Losing Generations: Adolescents in High-Risk Settings (1993)

Most of social interactions of families and adolescents are embedded within neighborhood settings. A ''neighborhood" can be defined both spatially, as a geographic area, and functionally, as a set of social networks. Neighborhoods are spatial units in which face-to-face social interactions and processes take place in relatively intimate, personal settings and situations. They are a place for social interaction, a place for education and human service—both formal and informal—and a place for preparing for and engaging in employment. A "community" may differ from a neighborhood in size or in residents' or outsiders' characterization (for example, a housing complex or a large area defined by accepted landmarks). Moreover, a "community" may have a political orientation. But, generally speaking, it is difficult for residents and social scientists to precisely distinguish the difference between social and political interactions and between neighborhood and community. As a result, the terms are often used interchangeably, as we do in this report.

However defined, a neighborhood is a key setting for adolescent development. This chapter summarizes research about neighborhoods, with a focus on poverty concentration and racial and ethnic stratification, and then examines the effects of these neighborhoods on adolescent development. We note, however, that there are both methodological and theoretical limitations to the research on neighborhoods.

Methodologically, it is difficult to identify causal relationships

between complex social settings and individual behavioral outcomes, and most results are open to more than one explanation (see Jencks and Mayer 1990a; Reiss and Roth, 1993). Also problematic is the definition of the areas under study: regions, standard metropolitan statistical areas (SMSAs), and cities are large, heterogeneous units with politically defined boundaries which are not coterminous with neighborhoods and may have no theoretical relevance for many research questions on adolescent outcomes; thus most researchers use intraurban units of analysis, usually census tracts, as proxies for "neighborhood," although they too do not necessarily correspond to neighborhoods. In addition, much research is based on decennial census data, but there is a significant time lag in the availability of those data for research. Thus, conclusions may be drawn for the period between 1970 and 1980, but one can only speculate from incomplete data regarding subsequent trends.

Almost all research examines metropolitan (i.e., urban and suburban) neighborhoods, with little attention to the unique characteristics and problems of rural areas. Moreover, the available research focuses almost exclusively on poor neighborhoods, and so there are few comparative data from affluent areas. Another limitation is that most research has focused on the structural features of neighborhoods, not on community-level interactions or processes. A theoretical bias toward consensual theories of community, rather than theories of a community-as-a-resource, has led researchers to look for internal sources of disorganization rather than external sources. As such, relatively few community studies attempt to analyze how private-sector actions, government policy, and regulatory law influence neighborhoods. There is also relatively little attention to a significant body of ethnographic research (see Anderson, 1991b; Skogan, 1990). In spite of the limitations, however, the available research on neighborhoods does provide significant insights into this setting for adolescents.

Household poverty and segregation by class and race are fundamental elements of metropolitan areas, and there is a clustering of poor people in many of these areas. By 1980, the 10 largest metropolitan areas accounted for almost half of all poor persons

who live in the poorest neighborhoods—those census tracts in which at least 40 percent of all residents have poverty-level incomes. The numbers of such residents are increasing. The 1970s saw a 29.5 percent increase in the number of poor people living in these neighborhoods, to an estimated 2.45 million. This growth was not evenly distributed. A little over half of all SMSAs showed no change or a decrease in concentrated poverty; the others, especially large cities in the Northeast and Midwest, showed increases in this concentrated poverty (Jargowsky and Bane, 1990; Farley, 1990). An analysis of the neighborhood changes from the 1990 census is not yet available, but evidence suggests that trends seen in the 1970s have continued through the 1980s.

Poor neighborhoods are also racially and ethnically stratified despite declines in stratification during the 1970s (Farley, 1990; Wilson, 1987; Massey and Denton, 1987). The average black family, for example, lives in a census tract in which 30 percent of families are poor; for the average Hispanic family, the figure is 23 percent; the average Asian family, 20 percent; and the average non-Hispanic white family, 14 percent (Massey and Eggers, 1990). However, there are regional and metropolitan characteristics that affect these differences. For example, trends during the 1970s toward racial integration were strongest in the smaller metropolitan areas and weakest in SMSAs with the largest black populations. For Hispanics, stratification was most pronounced in large metropolitan areas with low rates of population growth, in areas with large Hispanic populations, and in areas with high rates of Hispanic immigration (Massey and Denton, 1989). Nonetheless, the poorest neighborhoods remain the most highly stratified: an estimated 87 percent of all residents in such neighborhoods are members of racial or ethnic minority groups (Jargowsky and Bane, 1990).

