ infidelity in the context of migration. Mexican men in Durham, for example, are more than 5 times more likely than their counterparts in Mexico to report that if a woman has an extramarital affair her husband should just accept it. Comparable figures for Mexican women are similarly dramatic. Thus, Mexicans recognize that the reality of forced separation entails certain sacrifices with respect to fidelity for both male migrants and their wives left behind.

This finding is reinforced by the next item, which reports attitudes towards whether women need sex as much as men do. Likely in response to the reality of infidelity on the part of male migrants, both men and women are more likely to believe in biological explanations of sexual need in Durham as they are in Mexico.

Attitudes towards condoms undergo a radical transformation in Durham, again in ways that are suggestive of the issues highlighted above. The proportion of respondents who said that if they asked to use a condom their partner would suspect they had an STD is substantially higher among men and women in Durham, reflecting the higher incidence of infidelity and CSW use among migrants. Likewise, the proportion of men who report that condoms are only for sex with CSWs falls substantially in Durham, where they are likely to receive AIDS prevention information and where non-commercial extramarital relations are also more common.

Relative to Mexico, commitment to use condoms is higher among men but lower among women in Durham. This suggests growing awareness of the importance of condoms among men, but the more traditional pressures women feel in sexual relations in Durham.

The effect of migration on condom access is mixed, with men and women in Durham more likely than their Mexican peers to report that it is difficult to buy condoms, but fewer reporting that condoms are expensive or embarrassing to buy. Perhaps reflecting their greater experience with condoms in the U.S. than in Mexico, migrants are more likely to report that using condoms results in less pleasure and that they are tight and uncomfortable.

Conclusions

The continuous and increasing flow of Mexican migration to the U.S. has raised concerns about prospects for immigrant adaptation. One important yet overlooked aspect of immigrant incorporation is the reconstruction of sexual relationships after migration. The impact of migration on sexuality has important implications not only for public health, including the spread of STDs such as HIV, but also shapes immigrants’ emotional and familial well being. In this paper we followed a social constructionist perspective on sexuality that views sexual behaviors as embedded within the norms and context in which they take place. Our analytical strategy was to disentangle the role of migration and context by comparing the sexual practices of men and women in Durham, NC and sending communities in Mexico. Our analysis focused on understanding how sexual initiation, current sexual partner, and attitudes towards sexuality differ among Mexicans in Mexico and Durham, and how these differences vary by gender and marital status. The main expectation was that the relocation of persons across borders would profoundly alter patterns of sexual behavior, heightening migrants’ exposure to HIV risks and behaviors that undermine family stability, and influencing gender attitudes pertaining to sexuality.

Results corroborate differences in sexual behaviors that vary systematically by gender and marital status. Migration is associated with an uneven sex ratio, and men in Durham are less likely to be married and living with their spouse than are men in Mexico. Among women differences are more modest, with most women in both contexts in stable unions.

With respect to sexual initiation, migration is associated with a younger age at this important transition to adulthood for both men and women. Moreover, among men those initiated in the U.S. are less likely to be initiated with a spouse, due to the disruptive effect of migration which truncates many relationships that may have otherwise ended in marriage, and dramatically more likely to be initiated with a CSW compared to their non-migrant peers.

Migration is also closely associated with current partner behavior. This is especially the case for the use of CSWs which is relatively rare in Mexico but a common and generalized practice among migrants in Durham, not only among single men but also among unaccompanied married men whose wives continue to reside in Mexico. The flip side of this change is that single men in Durham are significantly less likely to have a stable girlfriend than their peers in Mexico. At the same time, a considerable proportion of unaccompanied married men form secondary unions in Durham, undermining union stability and the healthy functioning of migrants’ families.

Patterns are different among women. Overall, women’s behavior did not register dramatic change, highlighting that women’s exposure to sexual risks is mainly a by-product of their partners’ behaviors both in the U.S. and in Mexico. Only one group of women, namely single/separated women in the U.S., shows a pattern of migration-related change. This group tends to be overlooked in discussions of migration and sexual risks mainly because of their relatively small representation in the migration flow. However, the high prevalence of casual partnering among this group and their potential for acting as bridges connecting high and low risk groups suggests the need for targeting this population in public health interventions.

These differences by context also affect formative experiences and attitudes towards gender. Results documented a higher prevalence of sexual initiation with sex workers among men in the U.S. than in Mexico and a higher tolerance for infidelity, biological explanations of sexuality, and compliance with more rigid definitions of gender roles among both men and women in the U.S. than in Mexico. So, to the extent that behaviors connect with socialization processes, differences between Mexico and the U.S. can result in long-term changes in sexuality.

Arguably, the main implication of our study is that context matters. By and large the Mexican men and women in our study are comparable across contexts. While accounting for all the personal demographic and social characteristics affecting behaviors was beyond the scope of this analysis, the dramatic disparities in behaviors between Durham and sending communities suggests that individual-level personal characteristics alone cannot account for the large effects. Instead, results imply that processes connected to the dynamics of migration and particular characteristics of the context of reception lie at the root of these differences. Factors such as the male overrepresentation in the migration flow, the prevalence of family separation, and the lack of social support and control result in a context more conducive to sexual risks in the U.S. than in Mexico. At the same time, other particular conditions of the receiving community, such as the ready availability of CSWs and limited number of local organizations connecting migrants with the larger community, also contribute to the proliferation of health risks.

These contextual factors cannot easily be affected through local public health interventions. They are connected with larger dimensions of immigration policies or community structure. While politically unpopular, an emphasis on family reunification in immigration legislation could reduce health risks among immigrants and their communities by ameliorating the unbalanced sex ratio and reconstructing familial channels of social control. In the absence of such familial mechanisms of informal control, the structure of the community becomes increasingly important for health policy. The challenge is to recreate a context that would stimulate positive health outcomes (Coleman, 1993). This requires moving beyond individual informational campaigns to “structural interventions” aimed at altering the social disorganization affecting newly emerging immigrant communities.

Biographies

Emilio A Parrado is Associate Professor of Sociology at the University of Pennsylvania department of Sociology. His area of specialization is social demography, with particular emphasis on international migration, immigrant adaptation, family and fertility behavior, and health in Latin America and among Latinos in the U.S.

Chenoa A. Flippen is an Assistant Professor of Sociology at the University of Pennsylvania. Her primary research interests relate to racial and ethnic stratification, including such diverse topics as minority aging, inequality in wealth accumulation, the impact of residential segregation on minority housing, and Hispanic immigrant adaptation.