Black teenagers are more likely than whites to have intercourse, and a study using data from the National Longitudinal Survey of Youth (NLSY) suggests a link between residential racial segregation and that disparity.[1] Black participants who had lived in metropolitan areas with high levels of segregation, as measured across five dimensions, were more likely than whites from such areas to have initiated intercourse before age 20; racial disparities were generally not evident, however, among those from areas with moderate or low levels of segregation. Similarly, the odds of adolescent sexual activity were elevated for blacks who had lived in areas that ranked high on four or five of the dimensions of segregation, but not for those from more mixed areas.

The NLSY began in 1997–1998 with a cohort of 12–16-year-olds, who are followed up annually. To study the associations between racial segregation and racial differences in age at first sex, analysts examined data collected through 2005 from 4,311 participants (1,687 blacks and 2,624 whites) who lived in metropolitan areas with total populations of at least 100,000 and black populations of at least 5,000, as identified through census data. They conducted discrete time-to-event analyses to determine whether racial segregation is related to the odds of teenage sexual activity after a range of characteristics of respondents and features of metropolitan areas are taken into account. Using existing indices, they examined five dimensions of segregation: the black population's degree of isolation from other racial groups, concentration (or density), centralization (or proximity to the center of the metropolitan area), clustering (or living in contiguous neighborhoods) and unevenness (disproportionate representation in certain neighborhoods). Scores were categorized as very low, low, moderate or high.

Forty-seven percent of blacks and 51% of whites in the sample were male. At baseline, the mean age in both groups had been about 14 years; black respondents had been more likely than whites to live in a single-parent household, to live in a household including more than two children and to have parents who were high school graduates. Sixty percent of blacks and 21% of whites had lived in a central city at baseline.

Of the 105 metropolitan areas included in the analyses, 24% were classified as highly isolated, 61% as highly concentrated, 72% as highly centralized, 5% as highly clustered and 44% as highly uneven. Areas that were rated high on four or five dimensions were considered hypersegregated; 16% fell into this category.

Some 92% of black participants and 86% of whites had first had intercourse during adolescence; the average age at first sex was 15.3 years among the former and 16.2 among the latter. Blacks’ elevated likelihood of having initiated sexual activity before age 20 was borne out in an analysis controlling for all covariates except dimensions of segregation (odds ratio, 1.2), and results were essentially the same in models that included one segregation measure apiece. Of the segregation measures, only concentration was associated with participants’ overall odds of having had sex as teenagers: Youth in areas where the black population was moderately or highly centralized had been more likely than those in areas with very low centralization of the black population to engage in intercourse (1.5 and 1.4, respectively).

In a final set of analyses, the researchers compared the likelihood of adolescent sex by race at each level of each dimension of segregation. These models revealed that in areas that had a high level of segregation, by any measure, or were hypersegregated, black participants had been more likely than whites to have intercourse as teenagers (odds ratios, 1.2–1.4). The only other association in these analyses was an elevated risk among blacks from areas with very low levels of isolation (1.6).

The analysts point to limitations of their data that call for caution in interpreting the results. For example, the NLSY asked only about partners of the opposite sex and about where participants lived at baseline (which may not have been where they lived when they first had intercourse), some measures had substantial amounts of missing data, and the sample of metropolitan areas may not have been nationally representative. However, the researchers also note that little previous work has examined associations between segregation and sexual risk throughout adolescence. Emphasizing the complexity of the associations suggested by their analyses, they stress the need for further work to identify the causes and moderators of these relationships.—D. Hollander

Reference

1. Biello KB et al., Racial differences in age at first sexual intercourse: residential racial segregation and the black-white disparity among U.S. adolescents, Public Health Reports, 2013, 128(Suppl. 1):23–32.