Racial stratification is not limited to the core cities within SMSAs. Although suburban stratification is about 12 percent lower than in central cities, it is still high, especially for blacks in the Northeast and Midwest (Massey and Denton, 1988). In 1980, 86 percent of suburban whites still lived in census tracts with less than 1 percent black residents. In other words, black suburbanization seems to represent not integration, but an extension of racially segregated living patterns into neighborhoods adjacent to or near central cities (Judd, 1992; Logan and Schneider, 1984; Massey and Denton, 1988).

Adults in poor neighborhoods differ in important ways from those in more affluent areas. For example, high poverty neighborhoods have much higher proportions of unmarried mothers, single-parent families, and unemployed young men (Jargowsky and Bane, 1990).

Ricketts and Mincy (1990) argue that the social composition of an increasing number of neighborhoods has deteriorated during the 1970s, resulting in an increase in "underclass neighborhoods." They define "underclass" neighborhoods as census tracts with high values (at least one standard deviation above the national mean) for each of four indicators: working-age males not attached to the labor force, households headed by a woman with children, households receiving welfare, and dropouts among the school-age population. They use the term "concentrated poverty neighborhoods" to define those with 40 percent or more residents having poverty-level incomes. Although the poverty rate remained fairly constant from 1970 to 1980, there was a 75 percent increase in the number of census tracts with concentrated poverty, and, significantly, a 331 percent increase in the number of underclass neighborhoods (see Table 4-1). That is, over the course of the decade, the number of Americans living in neighborhoods with high concentrations of poverty had more than tripled.

Concentrated poverty goes hand-in-hand with many social problems, and the increases in both concentrated poverty and underclass

census tracts far surpass changes in other neighborhoods. For example, the underclass neighborhoods had a 355 percent increase in the measure of "idleness" (i.e., people with no regular attachment to the labor force or education), 5 times the increase in poor neighborhoods, and 20 times the increase for the overall U.S. population (see Table 4-2). By 1980, more than half of all poor neighborhoods had become underclass neighborhoods.

The growth in crack and cocaine markets since the early 1980s has placed additional stress on poor neighborhoods. The highly visible, lucrative, and violent drug markets have simultaneously accelerated the exodus of stable families and undermined the authority

of long-term community leaders. The operation of drug markets and the violence associated with them has weakened inhibitions against violence in all neighborhood contexts. The large amounts of money that can be made in the drug trade act as a magnet to draw children and adolescents into criminal activity (Reiss and Roth, 1993). Adolescents who are not involved as participants in drug markets are still influenced by their presence; some are victims of drug-related violence, while many more are unable to engage in normal neighborhood activities because of the dangers associated with drug markets.

Because the vast majority of students attend neighborhood schools and because school funding is related directly to family and neighborhood wealth, public schools tend to be stratified by class, race, and ethnicity. About 90 percent of all students attend public schools, and less than 5 percent of those students attend schools with substantial bussing programs (Mayer, 1991). Therefore, student bodies reflect the composition of the neighborhood.

As a result, poor children are concentrated in some schools and are largely absent from others. Almost half of all black children and almost 40 percent of Hispanic children live in poverty. About half of children in public schools in central cities are black or Hispanic, and most are poor. In 1989, 51.8 percent of public school students in central cities were black or Hispanic, a percentage that has increased steadily since the 1970s (Alsalem et al., 1992). Concentrations within individual schools are even more pronounced. According to analyses by Orfield and colleagues (1983, 1989, 1992:v), "school segregation of Hispanics has increased dramatically," while segregation has "remained relatively stable for Blacks" over the past 20 years. Consequently, many students in the public education system continue to attend segregated schools. As seen in Table 4-3, for example, 80 percent of all Hispanic students in the South attend schools with at least 50 percent minority population. Thirty-eight percent of these Hispanic students are in "intensely segregated" schools with 90 to 100 percent minority populations. Comparable concentrations exist for black students. In the Northeast, for example, over three-quarters of all black students attend "predominately minority" schools, with almost half attending intensely segregated schools.

Minority students are also most likely to attend the nation's poorest schools. Analyses of national data indicate that about 75

percent of black students and 46 percent of Hispanic students attend schools ranked in the bottom 20 percent by socioeconomic status (SES). By contrast, less than 7 percent of black and Hispanic students attend high-SES schools (Mayer, 1991).

