The Sexual Organization of the Citye

"In The Sexual Organization of the City Laumann and his team of researchers use the data from a 1995 survey of four Chicago neighborhoods to map out the mating habits of urbanites. . . . Neighborhoods play a critical role in how we choose partners for brief sexual encounters and long-term engagements."—Chicago

Please note that tables and footnotes do not appear in this online version of the chapter.

See also:

Chapter 1: The Theory of Sex Markets

A webpage for the Chicago Health and Social Life Survey


An excerpt from
The Sexual Organization of the City
Edward O. Laumann, Stephen Ellingson, Jenna Mahay, Anthony Paik, and Yoosik Youm, editors

Chapter 2
The Chicago Health and Social Life Survey Design
Martha Van Haitsma, Anthony Paik, and Edward O. Laumann

In 1992, researchers at the University of Chicago fielded the National Health and Social Life Survey (NHSLS), a national probability survey documenting how individual sexual expression, attitudes, and behaviors varied systematically across major social categories (gender, age, religion, education, marital status, race/ethnicity) of American society. The core findings of that survey were reported originally in two studies—The Social Organization of Sexuality (Laumann, Gagnon, Michael, & Michaels, 1994) and Sex in America (Michael, Gagnon, Laumann, & Kolata, 1994)—published at a time when the last, comprehensive scientific study of American sexuality to have been conducted was the two-volume Kinsey Report (Kinsey, Pomeroy, & Martin, 1948; Kinsey, Pomeroy, Martin, & Gebhard, 1953) and when the threat of HIV/AIDS was a mounting public health concern. A follow-up volume, Sex, Love, and Health in America (Laumann & Michael, 2001), presented a more detailed exploration of the NHSLS data set, focusing on the social and cultural contexts that frame such disparate topics as abortion, adult-child sexuality, circumcision, sexual dysfunction, modes of sexual expression, and changes in America's sexual mores in the wake of the Sexual Revolution of the late 1960s.

The NHSLS, and other studies also released in the 1990s (e.g., Catania et al., 1995; Tanfer & Schoon, 1992), provided critically needed baseline information about adult sexuality: who does what with whom, how often, and why. However, meeting and mating are fundamentally local processes, not activities that occur at the level of abstract, analytic demographic characteristics. The 1995-97 Chicago Health and Social Life Survey (CHSLS), the follow-up study to the NHSLS that provided the data on which The Sexual Organization of the City is based, was designed to better assess how people meet partners and organize their sexual relationships. This chapter provides an overview of our research strategies, describes the CHSLS data, and highlights some of the distinctive features of sexual expression in Chicago.

It is our immodest, wishful aspiration that the picture of urban sexuality presented by The Sexual Organization of the City replace that presented by such media representations as the television show Sex and the City, which portrays the sexual adventures of four professional, white women living in Manhattan. While the world of Sex and the City is decidedly narrow, it is at the center of several demographic trends (which no doubt have been thoroughly assessed by market researchers): upwardly mobile white women marry later, divorce more frequently, and have fewer children than do other women. In contrast, our study looks at sexuality as it is expressed across an entire city population. Many of our respondents are poor, are members of racial/ethnic minorities, and are unlikely to come into contact with the rarefied world portrayed in Sex and the City. The Sexual Organization of the City is about the diversity of sexual expression found in an American metropolis.

Cities are important places to study because they are at the locus of major societal patterns. Georg Simmel ([1903] 1971), and Robert Park ([1929] 1967) afterward, argued that the city was at the heart of modern life. Economic restructuring and racial/ethnic segregation are social processes affecting the whole United States throughout the twentieth century and into the twenty-first (Wilson, 1987; Massey & Denton, 1993), but they are considerably more dynamic in cities than in rural and suburban settings. Cities are demographically heterogeneous. They are the ports of entry for immigrants; they have high concentrations of racial and ethnic minorities. Indeed, in some cities, such as Los Angeles, white Americans are no longer the majority ethnic group. Young people, the nonmarried, and individuals with same-sex orientations tend to congregate in cities, making urban areas the most vibrant of all settings in which people search for sex partners. Because cities are also the unfortunate hosts to many social problems—poverty, homelessness, out-of-wedlock fertility, violence, drug abuse, etc.—they are important places to study if we are to understand the regulatory role of collective processes (e.g., social capital) and local institutions (Sampson, 2002). In short, cities are where the action is, and, given its variegated urban social processes, Chicago in particular is an exemplar.