Metropolitan areas in the Northeast and Midwest were hit especially hard by the economic downturn during the 1970s. The greatest deterioration in metropolitan neighborhoods, as measured by increased concentration of poverty, was also in the Northeast, followed by the Midwest. In addition, the number of underclass neighborhoods grew fastest in the Northeast (Jargowsky and Bane, 1990:Ch. 2; Ricketts and Mincy, 1990). On a regional level, therefore, it would be expected that the greatest increase (or least decrease) in social problems would also be in the Northeast and perhaps the Midwest.

Analyses of regional trends by Mayer (1991) from 1970 to 1980 support these expectations. As seen in Table 4-4, the Northeast had the smallest decrease in public assistance recipients and teenage births, the smallest increase in labor force participation rates, and the biggest increases in single-parent families (overwhelmingly female) and poverty rates. The Northeast also had the second-highest

increase in the murder rate. The South, in contrast, had the greatest decreases in public assistance recipients and poverty rates, the smallest increase in female-headed families, the greatest increase in high school graduates, and the second-smallest increase in murder rates. Although analyses of 1980-1990 data have not yet been done, there is no general evidence to suggest that these regional differences would have changed.

There are multiple factors that contribute to, and maintain, poverty concentration and racial stratification. Public policy in the last two decades has not strongly aided urban areas, but there

are other factors and difficult complexities in drawing conclusions about deteriorating neighborhoods. As highlighted by Massey and Eggers (1990:1186):

Once a neighborhood begins to substantially lose its economic base, however, other factors come into play. Adults and young people become socially isolated, losing the kind of networks and self-or group identifications that support customary behavior and prevent deviant behavior (Wilson, 1987; Fernandez and Harris, 1990; Harrell and Peterson, 1992; Crane, 1991b). Social institutions—schools, the social welfare system, and the criminal justice system—tend to anticipate and facilitate failure, shame, and hopelessness (Williams and Kornblum, 1985; Chapter 8). Parents also lose some degree of hope and have fears for the well-being of themselves and their children (National Commission on Children, 1991). Employers also respond: people living in underclass neighborhoods are less likely to be hired for a job than those in better-off neighborhoods, even with the same background and skills; and one study of Chicago-area employers found screening of prospective employees by address to be a surrogate for racial and socioeconomic status of neighborhoods (Kirschenman and Neckerman, 1991).

As their economic and social systems break down, the poorest of neighborhoods seem increasingly unable to restrain criminal or deviant behaviors (Wilson, 1987, Anderson, 1991a; Reiss and Roth, 1993). These disorganized areas are vulnerable to the processes that "hollow out" urban neighborhoods: fires in abandoned buildings, housing deterioration, fires in occupied buildings, housing abandonment, reduction of fire services, and accelerating outmigration by stable members of the community, leaving behind an isolated subgroup of residents who are unable or unwilling to move into accepted adult roles (Wilson, 1987; Rainwater, 1970, 1987). Not surprisingly, the result is that economically poor neighborhoods differ from more affluent neighborhoods in terms of diminished private economic activity, the types of public and social services available, limited recreational and youth development programs, and higher levels of crime (Littell and Wynn, 1989; Schneider and

Logan, 1982; Reiss and Roth, 1993). Adolescents themselves lose hope and have a diminished ability to use the few opportunities that are available (Sullivan, 1990; Eisenhower Foundation, 1990).

There is little reason to expect that the conditions summarized above are favorable for adolescent development. In recent years, researchers have examined potential influences from two perspectives. Ethnographic research has examined community-level processes and interactions, with an emphasis on identifying the ways in which adolescents respond to deteriorating neighborhood conditions. Quantitative research has examined the structural features of neighborhoods, such as concentrated poverty and the characteristics of adult residents, and their estimated effects on adolescents.

Neighborhood effects appear to work through the social milieu or settings in which adolescents live and act and can be considered as responses to a lack of legitimate economic opportunity. Taylor (1991), for example, reports that black males in poor neighborhoods often relinquish their belief in the possibility of conventional achievement and mobility in mainstream society. And, as documented by Ogbu (1984) and Farrel (1990), this perceived lack of opportunity, and the weight of discrimination, leads to diminished academic performance.

Studies on the structural influences of neighborhoods provide some support for this view. There is strong evidence, for example, that growing up in a high-SES neighborhood raises a teenager's educational attainment (i.e., promotion and graduation), independent of family characteristics (Jencks and Mayer, 1990a). Controlling for family education, income, and other variables, one 10-year study of urban adolescent males found that, overall, the higher the mean family income in a neighborhood (defined by ZIP codes), the greater the number of years of school completed. Moreover, the data suggest that black males stay in school longer if they have white neighbors, but drop out sooner if they are racially segregated (Datcher, 1982). A later study, using a different methodology, replicated these findings for young women (Corcoran et al., 1987).

Two recent studies, building on these investigations, have found

that adolescents living in low-SES neighborhoods are more likely to drop out of high school than adolescents with similar family backgrounds who live in advantaged neighborhoods. One study found that dropout rates were lowest in census tracts with a high percentage of adult workers in professional and managerial jobs and highest in neighborhoods in which fewer than 10 percent of workers were in ''high-status" occupations, especially among black adolescents. The other study, controlling for family background (e.g., mother's education, family structure, income-to-needs ratio) estimated that on average, a teenage girl's chance of dropping out of high school increases from 10.8 percent to 14.9 percent when the proportion of families in her ZIP code with incomes over $30,000 decreases by one standard deviation. An increase in the proportion of families that are headed by female single parents also increased the dropout rate among both blacks and other minority ethnic groups (Brooks-Gunn et al., 1993).

There is strong evidence that poor neighborhoods increase the likelihood of early pregnancy and childbearing. For example, researchers have found that many young women in poverty recognize the risks of early parenthood, but opt for this choice because it provides proof that they are attractive and successful. It also provides adult status more immediately than hoping for a job that may not come. Other teenage mothers lack the strong and sustained relationships with adults that are necessary to resolve basic identity issues. In this context, babies are a real and meaningful source of love in a setting with insufficient positive sanctions and caring attachments (Anderson, 1990; Musick, 1991).

Holding race and family background constant, births to unmarried adolescents are more likely in poor neighborhoods than in more affluent ones (Jencks and Mayer, 1990b). However, this effect is apparently modified by both race and family background. For example, higher numbers of high-status workers in a census tract reduce the chance that black, but not white, adolescents will have a baby. The number of black teenage girls having a baby increased from 7.4 percent to 19.8 percent as the proportion of high-status workers decreased from 31.2 percent to 3.5 percent (Crane, 1991b). On average, the chance of having a baby decreases from 10.6 percent to 6.9 percent as the average proportion of households with incomes over $30,000 increases by one standard deviation; among black girls, the decline is from 32.4 percent to 26.7 percent.

A higher proportion of families headed by females increases the likelihood of becoming pregnant for all teenage girls (Brooks-Gunn et al., 1993).

It is likely that poor neighborhoods also affect adolescent employment prospects. Not only are employers less likely to hire young people from such neighborhoods, but adolescents often believe it more profitable to enter underground or illegitimate employment markets, such as drug dealing or gambling, than formal labor markets (Williams, 1989; Reiss and Roth, 1993). And since many adolescents are unemployed, peer influences, especially those other youths of the same sex, offer significant barriers to gainful employment. Several studies show that growing up in an urban neighborhood that either is predominantly black or has a high rate of welfare dependency reduces young men's chances of finding well-paid jobs in adulthood. However, the median income of a neighborhood does not appear to affect young people's economic prospects independent of its racial mix or welfare recipient rate (Jencks and Mayer, 1990b). More recent analysis shows that, in the 50 largest cities in the United States, the higher the poverty rate of a census track, the more likely teenage males are to be idle (i.e., not in school, employed, or in the military). For example, a 16- to 19-year-old male is eight times more likely to be idle if he lives in a census tract with a poverty rate of at least 50 percent than if he lives in a tract with a poverty rate of 10 to 19 percent (Massey et al., 1991). Other studies have confirmed that the most highly distressed neighborhoods appear to have a stronger negative effect on employment opportunities than neighborhoods in a "middle range" (Crane, 1991a; Brooks-Gunn et al., 1993).

There is a relatively large body of research on neighborhoods, crime, and victimization (see Chapter 8), but few studies specifically examine adolescents or young adults (but see Sampson and Lauritsen, 1991; Reiss and Roth, 1993). However, age is one of the strongest individual-level correlates of offending: arrests for violent crimes peak around age 18 and decline gradually thereafter (Visher and Roth, 1986). We thus assume that most of the following discussion of research findings is generally applicable to young persons.

Urban areas have the highest rates of crime, and within urban areas the rates for both offending and victimization are highest in neighborhoods with concentrated poverty. Violent crime rates are also higher in neighborhoods with high percentages of people in the 12- to 20-year-old age group and large concentrations of single-parent households. And although concentrations of blacks, poor families, and single-parent households are also associated with homicide rates, the level of income disparity within neighborhoods is the strongest predictor (Reiss and Roth, 1993; Block, 1979; Sampson, 1985; Smith and Jarjoura, 1988).

Population density is often found to correlate with higher rates of violent crime, independent of neighborhood composition (Smith and Jarjoura, 1988; Beasely and Antunes, 1974; Sampson and Lauritsen, 1991). However, it is unclear whether it is neighborhood density per se or the density of residential structures that has the strongest influences. One study, for example, found that rates of crime victimization were 2 to 3 times higher in high-density neighborhoods, regardless of compositional factors such as age, race, and gender, but the analysis also revealed a strong positive association between rates of robbery and assault victimization and the percentage of housing with structures of five or more units (Sampson, 1983). Other studies have confirmed that the percentage of multiple-unit dwellings and renter-occupied housing are major predictors of crime rates in urban neighborhoods (Roneck, 1981; Scheurman and Kotrin, 1986).

Overall, these available studies suggest that a mix of neighborhood characteristics contribute to criminal offending. However, covariation among predictor variables and the use of cross-sectional designs in most studies limits the conclusions that can be drawn regarding causality. Two recent studies have sought to overcome these barriers. The first investigated changes in community structure and violent crime in Baltimore neighborhoods from 1970 to 1980 (Taylor and Covington, 1988). It found that neighborhoods with increasing concentrations of poor persons also experienced increasing violence rates. Less expected was an increased level of violence in gentrifying neighborhoods—those with increasing proportions of affluent persons (measured by owner-occupied housing and family income). The authors conclude that violence is associated with neighborhood change but that the underlying mechanisms are different for disparate neighborhoods. In increasingly poor neighborhoods, violence appears to be related to increased relative deprivation; in gentrifying neighborhoods, violence may be related to increased social disorganization.

The second study examined variation in homicide rates among SMSAs and cities over a 30-year period (Land et al., 1990), using common structural and demographic predictors discussed above. The analysts found that resource deprivation (a composite measure of poverty concentration and income stratification) had the largest effect on the homicide rate. All else being equal, cities with a large poor population and a high percentage of blacks, in conjunction with a high percentage of single-parent families and high divorce rates, had disproportionately high homicide rates. These predictors were more powerful for the 1970-1980 period, suggesting that increasing the concentration of disadvantaged persons within stratified neighborhoods has contributed to the recent increase in violent crime (Land et al., 1990).

Metropolitan neighborhoods in the United States, in both central cities and suburban areas, are characterized by a high degree of stratification by family income, race, and ethnicity. During the 1970s, there was an increase in the number of very poor neighborhoods and a substantial growth in underclass neighborhoods. Consequently, more children and adolescents are growing up in neighborhoods in which high proportions of adults are poor, unemployed, on welfare programs, or single parents. Blacks are most likely to be living in these neighborhoods, followed by Hispanics, Asians, and whites. In general economic and social integration is occurring in smaller metropolitan areas outside the Northeast and Midwest and in areas in which there are relatively small minority populations, but high levels of stratification still exist throughout the country.

Until analyses of 1990 census data are completed, trends during the 1980s cannot be confirmed. However, with overall trends toward higher poverty and unemployment, combined with the deteriorating labor market for young workers, it is likely that levels of concentrated poverty and residential segregation by class and race remained, at best, at 1980 levels.

Relatively few studies rigorously examine the hypothesis that variation in neighborhood composition is a causal factor in adolescent development. However, those few studies confirm that stratified neighborhoods independently contribute to dropping out of school, teenage parenthood, violent crime, and victimization. The economic and occupational mix of neighborhoods appears to have a stronger influence on adolescent development than racial

composition. However, it also appears that economic factors affect black and white adolescents in different ways and affect blacks more strongly. The data also suggest that adolescents may be differentially affected by different levels of neighborhood poverty: very poor neighborhoods differ both economically and socially from other neighborhoods, and these differences are likely to affect adolescent behavior.

Some youth do prosper in such disadvantaged settings, "superkids" who were sheltered from the influence of street life by parents and community institutions or were able to establish important relationships with other neighborhood adults (Williams and Kornblum, 1985). But in far too many cases, adolescents are overwhelmed by the worst aspects of their neighborhood settings.

Alsalem, N., L.T. Ogle, G.R. Rogers, and T.M. Smith 1992 The Condition of Education. Washington, D.C.: U.S. Department of Education.

Anderson, E. 1990 Neighborhood effects on teenage pregnancy. Pp. 375-398 in C. Jencks and P.E. Peterson, eds., The Urban Underclass. Washington, D.C.: The Brookings Institution.

1991a Alienation and Crime Among the Ghetto Poor. Unpublished manuscript, Department of Sociology, University of Pennsylvania.

1991b Streetwise: Race, Class and Change in an Urban Community. Chicago: University of Chicago Press.

Beasely, R.W., and G. Antunes 1974 The etiology of urban crime: an ecological analysis. Criminology 11:439-461.

Block, R. 1979 Community, environment, and violent crime. Criminology 17:46-57.

Brooks-Gunn, J., G. Duncan, P. Kato, and N. Sealand 1993 Do neighborhoods influence child and adolescent behavior. American Journal of Sociology, in press.

Corcoran, M., R. Gordon, D. Laren, and G. Solon 1987 Intergenerational transmission of education, income and earnings. Unpublished manuscript, University of Michigan, Ann Arbor.

Crane, J. 1991a Effects of neighborhoods on dropping out of school and teenage childbearing. In C. Jencks and P.E. Peterson, eds., The Urban Underclass. Washington, D.C.: The Brookings Institution.

1991b The epidemic theory of ghettos and neighborhood effects on dropping out and teenage childbearing. American Journal of Sociology 5:1226-1259.

Datcher, L. 1982 Effects of community and family background on achievement. Review of Economics and Statistics 64:32-41.

Eisenhower Foundation (Milton S.) 1990 Youth Investment and Community Reconstruction: Street Lessons on Drugs and Crime for the Nineties. Washington, D.C.: Milton S. Eisenhower Foundation.

Farley, R. 1990 Residential segregation of social and economic groups among blacks, 1970-89. In C. Jencks and P.E. Peterson, eds., The Urban Underclass. Washington, D.C.: The Brookings Institution.

Farrel, E. 1990 Hanging In and Dropping Out: Voices of At-Risk High School Students . New York: Teachers College Press.

Fernandez, R.M., and D. Harris 1990 Social Isolation and the Underclass. Unpublished manuscript, Center for Urban Affairs and Policy Research, Northwestern University.

Harrell, A. V., and G.E. Peterson, eds. 1992 Drugs, Crime, and Social Isolation: Barriers to Urban Opportunities. Washington, D.C.: Urban Institute Press.

Jargowsky, P., and M.J. Bane 1990 Ghetto poverty in the United States, 1970-1980. Pp. 235-273 in C. Jencks and P.E. Peterson, eds., The Urban Underclass. Washington, D.C.: The Brookings Institution.

Jencks, C., and S.E. Mayer 1990a The social consequences of growing up in a poor neighborhood. In L.E. Lynn and M. McGeary, eds., Concentrated Urban Poverty in America. Washington, D.C.: National Academy Press. 1990b Residential segregation, job proximity, and black job opportunities. In L.E. Lynn and M. McGeary, eds., Inner-City Poverty in the United States. Washington, D.C.: National Academy Press.

Judd, D.R. 1992 Segregation forever? The Nation 253(20):740-744.

Kirschenman, J., and K.M. Neckerman 1991 "We'd love to hire them, but …": the meaning of race for employers. Pp. 203-232 in C. Jencks and P.E. Peterson, eds., The Urban Underclass . Washington, D.C.: The Brookings Institution.

Land, K., P. McCall, and L. Cohen 1990 Structural covariates of homicide rates: are there any invariances across time and space? American Journal of Sociology 95:922-963.

Littell, J., and J. Wynn 1989 The Availability and Use of Community Resources for Young Adolescents in an Inner-City and a Suburban Community. Chicago: Chapin Hall Center for Children, University of Chicago.

Logan, J.R., and M. Schneider 1984 Racial segregation and racial change in American suburbs. American Journal of Sociology 89:874-878.

Massey, D., and N. Denton 1987 Trends in the residential segregation of Blacks, Hispanics and Asians: 1970-1980. American Sociological Review 52:802-825.

1988 Suburbanization and segregation in U.S. metropolitan areas. American Journal of Sociology 94(3):592-626.

1989 Hypersegregation in U.S. metropolitan areas: Black and Hispanic segregation along five dimensions. Demography 26(3):373-391.

Massey, D., and M. Eggers 1990 The ecology of inequality: minorities and the concentration of poverty. American Journal of Sociology 95:1153-1188.

Mayer, S.E. 1991 The Effect of School and Neighborhood Social Mix on Adolescents' Transitions to Adulthood. Commissioned paper prepared for the Panel on High-Risk Youth, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C.

Musick, J.S. 1991 The high-stakes challenge of programs for adolescent mothers. Pp. 111-138 in P.B. Edelman and J. Ladner, eds., Adolescence and Poverty: Challenge for the 1990s. Washington, D.C.: Center for National Policy Press.

National Commission on Children 1991 Speaking of Kids: A National Survey of Children and Parents. Washington, D.C.: National Commission on Children.

Ogbu, J.U. 1984 The Next Generation: An Ethnography of Education in an Urban Setting. New York: Academic Press.

Orfield, G. 1983 Public School Desegregation in the United States, 1968-1980. Washington, D.C.: Joint Center for Political Studies.

Orfield, G., and F. Monfort 1992 The Status of School Desegregation: The Next Generation. Alexandria, Va.: National School Boards Association.

Orfield, G., F. Monfort, and M. Aaron 1989 Status of school desegregation 1968-1986. In Segregation, Integration, and Public Policy: National, State, and Metropolitan Trends in Public Schools. Chicago: University of Chicago.

Rainwater, L. 1970 Behind Ghetto Walls: Black Families in a Federal Slum. Chicago: Aldine Press.

1987 Class, Culture, Poverty, and Welfare. Unpublished manuscript, Department of Sociology, Harvard University.

Reiss, A.J., and J.A. Roth, eds. 1993 Understanding and Preventing Violence. Panel on the Understanding and Control of Violent Behavior, Committee on Law and Justice, National Research Council. Washington, D.C.: National Academy Press.

Ricketts, E., and R. Mincy 1990 Growth of the underclass: 1970-1980. Journal of Human Resources 25(2):137-145.

Roneck, D. 1981 Dangerous places: crime and residential environment. Social Forces 60:74-96.

Sampson, R.J. 1983 Structural density and criminal victimization. Criminology 21:276-293. 1985 Neighborhood and crime: the structural determinants of personal victimization. Journal of Research in Crime and Delinquency 22:7-40.

Sampson, R.J., and J.L. Lauritsen 1991 Violent Victimization and Offending: Individual, Situational, and Community-Level Risk Factors. Unpublished manuscript, Panel on the Understanding and Control of Violent Behavior, University of Chicago.

Scheurman, L., and S. Kotrin 1986 Community careers in crime. Pp. 67-100 in A.J. Reiss, Jr. and M. Tonry, eds., Communities and Crime. Chicago: University of Chicago Press.

Schneider, M., and J.R. Logan 1982 Suburban racial segregation and black access to local public resources. Social Science Quarterly 63:762-770.

Skogan, W.G. 1990 Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods. New York: The Free Press.

Smith, D.R., and G.R. Jarjoura 1988 Social structure and criminal victimization. Journal of Research in Crime and Delinquency 25:27-52.

Sullivan, M. 1990 Getting Paid: Economy, Culture, and Youth Crime in the Inner City. Ithaca, N.Y.: Cornell University Press.

Taylor, R.L. 1991 Poverty and adolescent black males: the subculture of disengagement. Pp. 139-162 in P. Edelman and J. Ladner, eds., Adolescence and Poverty: Challenge for the 1990s. Washington, D.C.: Center for Public Policy Press.

Taylor, R., and J. Covington 1988 Neighborhood changes in ecology and violence. Criminology 26:553-590.

Visher, C., and J. Roth 1986 Participation in criminal careers. Pp. 211-291 in A. Blumstein et al., eds., Criminal Careers and Career Criminals. Washington, D.C.: National Academy Press.

Williams, T. 1989 The Cocaine Kids: The Inside Story of a Teenage Drug Ring. Menlo Park, Calif.: Addison-Wesley.

Williams, T., and W. Kornblum 1985 Growing Up Poor. Lexington, Mass.: Lexington Books.

Wilson, W.J. 1987 The Truly Disadvantaged: The Inner City, The Underclass, and Public Policy. Chicago: University of Chicago Press.