Sex in the city, then, takes place in an environment where search activities are heightened and potential opportunities are stunningly diverse. Sex is an important activity of many city dwellers, especially those who are not married. A considerable amount of socializing among urbanites is oriented toward meeting potential sex partners. The demographic heterogeneity and density of the city also allow for diverse sexual lifestyles and the possibility of mixing with socially distant others, both alternatives that are largely unavailable to many rural populations. At the same time, many people have crystallized views of whom they want as a sex partner and tend to share their preferences with those who are socially similar, creating ecological niches of sexual matching patterns in the city. Because of their large concentrations of men who have sex with men, American cities are also at the heart of one of the key social problems associated with sexuality: the transmission of HIV/AIDs and other sexually transmitted infections (STIs). Finally, there are many collective-level actors and institutions that must deal, directly or indirectly, with sexual expression and its concomitant problems. Ultimately, then, while a study of urban sexuality cannot claim to be representative of the sexuality of a nation as a whole, it does allow us to study sexual expression in an environment of diverse social contexts.

To capture these multiple urban social processes, we designed the CHSLS to be representative yet comparative and to have individual- and institution-level components. We began with the premise that sexual partnering occurs within real social groups and networks, not within analytic social categories. Sex partners meet each other because they attend the same church, belong to the same professional organization, live in the same neighborhood, work at the same firm, patronize the same bar, have mutual friends, or somehow share social and physical space at some point in time. In other words, sexual-partnering opportunities are heavily structured by the local organization of social life, the local population mix, and the local norms guiding the types of relationships that are sanctioned or supported. Thus, we adopted a multisample strategy to facilitate comparative analyses and a mixed-methods approach to capture both individual and organizational aspects of local communities. In the next section, we go into greater detail regarding the study design of the CHSLS, how we fielded the survey and key-informant interviews, and the design of the questionnaire.

Study Design

In a national survey, it is impossible to investigate detailed aspects of social context because respondents are spread too widely among locales with different norms, opportunities, and supports for sexual partnering and expression. The CHSLS looks at local communities with distinctive institutional configurations, local histories, and networks of social interaction. It emphasizes the social processes that produce intergroup differences as well as intragroup variation. What sexual behaviors occur, between whom, and how often are channeled by the social pressures and opportunities of the immediate social environment in which people live. The CHSLS was designed to uncover these social forces.

The research design itself consisted of two components: a household survey and key-informant interviews. The probability data are based on a representative, face-to-face household survey, conducted in five samples of persons aged eighteen to fifty-nine. The CHSLS utilized a similar methodology as the NHSLS in terms of mode of administration and target population. However, unlike the NHSLS, interviews were conducted in Spanish when necessary, and interviewers used programmed laptops rather than printed questionnaires to allow for complex skip patterns and self-answered questions. In total, we collected 2,114 cases. One sample, which included the city of Chicago as well as the inner suburban ring, was based on a two-stage survey representative of residents of Cook County. The remaining four samples were each household surveys representative of a community area within the city of Chicago. We also conducted key-informant interviews with institutional actors within each sampled community area. We completed a total of 160 interviews with community leaders and service providers within the four community areas. A limited amount of general observation of community life and events was also undertaken in each neighborhood in order to better characterize the local social milieu. The various types of interviews and samples were intended to produce results that would both stand alone and shed light on each other.

Choosing the Oversample Neighborhoods

Four focal communities were selected to provide a set of comparative case studies. These communities were chosen to represent various points along a continuum of perceived cohesiveness and insularity as well as to provide samples of different racial, ethnic, and sexual-orientation populations. The selection process involved three stages: choosing target neighborhood types; identifying general community areas that represent those types; and finalizing the choice of community areas to be studied.

As ours was a study of sexual expression and health, we focused on populations relevant for public-health policy: gay men; African Americans; and Hispanics. Gay men, African Americans, and Puerto Ricans have disproportionately high rates of HIV and other STIs. The Mexican American population in Chicago is also of special interest to public-health providers. This rapidly growing population includes many needing Spanish-language services as well as a subgroup of illegal immigrants who are often reluctant to access government programs.

Furthermore, Chicago is one of the few urban areas where sizable Puerto Rican and Mexican populations reside, allowing separate studies of both groups. Other studies have shown the obfuscation that results when Hispanics of different national origins are lumped together, particularly when they live in different parts of the country. Thus, it seemed important to take advantage of Chicago's population mix by selecting two different Hispanic groups for comparison.

The choice of community areas rested on several factors. The primary criterion was the relative degree of group concentration. An area characterized by its gay-male concentration was selected first because there is only one such neighborhood in Chicago. This neighborhood is interesting because it is also a singles' scene, drawing people from around the metropolitan area to its restaurants, clubs, theaters, and other social venues. There are a few community areas with a predominance of Mexicans, several contiguous neighborhoods where Puerto Ricans are concentrated, and numerous neighborhoods that are more than 95 percent African American. According to the 1990 census, the two most concentrated community areas of Mexican residence are each about 85 percent Mexican, but the most concentrated area of Puerto Rican residence is only 33 percent Puerto Rican.

The second and third criteria were avoiding areas of extreme poverty and attaining some degree of class comparability among the Mexican, Puerto Rican, and African American neighborhoods. The CHSLS was not meant to be a study of poverty. Although health and poverty are related, we anticipated that the effects of severe-poverty concentrations would dominate other local factors that would be of general interest and have policy relevance for sexual expression. We also hoped to avoid confounding race/ethnicity with class. The preselected area of gay concentration is one of high income and high levels of education, but there are no comparable neighborhoods with Puerto Rican or Mexican concentrations. There are middle-class black neighborhoods in the city, but these are not typical of the black urban population as a whole. We thus chose to look for working-class black and Hispanic areas. Class matching proved difficult in the case of blacks and Hispanics, however, because class markers are distributed so differently within each group. In the heavily immigrant Mexican neighborhoods, incomes are clustered around the poverty line, and education levels range from extremely low to average, but employment is high, and welfare use is low. In black neighborhoods with similar median-income levels, the dispersion of income is much wider, and education levels are higher, but unemployment and welfare reliance are greater as well. Puerto Rican neighborhoods are similar to African American neighborhoods in some respects and to Mexican neighborhoods in others. Because statistical measures of class are not entirely comparable between black and Hispanic areas, we relied on observation as well as census figures to locate suitable neighborhoods. We looked for areas with some on-the-ground similarity in terms of street life, local business activity, and residential housing stock.

Finally, we wanted neighborhoods that had some sense of boundedness as communities apart from Chicago in general. Again, it was not possible to do this in a strictly comparable way for both the Hispanic and the black communities. Chicago has historically been highly segregated by race, and this pattern continues today (see Massey & Denton, 1993). The black community is, thus, more commonly thought of in terms of large contiguous areas on the West and South Sides of the city—the traditional Black Belts—than in terms of smaller, neighborhood areas. Mexicans and Puerto Ricans are more widely dispersed throughout the residential areas of the city as a whole, although there are a few identifiable areas of higher concentrations.

Chicago is formally divided into seventy-seven community areas. These areas are useful designations because they refer to spaces that typically have real social meanings. These areas are made up of groups of census tracts and have been used by numerous city and social-service agencies as catchment areas. We thus began with census data for these defined areas and narrowed the choices to about twelve communities. We sent field-workers out to walk the main streets of several areas and spoke to service providers and residents of these communities, inquiring about the geographic boundaries as defined by local residents. Our final choices of areas do not correspond exactly to any of the seventy-seven officially designated community areas. Rather, we included most, but not all, tracts in chosen areas as well as a few tracts outside the community boundaries. Pseudonyms were assigned to final choices to reinforce the fact that areas do not strictly conform to official community-area boundaries. The general location of each pseudonymous area—"Shoreland" on the North Side (a neighborhood dominated by affluent young white singles and gay men), "Southtown" on the South Side (an African American neighborhood), "Westside" on the West Side (a predominantly Mexican neighborhood), and "Erlinda" on the Northwest Side (a mixed-Hispanic neighborhood)—will be clear to anyone familiar with Chicago. However, our choices of contiguous tracts distinct from the contiguous tracts that constitute the official community areas make it impossible to know with certainty whether any given census tract falls within one of our sample areas. Maps in later chapters indicate the general areas of the city in which interviews were conducted but do not accurately outline the sample areas.

Fielding the Study

The CHSLS fieldwork was conducted during two separate field periods. Data were collected on the first four samples—the Cook Country cross-sectional sample and the Shoreland, Southtown, and Westside oversamples—in 1995. The initial study was not fully funded, resulting in a delay while additional monies were secured to complete the neighborhood samples. Subsequent cost overruns truncated the first field period early so that resources were also needed to boost initial samples. Funding was obtained to collect data on the Erlinda oversample and to collect additional data on the Cook County cross-sectional and the Shoreland and Westside samples in 1997.

Response rates ranged from 60 to 78 percent across the samples. Readers may wonder whether low response rates were due to survey content. We do not believe this to be the case. The initial, 1995 field period ran into unanticipated, costly household-listing expenditures when the purchased sample proved to be of lower quality than expected. This led to cost overruns and a sudden, premature closing of the first field phase. Specifically, the low response rate in Southtown was due to the abrupt termination of field operations that resulted in many "in-process" interview prospects not being successfully pursued to completed interviews. However, no special difficulties were encountered in the field, and interviewers did not have problems gaining cooperation from Southtown respondents. Similarly, truncated field operations in the other samples were offset by reopening cases in 1997. During this second field period, additional cases were collected in a new sample area—Erlinda—and in three of the four original samples—the Cook County cross section, Shoreland, and Westside—but not in Southtown. This meant that low response rates from the shortened first field period could be remedied somewhat in all but the Southtown samples. Because limited funds precluded collecting more cases in all the neighborhoods and the number of completed African American cases in both the cross section and Southtown was sufficient to sustain independent analysis, the decision was made to spend what limited funding was available on collecting cases from areas where two other minority populations of interest—Hispanics and gay men—were more prevalent.

Shoreland, unlike Southtown, did provide some special field difficulties. These were due to contact rather than cooperation problems, however. Shoreland is characterized by many locked, high-rise buildings and gated developments as well as a population that works long hours and is also highly socially active and, therefore, seldom at home. The high response rate from Erlinda, which area was fielded entirely after the listing and contact problems that plagued the Shoreland collection process were resolved, demonstrates that survey content did not preclude high response rates.

The National Opinion Research Center (NORC) fielded the close-coded CHSLS survey. Interviewers were provided with written "Q-by-Q's"—question-by-question explanations of the purpose of questions, acceptable probes, and definitions. They were also required to attend an intensive all-day training session and a series of mock interview sessions. The all-day training included presentations by the principal investigators about the purpose and importance of the research and "desensitivity" sessions to help interviewers maintain a professional and neutral demeanor when asking respondents about sensitive topics. In addition, trainers gave interviewers practical advice and demonstrations of acceptable and unacceptable ways in which to obtain cooperation, probe for clarity and completeness of responses, and so forth. Mock interviews gave interviewers a chance to practice reading the questionnaire aloud and permitted supervisors to observe interviewing style and to correct any problems. Highly experienced operations staff supervised interviewers, providing them with ongoing oversight and direction throughout the field period.

Interviews in the neighborhood samples were conducted mostly by matched-race interviewers. Most interviewers in Southtown were African American, virtually all those in Westside and Erlinda were Hispanic and bilingual, and most interviewers in Shoreland were white. The countywide sample was more mixed in terms of the match between respondent and interviewers. Most of the interviewing staff were women, but NORC found that men were more successful in Shoreland and, thus, moved as many of the male interviewers as possible to that area. NORC also found that it was best not to send Puerto Rican interviewers to Mexican households but that sending Mexican interviewers to Puerto Rican households was usually not a problem.

People are often suspicious of the responses that interviewees provide to socially sensitive questions, and rightly so. Research into honesty in survey response consistently shows that, the more socially undesirable (desirable) the behavior, the more it will be underreported (overreported) (Sudman & Bradburn, 1982). While such social-desirability effects cannot be overcome entirely, there are techniques that can be employed to reduce the degree to which over- or underreporting occurs, including normalizing sensitive questions and providing greater privacy in which to respond to the most sensitive questions (Tourangeau, Rips, & Rasinski, 2000).

Interviewers for the CHSLS were carefully trained to conduct interviews in a professional manner and in private. In cases where spouses or partners insisted on being present during an interview, interviewers reported creative means of preventing the intruder from overhearing responses to questions most likely to be influenced by the presence of a second party. For example, interviewers reported asking a partner sitting within earshot of the interview for a glass of water at critical junctures and quickly running through the most sensitive questions while the partner was out of the room. Interviewers also reported utilizing seating arrangements that allowed interviewees but not others to view the computer screen and asking for responses by number. The use of hand cards—a technique to reduce social-desirability effects more generally—facilitated such practices. A self-administered portion of the questionnaire allowed interviewees to answer some of the most sensitive questions in a completely private manner.

In short, while it is impossible to know just how honest self-reports about sensitive and private behavior are, this study used available means to reduce this source of bias as much as possible in the context of an interviewer-administered questionnaire. Interviewers were carefully instructed in how to comport themselves so as to reduce bias, questions were worded in ways that normalized behaviors as much as possible, and some sections of the questionnaire were self-administered.

Key-Informant Interviews

To better understand the institutional regulation of sexual expression, the CHSLS conducted 160 interviews with community leaders and service providers in four institutional domains—medical, religious, legal, and social services. These domains were selected to cover the ways in which sex is most often framed in public discourse—as a health issue, as a moral issue, as an issue of social control, and as an issue affecting individual, family, or community well-being. Those interviewed were asked about community norms and behavior with regard to AIDS, other STIs, domestic violence, homosexuality, family formation, and the like. Interviews lasted between thirty minutes and four hours, with most taking about an hour and a half.

To ensure that we captured as much data as possible in these interviews, we adopted a unique, interviewer-assignment process in which two interviewers attended and independently wrote up each interview. In addition to ensuring greater completeness and reliability, this strategy allowed interviewers to specialize either in one community or in one institutional domain, with the result that most interviews were conducted by both a community specialist and an institution specialist. Eight research assistants in total were assigned to the key-informant interviews. The person responsible for a community area attempted to attend all interviews with local service providers, while the person responsible for an institutional domain tried to attend all interviews with relevant organizations. The eight interviewers were four men and four women—two African Americans (one man, one woman), one gay man, one lesbian, and two persons conversant in Spanish (one fluently). At times, we varied from the prescribed community/institution-specialist pairing and matched interviewer and informant characteristics (i.e., race, language, or sexual orientation) in order to gain informants' cooperation or put them at their ease.

The CHSLS Questionnaire

The survey instrument was administered to all respondents in each of the five probability samples—the Cook County sample and the four Chicago community areas. The CHSLS questionnaire was designed specifically to complement the NHSLS questionnaire. In order to facilitate comparison between the samples, basic questions from the NHSLS, such as those about frequency of sex and numbers of sex partners, were worded so as to mirror the NHSLS versions as closely as possible. Because the NHSLS already provided a baseline distribution of sexual behaviors, lifetime numbers of partners, and demographic characteristics for all partners, we deemed it unnecessary to collect this information in detail again. At the same time, we wanted to be able to compare the CHSLS distributions with the NHSLS distributions. Thus, numbers of partners in the last year and important lifetime facts such as age at first sexual experience, number of marriages, number of cohabitations, and total number of partners were collected again. Questions about sex practices, on the other hand, such as frequency of particular sex acts over the life course, were restricted to only the two most recent partnerships. Likewise, detailed social and demographic characteristics of partners were also limited to the two most recent sex partners.

To take maximal advantage of the local character of the CHSLS, we added several new sections and made modifications to the basic NHSLS questionnaire. Whereas the NHSLS was intended to provide representative statistics for a national sample, the strength of the CHSLS was that the survey focused on local sex markets. Persons interviewed could potentially partner with each other. Thus, patterns of partnering and nonpartnering would represent actual network cleavages rather than artifacts of the research design. In a national survey, it is impossible to control for local availability when looking at intergroup partnerships. So, for example, perhaps Hispanics married to non-Hispanics captured by a national study are those persons living in areas with few or no coethnics. In Chicago, we know that Mexicans are within reach of Mexican, Puerto Rican, black, white, and Asian partners so that matched partnering represents a positive choice. We wanted to use this fact to document how sexual partnering is socially structured on the ground.

We therefore decided that the questionnaire should focus on current rather than lifetime partnerships. We wanted to be able to ask about the actual physical locations where the partners met, which would permit us to geocode the neighborhoods where dating partners resided. We also wanted to ask respondents about their social networks more generally and about connections between their social and their sexual relationships. This, too, argued for focusing on current relationships, as it is difficult to report retrospectively about the characteristics of network ties. Thus, we added new questions about social networks, neighborhood characteristics, and the geographic parameters of respondents' sexual, social, and work lives.

Another modification—the heart of the CHSLS questionnaire—was a section designed to collect detailed information about the characteristics of respondents' two most recent sex partners. The NHSLS data documented the fact that most people's sex lives follow the pattern of periods of stable, long-term relationships punctuated by periods of numerous short-term relationships. That is, at any given point in time, most people are in the midst of long-term relationships; thus, cross-sectional data underrepresent more transient partnerships. By collecting information on the two most recent partners, we were typically able to capture the current primary partnerships of our respondents as well as a past short-term relationship. Details about relationship characteristics, as well as network and neighborhood information, allow us to specify the social contexts surrounding sexual partnerships.

The questionnaire was pretested in a paper-and-pencil format. This strategy was adopted because of the high costs associated with reprogramming laptops as the questionnaire was continually revised. That is, whereas the paper-and-pencil format was easily revised (by altering a computer file stored on a word processor and printing the results), the process by which the computerized format was revised was much more complicated. Since the software employed to program the laptops utilized linked screens, each change to the questionnaire required considerable work to ensure that skip patterns remained logical. Thus, it made sense to accumulate revisions using the paper-and-pencil format and to reprogram the CAPI instrument (the computerized format) only after accumulating a large number of revisions.

Using flyers, we recruited pretest respondents from a variety of demographic categories—men and women, young and old, gay and straight, minority and white. We paid pretest respondents for their participation and informed them that their interviews were pretests and were being observed. The questionnaire was administered to pretest respondents in a room with a two-way mirror, allowing members of the questionnaire-design team to observe. The questionnaire was revised when questions did not function as envisioned or when wording proved inappropriate for some respondents.

The programmed version of the questionnaire was tested by project staff rather than by actual respondents. Staff acted as respondents to test the limits of skip patterns when there were no partners, numerous partners, or unusual situations. Further, testers refused to answer critical screening questions or answered inconsistently to make sure that built-in cross-checks worked properly.

The final questionnaire was long, taking an average of ninety minutes to administer, and included complex skip patterns. However, CAPI automatically managed skips and text substitutions for gender and tense. A relatively lengthy self-administered section was included in the questionnaire to improve response rates on especially sensitive topics, such as sexual orientation, same-sex sexual experiences, domestic violence, and drug use. Because two of our oversample neighborhoods included large numbers of Spanish-speaking residents, the entire questionnaire was translated into Spanish (the Spanish-language version was available only in a paper-and-pencil format), and bilingual interviewers were used in these neighborhoods.

Describing the Samples and the Communities


Shoreland is an affluent, mostly white area with trendy retail stores, restaurants, bars, and clubs drawing daytime shopping and nighttime leisure traffic from around the city. High-rise apartment buildings, town homes, and condominiums predominate. Streets are lined with late-model cars; parking is difficult. Many residents are young adults (i.e., between eighteen and twenty-nine): 43 percent, compared to only one-quarter for the city as a whole. Fully three-quarters of the locals live in nonfamily households, an astounding 62 percent living alone. There are few children, fewer teenagers. Two-thirds of the population lived in a different residence five years earlier. These percentages correspond to key-informant descriptions of Shoreland as a transitional neighborhood: singles meet there, marry, and then leave, particularly once a child is born or reaches school age.

In terms of status characteristics, Shoreland residents are upwardly mobile. Income and education-attainment levels are high. Over half the residents have a bachelor's degree, a fifth an advanced degree. Although the poverty rate in Shoreland is higher than it is in suburban areas, it is half that of the city as a whole. The residents—most with high-paying jobs, and few with children—have considerable disposable income, a fact reflected by the luxury goods and services offered by local businesses.

Shoreland is also a highly secular community, with roughly 25 percent of the population professing no religious affiliation. About 30 percent are Catholic and 30 percent Protestant, with a slight majority of the Protestants attending liberal, mainline churches. The last 14 percent are split between Jewish and other religious affiliations. Attendance at weekly services is, however, low.

A distinctive feature of Shoreland is its visible concentration of businesses, from bars to bookstores, geared toward gay men and lesbians. The preeminent social-service agency for gays and lesbians in Chicago is located in this area, and the annual gay and lesbian pride parade is routed through the neighborhood. It is not unusual to see same-sex couples walking down the street holding hands, and such behavior does not draw public attention.


Westside is a port of entry for Mexican immigrants, both legal and illegal. Roughly half the population is foreign born, a fifth of these individuals having immigrated to the United States in the last five years. Westside was formerly an Eastern European enclave. One-tenth of the local population is non-Hispanic white, mostly elderly homeowners whose grown children have left the neighborhood. The local culture is firmly, palpably Mexican. Signs in stores are in Spanish. There are Aztec-themed murals on the walls. Radios blare ranchero music. Restaurants offer Mexican cuisine. A lively street trade in fresh fruit and vegetables is geared toward Mexican cooking. Shop windows display paraphernalia associated with the many church- and family-centered ritual occasions that organize community life—baptisms, first communions, confirmations, quinceañeras, graduations, and weddings.

Westside is poor, a working-class neighborhood. Although the business strip is thriving, barred windows are common. Most Westside immigrants hail from rural areas of Mexico; many lack even a secondary education. Men are employable but earn little for their labor. Indeed, household incomes are clustered around the poverty line, although welfare dependence is quite low, as would be expected in an immigrant neighborhood. Reflecting the traditional division of labor preferred by most Westside families, women work at rates below those of women in Chicago generally. In keeping with their Mexican origins, most Westsiders are Catholic—over 80 percent. The remainder are mostly Protestant. Westside has an unusually high proportion of Catholics who claim to have been born again. This pattern likely reflects the expressive character of Mexican Catholicism in contrast to that of the Catholicism of European immigrant groups. Westsiders also attend church more frequently than do residents of Shoreland, Southtown, or Erlinda or residents of Cook County generally.

Westside houses are mostly two-flats or three-flats, with some three- and four-story walk-up apartment buildings. Houses are often built one in front of the other on narrow lots, and residences are among the most crowded in the city. The overall age profile is very young, with a quarter of the residents under age twelve. Nuclear families predominate, with almost two-fifths of households composed of a married couple and their children. Owing to the presence of unattached immigrant men, nonfamily households are more common here than they are in Erlinda or Southtown. Gangs are also prevalent, and their graffiti is liberally applied to garage doors and park benches.


Erlinda, another community area in which the population is predominantly Hispanic, mirrors Westside in many respects. However, Erlinda is more racially and ethnically diverse than is Westside and houses fewer immigrants and more later-generation Hispanics. The area is commonly thought of as a Puerto Rican neighborhood—and it is, in fact, 40 percent Puerto Rican—but Mexicans and other Central and South Americans make up a third of the population, one-fifth of the population is non-Hispanic white, and 5 percent is African American.

The Erlinda area is a somewhat more desirable section of the larger Puerto Rican/Hispanic Northwest Side, a place where relatively better off residents have moved as gentrification has squeezed them westward and as upward mobility has allowed them to purchase modest homes. Houses are small but well kept. Despite the fact that the household incomes of a quarter of the population are below the poverty line, there is a local sense of progress, of having "moved up" from the more dangerous and poorer parts of the neighborhood.

Like both Westside and Southtown, Erlinda is dominated by families. Fully 80 percent of the residents live in family households. But only one-third of all households in Erlinda are nuclear-family households, 5 percentage points fewer than in Westside, and 17 percent are female-headed households, contrasting with Westside's 10 percent, and identical to Southtown's 17 percent. As is also true in Westside, one-third of Erlinda's population is under the age of eighteen.

Education-attainment levels in Erlinda are very low, although more Erlinda residents than Westside residents have completed high school. Employment is relatively high given poverty rates, indicating that many of the poor are working rather than unemployed. Like Westside, Erlinda is heavily Catholic, and rates of church attendance are high. In contrast to Westside, however, Erlinda has a larger group claiming no religious affiliation and a larger group of Type II Protestants—Baptists, conservative fundamentalists, and members of Pentecostal sects. This is consistent with trends in Puerto Rico, where many have become disaffected with the Catholic Church and proselytizing Protestants have won many converts.


Southtown is an African American neighborhood that has seen better days. The large industries in the surrounding neighborhoods that once provided Southtowners with good, blue-collar jobs have downsized or shut down entirely. Now, only half the men and women are employed. Once home to a shopping district with national chain stores serving the entire South Side of the city, Southtown is left mostly with liquor stores and miscellaneous dry-goods stores (advertising wigs, jewelry, and gym shoes)—all heavily barred. Many shop owners are Korean or Arab and live outside the area. There is also a red-light district located on the main street. Drug activity and gang activity are both concerns along the main street.

At the height of factory closings in the 1970s, over nine hundred mortgages were foreclosed in the surrounding area, and vacant homes and commercial buildings remain a concern in some parts of the neighborhood. The housing stock is mostly single-family homes, well-constructed one- and two-story brick bungalows on lots with yards. Many of these homes are occupied by older couples with grown children and by white-collar workers who have retained their jobs. However, a growing proportion of the homes are shared by multigenerational, female-headed families (e.g., mother, daughter, and daughter's children) or by single mothers doubling up with one another. Seventeen percent of household heads are women with minor children. The poverty rate for Southtown is right at the city average—a fifth of the households are under the poverty line—but a third of the households earn upward of $40,000. Persons interviewed described the area as mixed income, crediting the single-family homes with retaining the better-educated, employed segment of the population.

Southtown is also known as a community of churches. The black church is the social center of the black community. There are at least two or three churches to be found on every block of the main street. There are large churches with high-profile ministers that pull congregants from all over the South Side as well as many small churches, including numerous storefront churches. There is, however, little interaction between the two.

There is a great deal of activity on the main street during the day. Many people hang out there. Most are between sixteen and thirty, and most are men, but some are young women, often teenagers with infants and/or toddlers. Residents of this neighborhood have attained higher levels of education than have Hispanics in Westside or Erlinda; still, 18 percent of the residents never graduated from high school, and only 11 percent have a college degree. The majority of Southtown residents are single (72 percent), but the average age of singles is Southtown is somewhat older than that of singles in other neighborhoods. And Southtown is highly segregated; 98 percent of Southtown residents are African American.

Sexual Expression in the CHSLS Communities

As mentioned earlier, a central objective of the CHSLS was to capture the diversity of sexual expression in urban life.

Sexual Attitudes and Sexual Identity

There is substantial variation across the samples and between men and women with respect to sexual attitudes. In general, many of the respondents in the CHSLS exhibit nonpermissive attitudes toward extramarital sex, homosexuality, pornography, and abortion. Overwhelming majorities of the respondents in Westside, for example, believe that extramarital sex is always wrong (74 percent of men, 81 percent of women), that sex with a same-sex partner is always wrong (76 percent of men, 72 percent of women), that the sale of pornography should be illegal (57 percent of men, 82 percent of women), and that abortion for any reason should be illegal (70 percent of men, 76 percent of women). Respondents in Shoreland, by contrast, exhibit much more permissive attitudes. There are interesting gender differences as well. Men in all the samples tend to have more permissive attitudes toward pornography and extramarital sex, which is interesting considering the fact that they are more likely than women to cheat on their spouses and to purchase pornography (Laumann, Gagnon, Michael, & Michaels, 1994). Women, in contrast, are more tolerant of homosexuality and are more likely to hold proabortion views than are men.

One implication of these differences for sex markets is the importance of local context. For example, individuals with an interest in forms of sexuality considered taboo in their neighborhoods will be forced to pursue those interests in other parts of the city. Social disapproval can also result in the strict segregation of certain sex markets, even forcing some underground. The subject of the importance for sex markets of local context generally is taken up by Mahay and Laumann (chapter 3 in this volume). An extended discussion of the construction of same-sex sex markets specifically can be found in Ellingson and Schroeder (chapter 4 in this volume).

Another implication of these differences is the resulting differential management of or response to sexual issues as public-health issues. Ellingson (chapter 10 in this volume) discusses institutional approaches to sexual problems generally. Ellingson, Van Haitsma, Laumann, and Tebbe (chapter 11 in this volume) discuss the religious response specifically.

Sexual Behavior, Search, and Commercial Sex

Closely connected to sexual attitudes, sexual behavior exhibits similar patterns. Respondents in Southtown and Shoreland are more oriented toward multiple partnering strategies. About half the men in each sample reported having had at least two sex partners during the twelve months prior to being interviewed; 61 percent of the Southtown men and 78 percent of the Shoreland men have had more than two one-night stands during their lifetime. In Westside and Erlinda, vast majorities of the men and women are monogamous, and few have ever engaged in a one-night stand. Overall, men tend to have more partners than do women.

Shoreland respondents are distinctive in that they are the most likely to engage in active searches for potential sex partners. For example, during the year prior to being interviewed, 20 and 21 percent of unmarried women and men, respectively, went to the gym partly with the hope of meeting someone. The comparable figure for Cook County is 4 percent of women and 10 percent of men. Indeed, Shoreland respondents are the most likely of all our respondents to engage in search behavior, from joining sports teams to taking classes, from going to bars to being set up with dates.

Overall, the majority of CHSLS respondents do not engage in public commercial sexual activities, such as going to a "gentlemen's" club, watching pornographic movies in public theaters, or hiring a prostitute. That is not to say that no one in Chicago engages in such behavior. Indeed, substantial numbers of Chicagoans, particularly among the men, appear to participate in the commercial sex market. The most popular form of commercial sex is watching a pornographic video at a private residence. Forty percent of the men in Cook County did so during the year prior to being interviewed; the comparable figure in Shoreland is even higher (65 percent). In contrast, prostitution is quite rare. Only 2-4 percent of men, and essentially no women, hired a prostitute in the previous twelve months.

These patterns raise certain issues. One is the question of what leads men and women to engage in search behavior, a topic taken up by Mahay and Laumann (chapter 5 in this volume). Another is the fact that the prevalence of multiple partnering strategies in Shoreland and Southtown suggests that sexual partnering is organized differently among different racial/ethnic groups. Aspects of this issue are discussed by Youm and Paik (chapter 6 in this volume) and by Paik, Laumann, and Van Haitsma (chapter 7 in this volume).

Health and Well-Being

Overall, exchanging sex for drugs is relatively rare, although we note that this practice is significantly more common among respondents in Southtown than among those in the other samples. On the other hand, drug use generally is quite common, especially in Shoreland, where one-third of the men and one-quarter of the women have used either crack or cocaine in the past.

One striking feature is the prevalence of forced sex, particularly among women. Twenty percent of Cook County women and nearly a third of the Shoreland women report ever having been forced to have sex. These statistics are significant for our argument since an economic view of sex markets would imply negotiated transactions without force or fraud. Clearly, the use of force, primarily by men against women, is an important feature of sex markets. This topic is taken up by Buntin, Lechtman, and Laumann (chapter 8 in this volume).

Reported experience with STIs varies across the samples, although men and women appear to be equally at risk. This topic is discussed by Youm and Laumann (chapter 9 in this volume).

We stress, however, that sexual expression involves more than negative outcomes. Indeed, most men and women in the CHSLS samples find their sexual relationships to be very satisfying, both physically and emotionally. Thus, while sexual activity involves certain risks, it certainly has its rewards as well. The following chapters touch on these issues as well as others.

The Sexual Organization of the City cuts across several sexual scenes, each distinctly marked by the types of people and the sexual attitudes and behaviors found there. It is able to do so because it is based on a unique data set, one that allows us to tell a story about sexual expression in Chicago that is representative and general as well as site specific and contextual. Thus, we capture both the diversity and the broad patterns of sexual partnering in Chicago.

Copyright notice: Excerpted from pages 1-xxxx of The Sexual Organization of the City edited by Edward O. Laumann, Stephen Ellingson, Jenna Mahay, Anthony Paik, and Yoosik Youm, published by the University of Chicago Press. ©2004 by The University of Chicago. All rights reserved. This text may be used and shared in accordance with the fair-use provisions of U.S. copyright law, and it may be archived and redistributed in electronic form, provided that this entire notice, including copyright information, is carried and provided that the University of Chicago Press is notified and no fee is charged for access. Archiving, redistribution, or republication of this text on other terms, in any medium, requires the consent of the University of Chicago Press and of the author.

Edward O. Laumann, Stephen Ellingson, Jenna Mahay, Anthony Paik, and Yoosik Youm, editors
The Sexual Organization of the City
©2004, 424 pages, 1 map, 21 figures, 39 tables
Cloth $35.00 ISBN: 0-226-47031-8
Paper $23.00 ISBN: 0-226-46897-6

For information on purchasing the book—from bookstores or here online—please go to the webpage for The Sexual Organization of the City.

See also: