Neighborhoods and Self-Control: Toward an Expanded View of Socialization

Page 1

Social Problems, Vol. 56, Issue 1, pp. 205–222, ISSN 0037-7791, electronic ISSN 1533-8533. © 2009 by Society for the Study of Social Problems, Inc. All rights reserved. Please direct all requests for permission to photocopy or reproduce article content through the University of California Press’s Rights and Permissions website at www.ucpressjournals.com/reprintinfo/asp.DOI: 10.1525/sp.2008.56.1.205.

Neighborhoods and Self-Control: Toward an Expanded view of socialization

Brent Teasdale, Georgia State University

Eric Silver, The Pennsylvania State University

In this article we develop and test a multi-level theory of the sources of self-control among adolescents. We argue that neighborhoods are an important structural source of self-control. We test this idea using data from the National Longitudinal Study of Adolescent Health (n = 9,171). Results from a multilevel structural equation model indicate that neighborhood disadvantage is a significant predictor of adolescent self-control, controlling for demographics, family characteristics, and social integration. Implications for future research on the role of neighborhood context in the development of self-control among youth are discussed. In addition, we discuss the implications for policy of multilevel theorizing. Keywords: self-control, neighborhood disadvantage, socialization, multilevel analysis.

A good deal of criminological research has been devoted to examining Michael Gottfredson and Travis Hirschi’s (1990) general theory of crime (Burton et al. 1998; Grasmick et al. 1993; Longshore 1998; Longshore et al 1996; for a meta-analytic review see Pratt and Cullen 2000). Together these studies show that self-control is a robust predictor of criminal and other devi-ant behaviors. However, few studies have investigated the precursors of self-control (for ex-ceptions see Cochran et al. 1998; Gibbs et al. 1998; Hay 2001; Perrone et al. 2004; Polakowski 1994; Pratt et al. 2004), and the majority of these have focused exclusively on parenting practices. Thus little is known about the extent to which neighborhood conditions affect self-control (but see Pratt et al. 2004). To address this issue, the current study blends self-control theory with the communities and crime literature to develop a multi-level perspective on the development of self-control. Specifically, we argue that neighborhood disadvantage influenc-es self-control by decreasing the availability of social capital that parents and neighbors may draw on in their efforts to socialize local youth. To the extent that neighborhoods facilitate or inhibit the socialization efforts of parents (Sampson 1997), and parenting is important in the formation of self-control (Gottfredson and Hirschi 1990), it seems logical to propose that neighborhoods play a role in the formation of self-control. That is, we propose that neighbor-hoods influence self-control by creating environments in which the socialization activities of parents may be more or less successful.

This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill, NC 27516–2524 ([email protected]). No direct support was received from grant P01-HD31921 for this analysis. A previous version of this article won second place in the American Society of Criminology’s Gene Carte Student Paper Competi-tion. The authors would like to thank Wayne Osgood for statistical advice and several anonymous reviews for helpful feedback. Direct correspondence to: Brent Teasdale, Department of Criminal Justice, Georgia State University, Atlanta, GA 30302. E-mail: [email protected].

SP5601_11.indd 205 1/8/09 1:58:32 PM

Page 2

206 Teasdale/silver

Multilevel theorizing is important not only for our understanding of the etiology of ado-lescent deviance, but has important implications for policy making. We argue that policy deci-sions about social problems, such as crime, require a multilevel understanding, such as that provided in the current examination. That is, individual-level theories of social problems, such as self-control, necessarily focus policy solutions at the individual-level. Consequently, policies such as incapacitation are favored and prison populations rise. In contrast, if an in-dividual’s capacity for self-control is influenced in part by neighborhood context, a different type of policy intervention is made plausible. Multilevel theorizing points policy makers to multilevel interventions at the community, family, and individual levels. We argue that a complete understanding of the problem, and consequently better policy interventions, require this contextual viewpoint, a point we return to in greater detail in our discussion.

In the current study, we use data from the National Longitudinal Study of Adolescent Health (Add Health) to test whether such a contextual view of self-control is warranted. Add Health is a nationally representative study of adolescents in grades 7 through 12. The data include 1990 Census variables that allow us to construct objective measures of the neighbor-hood context. In addition, the data contain indicators of self-control taken from the child’s perspective. Finally, the data contain information on a wide range of theoretically relevant predictors of self-control, including parental network involvement, intergenerational closure, parental availability, parental bonding, and social integration (e.g., social ties).

Self-Control Theory

Gottfredson and Hirschi’s (1990) self-control theory is perhaps the most prominent individual-level theory in criminology (Pratt and Cullen 2000). According to Gottfredson and Hirschi, self-control is a stable individual characteristic established early in life that can account for crime and its correlates (i.e., low SES [socioeconomic status], gender, deviant peers, etc.). As such, this single personality characteristic is offered as a replacement for other criminologi-cal perspectives. According to Gottfredson and Hirschi, all other theories of crime, including macro-level theories, are unnecessary because they may be derived from the proposition that criminals have low self-control.

Gottfredson and Hirschi describe people with low self-control as “impulsive, insensitive, physical (as opposed to mental), risk-taking, short sighted, and nonverbal” (p. 90). They argue that self-control is fixed by about the age of seven or eight, prior to the age of responsibility for crime and that it tends to persist throughout life (p. 272). Gottfredson and Hirschi further explain that self-control is inversely related to criminal behavior because crime is an activity that produces immediate gratification and generally requires few skills and little advanced planning. That is, crime is the kind of activity that is likely to be engaged in by people with low self-control.

The notion that self-control is related to crime and analogous behaviors has received much empirical support in the criminological literature (Chapple 2005; Chapple, Tyler, and Bersani 2005; Doherty, 2006; Feldman and Weinberger 1994; Grasmick et al. 1993; LaGrange and Silverman 1999; Latimore, Tittle, and Grasmick 2006; Longshore 1998; Long-shore and Turner 1998; Longshore, Chang, and Messina 2005; Longshore, Turner, and Stein 1996; Vazsonyi and Crosswhite, 2004; Vazsonyi et al. 2001, 2004; for meta-analytic reviews see Miller and Lynam 2001; Pratt and Cullen 2000). In addition, support has been found for the theory’s prediction that the relationship between self-control and offending is con-ditioned by opportunity (Burton et al. 1998; Longshore 1998; Nakhaie, Silverman, and La Grange 2000), as well as the idea that self-control is correlated with negative life outcomes other than crime, including friendship quality, family relations, and educational attainment (Evans et al. 1997).

SP5601_11.indd 206 1/8/09 1:58:32 PM

Page 3

Neighborhoods and self-Control 207

For Gottfredson and Hirschi, socialization is the key cause of self-control (p. 97). That is, high self-control develops only in the presence of good parenting. They base this idea on two early studies that found that parental discipline, supervision, and affection were lacking in the homes of delinquent youth (Glueck and Glueck 1950; McCord and McCord 1959). Based on these studies, Gottfredson and Hirschi (p. 97) conclude that there are three necessary condi-tions for adequate child rearing: (1) monitoring the child’s behavior; (2) recognizing misbe-havior when it occurs; and (3) punishing misbehavior. Thus, their theory is one of individual differences caused primarily by inadequate parental socialization, resulting in a propensity toward crime that remains stable throughout life.

Gottfredson and Hirschi’s suggestions about the role of socialization in the production of self-control have found much support (Cochran et al. 1998; Gibbs et al. 1998; Hay 2001; Hope et al. 2003; Polakowski 1994; Pratt et al. 2004). Specifically, Trina Hope, Harold Grasmick, and Laura Pointon (2003) find that parent’s education, supervision, and attachment are all significant predictors of adolescent self-control. Carter Hay (2001) finds parents who provide monitoring and discipline have high school students with higher self-control, but he moves beyond the Gottfredson and Hirschi explanation for self-control by arguing that self-control is “explained more thoroughly when this broader range of parenting variables is considered” (p. 721). This broader range of parenting variables refers to an authoritative style of parenting that includes acceptance-involvement, psychological autonomy, fair discipline, and nonphysical discipline (Hay 2001).

An Expanded View of Socialization

While appreciating the parsimony of their explanation, we argue that Gottfredson and Hirschi (1990) take an unnecessarily narrow view of socialization by restricting it primarily to processes occurring in the family. In contrast, we find it useful to view self-control as an outgrowth of social processes that extend beyond the family to include the neighborhood contexts within which children and families live. To the extent that neighborhood condi-tions influence the likelihood that youth misbehavior will be detected and dealt with by parents and adults outside the family (Sampson 1997; Sampson and Groves 1989), it is reasonable to posit that neighborhood characteristics will play a role in the development of self-control.

Gottfredson and Hirschi (p. 272) argue that self-control is fixed at age seven or eight. However, several recent studies suggest that self-control may not be so stable (Burt, Simons, and Simons 2006; Tittle and Grasmick 1997:341; Wright et al. 1999:496, but see Hay and For-rest 2006 for a counter point). Using data from the Thirteenth Annual Oklahoma City Survey, Tittle and Grasmick (1997) found significant variation across age categories in self-reported levels of self-control. Although their study was not longitudinal and therefore could not con-trol for cohort effects, their finding is consistent with the notion that self-control changes as children age. A second study by Bradley Wright and colleagues (1999) found small but significant correlations between childhood and adolescent measures of self-control (ranging from r = .02 to r = .26) using self-, parent-, and teacher-reports of self-control. Similarly, Callie Burt, Ronald Simons, and Leslie Simons (2006) find that “individuals’ rankings in self-control were substantially altered between the two waves” (p. 375). In addition, Burt and colleagues (2006) conclude that their study fails to support Gottfredson and Hirschi’s (1990) strict stabil-ity postulate. In contrast to these studies, Hay and Walter Forrest (2006) conclude that there is “strong absolute and relative stability of self-control for more than 80 percent of the sample” (p. 739). Importantly for our purposes, however, they also find that “parental socialization continued to affect self-control during adolescence, even after accounting for both prior self-control and exposure to parental socialization” (pp. 739–40). This indicates that social influ-ences in adolescence (parents and possibly other sources of socialization) may continue to have impacts beyond those present in childhood.

SP5601_11.indd 207 1/8/09 1:58:32 PM

Page 4

208 Teasdale/silver

These results on average contradict the claim that rankings of self-control are stable over the life course. In short, Gottfredson and Hirschi’s position on the stability of self-control throughout life appears overstated (for additional discussion, see Sampson and Laub 1993). The stability issue is important for the current project because to the extent that self-control is less stable than originally posited, consideration of the social processes within and outside of the family that may influence the development of self-control during later childhood and adolescence are worthy of examination. We believe the body of literature on this point sup-ports the idea that influences on self-control in adolescence are worthy of examination (see Hay and Forrest 2006 on this point).

The current study thus rests on a broader conception of the factors that influence self-control than was originally formulated by Gottfredson and Hirschi (see also Hay 2001). Specifi-cally, we argue that self-control is influenced by both family and neighborhood level processes, and that such influences persist into late childhood and adolescence. Based on these assump-tions, we examine whether neighborhood-level social processes influence levels of self-con-trol among adolescent youth.1

A number of studies have examined the role of parenting in the creation of self-control (Cochran et al. 1998; Gibbs et al. 1998; Hay 2001; Polakowski 1994; Pratt et al. 2004). In contrast, only two prior studies have examined the link between self-control and neighbor-hood characteristics. In the first study, based on a large sample of 13-year-old inner-city boys, Donald Lynam and colleagues (2000) found a significant, though modest, association between neighborhood disadvantage (measured as a scale combining poverty, unemployment, house-hold income, single-parent families, public assistance, and percent black in the census tracts where subjects lived) and self-control (measured as a series of 11 self-, parent-, and teacher-reported items tapping impulsivity). Their results showed that self-control and neighborhood disadvantage were related at the bivariate level; however, the study was not aimed at under-standing the sources of self-control. Consequently, Lynam and colleagues did not examine the independent effect of neighborhood disadvantage on self-control within a multivariate framework controlling for other relevant factors, such as parenting, nor did they attempt to explain theoretically why low self-control was more prevalent among youth living in socio-economically disadvantaged neighborhoods.

The second study to investigate neighborhood problems as a source of self-control was conducted by Travis Pratt, Michael Turner, and Alex Piquero (2004). Pratt and colleagues argue that neighborhood context affects the development of self-control in children through its effect on parenting. However, their analyses of the National Longitudinal Study of Youth (NLSY) provided only partial support for this hypothesis. First, somewhat surprisingly, they found that mothers are more likely to monitor and discipline their children in neighborhoods where they perceive more problems (such as people not having respect for the rules, too many parents not supervising their children, and people not caring about what goes on in the neigh-borhood). Consequently, parenting practices (monitoring and discipline) do not explain why children in more problematic neighborhoods have lower levels of self-control. In addition, Pratt and colleagues found that a mother’s perception of the neighborhood as problematic was inversely related to her child’s self-control (as rated by the mother on the behavior problems index [BPI]). Although supportive of the notion that neighborhood context influences the de-velopment of self-control in youth, the measures used by Pratt and colleagues leave much to be desired. This is because both the neighborhood context and the child’s level of self-control were measured via the same source, namely the mother. Thus, to the extent that mothers who view their neighborhoods as problematic also tend to view their children as problematic,

1. We acknowledge that this is not a test of Gottfredson and Hirschi’s ideas about the sources of self-control be-cause we lack data on children before the age of eight. Therefore, their thesis that self-control is solely caused by factors originating within the family home cannot be refuted by this study. We argue that extension may be necessary and that future research should consider neighborhood variables as predictors of a child’s level of self-control.

SP5601_11.indd 208 1/8/09 1:58:33 PM

Page 5

Neighborhoods and self-Control 209

we cannot be certain of the validity of the strength of the relationship observed. Additionally, the measure of self-control is based on the behavioral problem index, which reflects concerns not traditionally associated with the general theory of crime, such as mood swings, feeling unloved, being tense, nervous, or anxious, etc. (for a complete list of items, see Pratt et al. 2004).

Neighborhood Structure and the Socialization of Youth

Sociologists have long been interested in the effect of social structure on the socialization of youth (Elliott et al. 1996; Sampson 1997; Shaw and McKay [1942]1969). In their seminal studies of juveniles in Chicago, Clifford Shaw and Henry McKay ([1942]1969:171) called attention to differences in norms and standards of behavior associated with neighborhood location. They argued that children growing up in economically disadvantaged, residentially mobile, and ethnically heterogeneous neighborhoods were likely to be exposed to “compet-ing ways of life” as they encountered examples of conventional and unconventional lifestyles in the neighborhood. Shaw and McKay found that poverty, residential mobility, and racial/ethnic heterogeneity were associated with higher rates of juvenile delinquency and argued that these structural conditions severely undermined the ability of neighborhood residents to transmit conventional codes of conduct to local youth, a situation they referred to as social disorganization.

More recently, Robert Sampson and colleagues have provided further specification of the mechanisms that mediate the link between social structure and socialization by focusing on the differential distribution across neighborhoods of (1) informal social controls (Sampson and Groves 1989; see also Bursik 1988); (2) social capital (Sampson 1992, 1997); and (3) col-lective efficacy (Morenoff et al. 2001; Sampson et al. 1999). Each of these mechanisms has been conceptualized as enhancing the capacity of local residents to monitor and supervise youth in the neighborhood.

For almost two decades, informal social control, or the ability of residents to monitor teenage peer groups, has been conceptualized in the criminological literature as rooted in neighboring activities and social ties between community residents (Bellair 1997; Bursik 1988; Sampson and Groves 1989; Silver and Miller 2004). For example, in a pair of studies of over 11,000 residents measured as part of the 1982 and 1984 British Crime Surveys, Sampson and Byron Groves (1989) found that social ties between friends and neighbors mediated a significant amount of the association between neighborhood conditions and crime victim-ization. They interpreted this finding as indicating that social ties among neighbors facilitate neighborhood social controls. Paul Bellair (1997) further examined the relationship between social ties and informal social control using a sample of 60 urban neighborhoods in the United States. He found that frequent (e.g., strong) social ties among neighbors produced the great-est informal social control effect when accompanied by infrequent (e.g., weak) social ties. In addition, a recent study by Delbert Elliott and colleagues (1996) based on a sample of 1,045 families from Denver and Chicago found that a measure of informal social control mediated the associations between neighborhood disadvantage and the following outcomes: adolescent prosocial competence, numbers of conventional friends, and adolescent problem behaviors. Informal social control thus holds a central place in the communities and crime literature as a key mechanism influencing youth behavior.

Social capital refers to the availability of social resources within a community that tran-scend and condition “the more proximate family processes and relationships observed inside the home” (Sampson et al. 1999:634; see also Bourdieu 1986; Coleman 1988). Sampson, Jeffrey Morenhoff, and Felton Earls (1999) conceptualize social capital in terms of the fol-lowing two dimensions: (1) intergenerational closure is the extent to which adults and children in a community are linked to one another; and (2) reciprocated exchange is the intensity of

SP5601_11.indd 209 1/8/09 1:58:33 PM

Page 6

210 Teasdale/silver

interfamily and adult interaction with respect to childrearing.2 Together, these dimensions constitute important mechanisms through which neighborhood social networks contribute to the socialization of youth. To this conceptualization, Sampson and colleagues (1999) add ex-pectations for informal support and mutual control of children (or “collective efficacy for children”), which they define as the “shared expectations and mutual engagement by adults in the active support and social control of children” (Sampson et al. 1999:635). Although the concepts of social capital and collective efficacy for children are conceptually related, they are distinguish-able in that the latter refers to the process of activating social ties to achieve desired outcomes, while the former refers to the ties themselves. Together, however, these concepts provide a theoretical basis for extending the concept of socialization beyond what goes on directly under the family’s roof to include social factors rooted in the neighborhood.

The Current Study

We argue that informal social controls, social capital, and collective efficacy contribute to the development of self-control in youth by increasing the likelihood that neighborhood residents will participate in the socialization of children in the local area. Deficits in these areas are theorized to lead to lower levels of social regulation, which in turn lead to higher rates of unmonitored, and therefore unpunished, deviant behavior (Sampson 1997). To the extent that monitoring and punishment of youth misbehavior are proximal mechanisms through which self-control is formed (Gottfredson and Hirschi 1990), then neighborhoods that provide less collective regulation should be more likely to produce children with low self-control. In short, this study hypothesizes the following:

Neighborhoods with greater socioeconomic disadvantage will produce adolescents with lower (1) levels of self-control, controlling for other theoretically relevant predictors of self-control.The relationship between neighborhood disadvantage and self-control will be mediated, in (2) part, by family socialization activities and social regulation.

Methods

Data

The data for this study come from Waves One and Two of the National Longitudinal Study of Adolescent Health (Add Health) (Bearman, Jones, and Udry 1997). The first wave of data for Add Health was collected in 1995, using a clustered sampling design consisting of students nested within schools. To be eligible for sampling, a school was required to have an eleventh-grade class and a total enrollment of more than 30 students. The initial sample was represen-tative of U.S. schools with respect to region of the country, urbanity, school type (e.g., public versus private), ethnicity, and size. In 1996, students who participated in the initial in-home interview (Wave One) completed a second in-home interview (Wave Two). The response rate for the in-home sample was 75 percent for Wave One. Approximately 88 percent of the Wave One sample participated in the Wave Two follow-up. Students who were in the twelfth grade in 1995 (at Wave One) and those who were part of special over-samples were not reinter-viewed for Wave Two (Bearman et al.1997).

Several questionnaires were part of the Add Health study. The first was the in-school questionnaire, administered during one 40 to 60 minute class period. The second was an interview of adolescents in their own homes. These questionnaires took from one to two

2. Sampson’s conception of social capital is clearly more related to Coleman’s concept of social capital than that of Bourdieu.

SP5601_11.indd 210 1/8/09 1:58:33 PM

Page 7

Neighborhoods and self-Control 211

hours to complete. All students in a school were eligible for inclusion in the in-home sample, but not all were interviewed. The in-school survey was shorter and contained fewer questions than the in-home survey. For this reason, the in-home interview data are used in the current analyses. Students were chosen for the in-home interview based on a stratified sampling tech-nique. Within each school, students were stratified by grade and sex. Approximately 17 stu-dents were randomly selected from each stratum resulting in the selection of approximately 200 adolescents from each of the 80 pairs of schools (Bearman et al. 1997). In addition to the in-school and in-home student surveys, one parent of each subject in the in-home sample also completed a questionnaire. The parent questionnaire provided information about the family’s socioeconomic status and parent networking behaviors. Finally, census data for each individual’s block group were appended to each record in the data set. This information is used in the current analyses to measure block-level socioeconomic disadvantage.

Simulation studies by Cora J. M. Maas and Joop J. Hox (2005) suggest that group size has relatively little influence on estimates, when the number of groups is large (greater than 50). Since we have more than 50 level two units, the small number of individuals per level two unit should not bias estimates, per Maas and Hox. We limited our sample to subjects who resided in block groups in which at least four other subjects also resided. Using this criterion, our analysis sample includes 9,171 subjects (down from 13,412), residing in 711 block groups.3

Measurement

Self-Control. We measure self-control following the approach taken by Dina Perrone and colleagues (2004).4 Using Add Health data, they measure self-control using five items tapping the orientation to simple tasks, physical activities, and impulsivity dimensions of self-control (Perrone et al. 2004). The items ask about “problems keeping your mind on what you were doing,” “trouble getting your homework done,” “difficulty paying attention in school,” and “trouble getting along with teachers.” Higher scores on this scale reflect higher levels of self-control; lower scores reflect short attention span, lack of planning, insensitivity to normative expectations, and an unwillingness to delay gratification, all of which are part of Gofftredson and Hirschi’s (1990) conception of low self-control. It is important to note that this scale was measured using the Wave Two data and is therefore appropriately separated in time from the independent measures, all of which were measured at Wave One. This variable was measured as a latent factor with four observed indicators (alpha = .67). The Perrone and colleagues’ item regarding “feeling like you are doing everything just about right” was dropped, because it did not load with the other four items in our analysis.

Neighborhood Disadvantage. We used the census block group to represent respondents’ neighborhoods. Census block groups contain approximately 400 housing units (range: 250 to 550), and are the smallest spatial units about which the U.S. Census Bureau collects informa-tion. Block groups are based on physical boundaries (such as railroad tracks, major roads, and streams) and are designed to approximate the usual conception of a neighborhood.

Neighborhood socioeconomic disadvantage was measured using the following four 1990 U.S. Census items: proportion persons in 1989 with income below the poverty level, propor-tion households with public assistance income, unemployment rate, and percent female-headed households. Higher scores on this scale indicate greater socioeconomic disadvantage (alpha = .84).5 This variable was measured as a latent factor based on the four observed indicators.

3. The mean number of subjects within block groups was 13, ranging from 5 to 101. The median was 10. There were no significant differences with respect to age or gender between the original and the reduced sample. However, the reduced sample contained significantly fewer white subjects than the original sample.

4. See Appendix A for the principal components analysis results.5. The same scale based on the census tract yielded results similar to those reported below.

SP5601_11.indd 211 1/8/09 1:58:33 PM

Page 8

212 Teasdale/silver

Social Regulation. Drawing on the communities and crime literature, we constructed three scales designed to tap social regulation: social integration (e.g., social ties), parental involvement in organizations, and intergenerational closure. The social integration scale measures the extensive-ness of local social ties as perceived by the adolescent. We created this scale as a latent factor with three observed indicators (alpha = .60) that asked for the respondent’s level of agreement with the following statements: “You know most of the people in your neighborhood”; “In the past month, you have stopped on the street to talk with someone who lives in your neighbor-hood”; “People in this neighborhood look out for each other.” Response categories for these items were 1 = true and 0 = false.

The parent involvement in organizations scale consists of two items from the Wave-One parent interview. Parents were asked about their involvement in “parent-teacher organiza-tions” and “civic or social organizations.” These two items were summed to create the parental involvement in organizations index (alpha = .39).

In order to tap intergenerational closure, parents were asked if they could name the school that their child’s best friend attended, if the parent had met their child’s best friend, and if the parent had met the parent’s of their child’s best friend. This variable was created as a latent factor with the three observed indicators (alpha = .64).

Parental Controls. In addition to social integration and parent network involvement, we created indices of parental availability and attachment. Parental attachment and availability are important variables to control for in the current analyses, because Gottfredson and Hirschi (1990) argue that attachment and availability for supervision are necessary for socializing children with high self-control. That is, these measures reflect the “under the roof” paren-tal socialization that Gottfredson and Hirschi propose should explain the creation of self-control.

We measure parental availability based on the frequency with which each parent was physically present in the home at three points in the day: when the adolescent left for school in the morning, returned home in the afternoon, and went to bed in the evening. Responses ranged from 1 = never to 5 = always. We took the mean of these six items (three for people with one parent), with higher scores on the scale indicating more parental availability for supervision.

We measure parent attachment using three items from the Wave One interview that re-flect emotional closeness to parents (alpha = .66). This variable was created as a latent factor with three observed indicators, including two questions measuring how close the subjects felt to their mothers and fathers, and one question asking how much their mothers cared about them. An important weakness of our study is that we do not have access to a measure of pa-rental self-control. However, Gottfredson and Hirschi suggest that emotional attachments are less likely to form to parents with low self-control. Thus our measure of parental attachment may, in part, reflect (and thus help us to control for) the self-control of parents.

Control Variables. We include age, race, sex, family structure, and socioeconomic status as controls, so that relationships among the theoretically relevant variables and self-control are not confounded by demographic differences in the sample. Race was measured by three dummy variables, including Hispanic, black, and other, with white as the omitted refer-ence category. Other control variables include a dichotomous variable representing sex (1 = female), age measured in years, and length of residence measured in years. We measured family structure as a dichotomous variable coded 1 if the individual lived with both biological parents and 0 if not.

Family socioeconomic status was measured by parents’ self-reports of their educational attainment and adolescent’s reports of their parents’ occupational prestige (for both moth-er and father, when available). Occupational prestige was defined using the U.S. Census

SP5601_11.indd 212 1/8/09 1:58:33 PM

Page 9

Neighborhoods and self-Control 213

occupational group codes. Educational attainment was measured in years of schooling com-pleted. These four items were z-scored and the mean of parent’s education and occupational status was taken to create the SES index (or two items if the adolescent had only one parent). Family SES is an important variable to include for several reasons. First, family SES is the primary mechanism sorting families into neighborhoods. Second, parent’s SES may, in part, reflect par-ents’ level of self-control. Because the selection argument—that parents with low self-control produce low self-control in their offspring and cause them to self-select into disadvantaged communities—is the primary alternative to the theory we propose, it is crucial that parent’s SES be included in the analyses.

Data Analysis

The clustered sample design of the Add Heath data introduces a problem of statisti-cal dependence, which may produce correlated errors in a regression model. Specifically, regression residuals may be correlated because adolescents are clustered within neighbor-hoods. This correlation violates the OLS regression assumption of independent observa-tions. To correct this problem, we use a multilevel structural equation model estimated in Mplus (version 5.0) to analyze the data (Muthén and Muthén 2006). In addition to the usual individual-level residual term (σ2), these models include a second residual term for the neighborhood-level units (µ

0). This neighborhood-level error term allows for a mean

level of error common to all individuals within a block group thus ensuring unbiased stan-dard errors and significance tests.6 In addition, the latent variable framework of Mplus allows us to model both a measurement model and a two-level structural model. This allows us to take into account measurement error in the multilevel structural equation models we present below.

Results

Descriptive statistics for the sample are provided in Table 1. As shown, the sample is evenly split between males (49 percent) and females (51 percent) and mostly white (52 percent). The average respondent is approximately 16 years of age and has lived at their current resi-dence for approximately 7.6 years. The majority of respondents live with both biological par-ents (63 percent). Correlations are presented in Table 2. As shown in Table 2, neighborhood disadvantage shows a small, but statistically significant bivariate association with self-control (r = -.045).

The first step in examining the association between neighborhood disadvantage and low self-control was to examine whether significant variation existed in self-control across the neighborhood-level units. Before a neighborhood-level variable can be related to an out-come, the outcome must vary significantly across neighborhoods. Therefore, we estimated a null model in HLM (version 6.04) to determine if there was significant variance across block groups in self-control. The variance component from this analysis (µ

0 = 0.013) indicated that

self-control does vary significantly across block-groups (p < .001). Approximately 5 percent of the variance in self-control is between neighborhoods.

The next step in our analysis was to include neighborhood socioeconomic disadvantage as a predictor of self-control in a multilevel structural equation model in Mplus (version 5.0).

6. All regression models were reestimated using STATA (version 9) and HLM (version 6.04) in order to take into account design effects. Findings were consistent in terms of direction and significance across statistical procedures. We report the results from the Mplus models because these allow us to simultaneously model the multilevel nature of the data and include a measurement model.

SP5601_11.indd 213 1/8/09 1:58:33 PM

Page 10

214 Teasdale/silver

Table 3 presents the results for the two-level structural portion of the model. A diagram of the estimated model is shown in Appendix B. As shown in Model 1 of Table 3, neighborhood disadvantage was significantly related to self-control at the bivariate level. Additionally, neighborhood disadvantage alone accounted for 27 percent of the neighborhood variance in self-control.

To ensure that this bivariate relationship was not spurious due to demographic factors, we included individual- and family-level predictors of self-control. As shown in Model 2 of Table 3, neighborhood disadvantage remained significantly related to adolescent self-control, after controlling for demographics, family factors, and social regulation variables. Individuals who resided in more disadvantaged neighborhoods had lower self-control than individuals resid-ing in less disadvantaged neighborhoods. Also consistent with our community perspective, adolescents from neighborhoods with higher levels of social integration and lower levels of residential mobility exhibited significantly higher levels of self-control. This model exhibited acceptable fit to the data. The CFI value was .922. The TLI value was .900. The RMSEA value was .027 and the SRMR value was .029.

Other important predictors of self-control were sex, age, race, family structure, and pa-rental attachment. Females, older adolescents, and African Americans exhibited lower levels of self-control. Adolescents from two parent households reported significantly higher lev-els of self-control than did adolescents from other family types. Consistent with Gottfredson and Hirschi’s (1990) emphasis on parenting, adolescents with stronger parental attachments exhibited higher levels of self-control than those with weaker parental attachments. Also consistent with Gottfredson and Hirschi’s arguments, parental availability for supervision was associated with higher levels of self-control (net of family structure, family SES, and parental attachment). Including these variables increased the explained variation in self-control across block groups to 60 percent (µ

0 decreased from .013 to .005 between the null model and Model

2 of Table 3).In sum, controlling for demographics, family factors, and social regulation, neighbor-

hood disadvantage, residential mobility, and social integration were found to influence self-control among adolescents in this sample. Moreover, comparing the disadvantage co-efficients in Models 1 and 2 reveals a 36 percent decrease in the disadvantage-self-control relationship after controlling for these measures. Thus, part of the reason why adolescent self-control was lower in more socioeconomically disadvantaged neighborhoods was that these neighborhoods had lower levels of social integration and lower levels of parental at-tachment and supervision.

Table 1 • Descriptive Statistics (observed variables)

Variable Mean SD Range

Person level

DemographicsFemale .51 .50 .00–1.00White .52 .50 .00–1.00Blacka .19 .39 .00–1.00Othera .05 .29 .00–1.00Hispanica .24 .43 .00–1.00Age 15.83 1.57 11.55–20.64Length of residence 7.56 5.43 .00–20.52

Family characteristicsFamily SES -.04 .89 -1.99–1.58Both biological parents .63 .48 .00–1.00Parent availability 2.36 .70 1.00–5.00

aWhite is the reference category.

SP5601_11.indd 214 1/8/09 1:58:33 PM

Page 11

Tab

le 2

• C

orre

lati

ons Se

lf-

Con

trol

Dis

adva

nta

geM

obil

ity

Fem

ale

Bla

ckO

ther

His

pan

icA

geL

engt

hF

amil

y SE

SB

oth

Bio

Par

ent

Att

ach

Ava

ilab

ilit

ySo

cial

Inte

grat

ion

Par

ent

Net

wor

k In

volv

emen

t

Sel

f-co

ntr

ol

—D

isad

van

tage

-.045**

—R

esid

enti

al

mobi

lity

-.028**

.074**

*—

Fem

ale

-.086**

.022**

-.003

—B

lack

-.042**

.525**

*.0

79**

*.0

21**

—O

ther

-.011

-.083**

*-.

070**

*-.

015*

-.223**

*—

His

pan

ic-.

031**

-.105**

*-.

141**

*-.

016*

-.302**

*-.

547**

*—

Age

.023**

-.016*

.010

-.045**

*-.

030**

*.0

77**

*.1

02**

*—

Len

gth

of

resi

den

ce.0

20**

*-.

019**

.199**

*-.

015*

-.043**

*-.

047**

*-.

083**

*.0

87**

*—

Fam

ily

SE

S.0

32**

-.028**

*-.

004

-.011

-.001

-.057**

*-.

161**

*-.

048**

*.0

64**

*—

Both

bio

.037**

-.233**

*.0

38**

*-.

031**

*-.

251**

*.0

54**

*.0

37**

*-.

046**

*.2

20**

*.0

68**

*—

Par

ent

atta

chm

ent

.176**

*-.

036**

*-.

018**

*.0

26**

*-.

042**

*.0

42**

*.0

57**

*.1

50**

*-.

008

-.062**

*-.

027**

*—

Par

ent

avai

labi

lity

.062**

*-.

039**

*.0

27**

.041**

*.0

03

-.016

-.061**

*-.

024**

.026**

.202**

*.1

03**

*.0

74**

*—

Soci

al in

tegr

atio

n.0

45**

*-.

044**

*-.

133**

*.0

60**

*-.

074**

*.1

15**

*.1

36**

*.0

99**

*-.

204**

*.0

27**

*-.

046**

*.0

71**

*.0

29**

—Par

ent

net

work

in

volv

emen

t.0

02

-.066**

*.0

51**

*.0

16*

-.073**

*-.

108**

*-.

168**

*-.

136**

*.1

22**

*.2

46**

*.1

56**

*-.

040**

*.0

36**

*-.

103**

*—

Inte

rgen

erat

ion

al

closu

re.0

21*

-.123**

*.0

38**

*.0

27**

-.102**

*-.

092**

*-.

132**

*-.

027**

*.0

92**

*.1

54**

*.1

11**

*.0

12

.043**

*-.

056**

*.2

73**

*

*p <

.05

**

p <

.01 **

*p <

.001 (

two-t

aile

d t

ests

)

SP5601_11.indd 215 1/8/09 1:58:33 PM

Page 12

216 Teasdale/silver

Together, these findings support our hypothesis that neighborhood disadvantage affects self-control (Hypothesis 1), as well as our hypothesis that family socialization activities and social regulation mediate some of this relationship (Hypothesis 2).7 It is important to highlight that a significant relationship remained between neighborhood disadvantage and self-control, after controlling for a variety of theoretically relevant predictors of self-control (a point we discuss below).

Discussion

This study examined the relationship between neighborhood context and adolescent self-control using data from Add Health. We found neighborhood disadvantage to be a significant predictor of self-control in adolescents, controlling for demographics, family characteristics, and individual-level social regulation measures. Thus, in contrast to Gottfredson and Hirschi’s (1990) arguments, our findings indicate that self-control is influenced, in part, by factors originating outside of the family. More specifically, our study suggests that the neighborhood environment within which adolescents live affects the level of self-control they exhibit. Our results also indicate that family factors and neighborhood social integration mediate part of the relationship between neighborhood disadvantage and self-control. Both of these find-ings represent departures from Gottfredson and Hirschi’s conceptualization of self-control

7. We are unaware of a statistical test for the change in coefficients for nested models within an HLM framework.

Table 3 • Results from Two Level Structural Equation Model Predicting Self-Controla

Model 1 Model 2

Neighborhood level (n = 711) b se b seSocioeconomic disadvantage -.875*** .114 -.564*** .131Residential mobility — -.123* .053

Person level (n = 9171)

DemographicsFemale — -.151*** .015Blacka — -.104** .023Othera — .013 .027Hispanica — -.091** .024Age — .019** .005Length of residence — .002 .001

Family characteristics Family SES — .019 .012Both biological parents — .081** .017Parent attachment — .287*** .019Parent availability — .033* .012

Social regulationSocial integration — .084** .026Parent network involvement — .036 .082Intergenerational closure .227 .135

Neighborhood level R2 .271 .615Individual level R2 .011 .090

*p < .05 **p < .01 ***p < .001 (two-tailed tests)a Results from the structural model only. Measurement model not reported.b White is the reference category

SP5601_11.indd 216 1/8/09 1:58:33 PM

Page 13

Neighborhoods and self-Control 217

as originating exclusively within the family, and support an expanded view of the socialization processes that contribute to the development of self-control.

Although collective regulation processes were found to influence self-control, a signifi-cant relationship between neighborhood disadvantage and self-control remained after con-trolling for collective regulation. One possible explanation for this result is that our mea-sures of collective regulation did not adequately capture the full range of regulation processes at work in the neighborhood. For example, we were unable to measure collective efficacy (Sampson, Raudenbush, and Earls 1997) using the Add Health data. Collective efficacy com-bines informal social control with cohesion and trust among neighbors, both of which are likely to contribute to the socialization of youth outside the home. Omitting this measure may have undermined our ability to explain the association between neighborhood disadvantage and self-control. Future research on the relationship between neighborhood context and self-control should therefore attempt to measure a more complete range of collective regulation processes than we were able to measure here.

Another explanation for the finding that the relationship between neighborhood disad-vantage and self-control remained after controlling for collective regulation is that the asso-ciation may be due to factors other than collective regulation. One possibility in this regard is culture. As we noted earlier, Shaw and McKay ([1942]1969) emphasized the idea of compet-ing ways of life to explain the relationship between neighborhood structure and delinquency. According to their argument, residents of disorganized neighborhoods vary widely in the ex-tent to which they are exposed to, influenced by, and adhere to conventional values, includ-ing self-discipline, long-term planning, and delayed gratification, each of which is found in relatively low supply among residents of disadvantaged neighborhoods, particularly those en-gulfed in an oppositional street culture (Anderson 1999; Newman 1999). In his ethnographic study of street life in inner city Philadelphia, Elijah Anderson (1999) describes the street cul-ture as a “mediating influence under which children may come to reconsider and rearrange their personal orientations”(p. 67). To the extent that disadvantaged neighborhoods offer normative environments in which short-term thinking, impulsive behavior, and immediate gratification are rewarded (and bring status and esteem on the street), such environments may function as powerful socializing mechanisms hindering the development and exercise of self-control in youth.8

A third explanation for our results is that they are the product of selection bias. In this study, as with all research aimed at relating contextual factors to individual-level outcomes, the problem of selection bias must be considered and, to the best extent possible, ruled out. People choose neighborhoods on the basis of factors (especially SES, but also unmeasured and unmeasurable ones) that may be related to the outcome in question. To what extent the relationship between neighborhood disadvantage and self-control among adolescents is de-termined by factors related to the systematic selection into disadvantaged neighborhoods of youth with low self-control cannot be determined from our data. However, there are several reasons to believe that selection bias is not a significant problem in our study. First, adoles-cents do not choose their neighborhoods, their parents do. Thus, for selection bias to operate, parents would have to choose neighborhoods, in part, based on their children’s level of self-control, an assumption that does not seem plausible. Rather, the key factors known to affect residential attainment among adults are SES and employment (Logan et al. 1996), each of which was controlled in our analyses.

8. It is important to note, however, that although the cultural argument is consistent with the data, it contradicts Gottfredson and Hirschi’s (1990) assumption that low self-control is not learned, but is the product of a breakdown in socialization within the family. Thus, to posit a cultural influence on self-control is to assume that low self-control may be in part learned. Clearly, this issue cannot be resolved within the current analysis. However, it would make an interesting focus for future research in this area.

SP5601_11.indd 217 1/8/09 1:58:33 PM

Page 14

218 Teasdale/silver

A more serious selection concern is that parents who themselves have low self-control self-select into disadvantaged neighborhoods due to their inability to plan for the future and achieve long-term economic goals. Moreover, parents with low self-control may be more like-ly to produce offspring with low self-control, due to their poor parenting skills (Gottfredson and Hirschi 1990). Thus, the association between neighborhood disadvantage and low self-control observed here may be spuriously related to parental levels of self-control. Although we have no direct measure of parents’ self-control, two of the measures included in our study help to mitigate this concern. First is parental SES. According to Gottfredson and Hirschi, par-ents with low self-control are likely to achieve lower levels of socioeconomic status and, it is their lower SES that is likely to lead to their location in a disadvantaged neighborhood. Thus, to the extent that parental SES and parental self-control are related, controlling for paren-tal SES (as we have) should provide some purchase on controlling for parental self-control. Secondly, we control for parental attachment and monitoring. According to Gottfredson and Hirschi, parents with low self-control are less likely to develop meaningful attachments to their children due to their erratic parenting and disciplinary practices. To the extent that this is true, controlling for parental attachment and availability for monitoring (as we have) should provide additional purchase on our attempt to rule out a spurious relationship due to parental self-control.

Finally, it is possible that the results we estimated are attenuated because adolescents with the lowest levels of self-control are more likely to drop out of school. Since we use data from a school-based sample, any estimation of the effects of neighborhoods on self-control will be attenuated. Although we cannot completely rule out the possibility that the relationship between neighborhood disadvantage and low self-control is due to the effects of selection, we believe that by controlling for parent SES, parental attachment, and parental monitoring, we have gone a long way toward reducing this concern. Nonetheless, measurement of parental self-control should be an important priority for future research in this area.

The theoretical contribution of this study lies in bringing together macro- and micro-level variants of control theory. Although criminologists have devoted attention to the topic of theoretical integration (Bernard and Snipes 1996; Messner, Krohn, and Liska 1989), and some very interesting approaches have emerged (Elliott, Ageton, and Cantor 1979; Tittle 1995; Thornberry 1987), there have been few attempts to develop multilevel perspectives that bring together “types of places” with “types of people” explanations (Stark 1987). In the current study, we drew upon two well-known variants of control theory, social disorganization theory and the general theory of crime, to examine the ways in which community context might influence the development of low self-control in youth. We argued that the two theories have significant overlap, especially in the area of socialization: social disorganization theory is concerned with the impact of community structure on socialization, while the general theory is concerned with the consequences of socialization for the development of self-control. Com-bining these variants of control theory enabled us to develop and test a multi-level perspective on the etiology of self-control in youth. Although in the end we were unable to fully explain the observed association between neighborhood disadvantage and low self-control, the fact that such an association exists, net of demographic-, family-, and individual-level social regu-lation variables, underscores the value of studying the development of self-control from a multilevel perspective. We hope our findings stimulate further research and theorizing along these lines.

An important implication of this study is the notion that policy initiatives based on indi-vidual-level characteristics are, in part, a response to naïve theorizing. Selective incapacita-tion and truth in sentencing legislation, for example, have at their root an individual-level understanding of the crime problem. Criminological theories, such as self-control, reinforce these notions. We believe that a more complete understanding of the crime problem based on a multilevel theory of criminal offending might support the development of better-in-formed crime policy. It is our hope that the current study provides an impetus to further examining these issues.

SP5601_11.indd 218 1/8/09 1:58:34 PM

Page 15

Neighborhoods and self-Control 219

Table A1 • Results from Principal Components Analysis

AlphaStandardized

Factor Loading

Self-control .67 Trouble paying attention .827

Trouble getting homework done .767 Trouble with teachers .614 Trouble keeping focused .598

Socioeconomic disadvantage .84

Proportion HH with public assistance income .866 Proportion indiv. with income less than poverty level .830 Unemployment rate .785 Proportion female-headed HH .762

Parent attachment .66

Mother warm and loving .736 Father warm and loving .532 Mother cares about you .650

Social integration .60 Know most people .726 Talked to neighbors on the street .550 Neighbors look out for one another .466

Parent network involvement .39

Member PTO .526 Member of a civic organization .509

Intergenerational closure .64

Know school of best friend .712 Met best friend .870 Met parents of best friend .752

Self-control

% poverty

% Fem. headed

Unemployment

Pub. assist.

Within Level

Between Level

Homework

Teachers

Attention

Focused

Disadvantage

Parent attachment

Social integration

Intergenerational closure

Know

Talk

Look out

Know parents

Know friend

Met friend

Mom warm

Dad warm

Mom cares

Mobility

Demographics

Figure B1 • Two-Level Structural Equation Model

Appendix B

Appendix A

SP5601_11.indd 219 1/8/09 1:58:34 PM

Page 16

220 Teasdale/silver

References

Anderson, Elijah. 1999. The Code of the Street: Decency, Violence, and the Moral Life of the Inner City. New York: W. W. Norton.

Bearman, Peter S., Jo Jones, and J. Richard Udry. 1997. The National Longitudinal Study of Adolescent Health: Research Design. Retrieved November 2004 (http://www.cpc.unc.edu/addhealth).

Bellair, Paul E. 1997. “Social Interaction and Community Crime: Examining the Importance of Neighbor Networks.” Criminology 35:677–703.

Bourdieu, Pierre 1986. “The Forms of Capital.” Pp. 241–58 in Handbook of Theory and Research for the Sociology of Education, edited by J. Richardson. New York: Grenwood Press.

Bernard, Thomas J. and Jeffery B. Snipes. 1996. “Theoretical Integration in Criminology.” Pp. 301–48 in Crime and Justice: A Review of Research (vol. 20), edited by Michael Tonry. Chicago: University of Chicago Press.

Bursik, Robert J., Jr. 1988. “Social Disorganization and Theories of Crime and Delinquency: Problems and Prospects.” Criminology 26:519–52.

Burt, Callie H., Ronald L. Simons, and Leslie G. Simons. 2006. “A Longitudinal Test of the Effects of Parenting and the Stability of Self-Control: Negative evidence for The General Theory of Crime.” Criminology 44(2):353–96.

Burton, Velmer S., Jr., Francis T. Cullen, T. David Evans, Leanne F. Alarid, and R. Gregory Dunaway. 1998. “Gender, Self-Control, and Crime.” Journal of Research in Crime and Delinquency 27:45–54.

Chapple, Constance L. 2005. “Self-Control, Peer Relations, and Delinquency.” Justice Quarterly 22(1):89–106.

Chapple, Constance L., Kimberly A. Tyler, and Bianca E. Bersani. 2005. “Child Neglect and Adolescent Violence: Examining the Effects of Self-Control and Peer Rejection.” Violence and Victims 20(1):39–53.

Cochran, John K., Peter B. Wood, Christine S. Sellers, Wendy Wilkerson, and Mitchell B. Chamlin. 1998. “Academic Dishonesty and Low Self-Control: An Empirical Test of a General Theory of Crime.” Devi-ant Behavior 19:227–55.

Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94:S95–S120.

Doherty, Elaine E. 2006. “Self-Control, Social Bonds, and Desistance: A Test of Life-Course Interdepen-dence.” Criminology 44(4):807–34.

Elliott, Delbert S., Suzanne S. Ageton, and Rachelle J. Cantor. 1979. “An Integrated Theoretical Perspec-tive on Delinquent Behavior.” Journal of Research in Crime and Delinquency 16:3–37.

Elliott, Delbert S., William J. Wilson, David Huizinga, Robert J. Sampson, Amanda Elliot, and Bruce Rankin. 1996. “The Effects of Neighborhood Disadvantage on Adolescent Development.” Journal of Research in Crime and Delinquency 33:389–426.

Evans, T. David, Francis T. Cullen, Velmer S. Burton JR, R. Gregory Dunaway, and Michael L. Benson. 1997. “The Social Consequences of Self-Control: Testing the General Theory of Crime.” Criminology 35:475–501.

Feldman, S. Shirley and Daniel A. Weinberger. 1994. “Self-Restraint as a Mediator of Family Influences on Boys’ Delinquent Behavior: A Longitudinal Study.” Child Development 65(1):195–211.

Gibbs, John J., Dennis Giever, and Jamie S. Martin. 1998. “Parental Management and Self-Control: An Empirical Test of Gottfredson and Hirschi’s General Theory.” Journal of Research in Crime and Delin-quency 35:40–70.

Glueck, Sheldon and Eleanor T. Glueck. 1950. Unraveling Juvenile Delinquency. Cambridge, MA: Harvard University Press.

Gottfredson, Michael R. and Travis Hirschi. 1990. A General Theory of Crime. Stanford, CA: Stanford Uni-versity Press.

Grasmick, Harold G., Charles R. Tittle, Robert J. Bursik Jr., and Bruce G. Arneklev. 1993. “Testing the Core Empirical Implications of Gottfredson and Hirschi’s General Theory of Crime.” Journal of Research in Crime and Delinquency 30:5–29.

Hay, Carter. 2001. “Parenting, Self-Control, and Delinquency: A Test of Self-Control Theory.” Criminology 39:707–36.

Hay, Carter and Walter Forrest. 2006. “The Development of Self-Control: Examining Self-Control Theo-ry’s Stability Thesis.” Criminology 44(4):739–74.

Hope, Trina, Harold G. Grasmick, and Laura Pointon. 2003. “The Family in Gottfredson and Hirschi’s General Theory of Crime.” Sociological Focus 36:291–311.

SP5601_11.indd 220 1/8/09 1:58:34 PM

Page 17

Neighborhoods and self-Control 221

LaGrange, Theresa C. and Robert A. Silverman. 1999. “Low Self-Control and Opportunity: Testing the General Theory of Crime as an Explanation for Gender Differences in Delinquency.” Criminology 37:41–72.

Latimore, T. Lorraine, Charles R. Tittle, and Harold G. Grasmick. 2006. “Childrearing, Self-Control, and Crime: Additional Evidence.” Sociological Inquiry 76(3):343–71.

Logan, John R., Richard Alba, Thomas McNulty, and Bonnie Fisher. 1996. “Making a Place in the Metro–polis: Locational Attainment in Cities and Suburbs.” Demography 33:443–53.

Longshore, Douglas. 1998. “Self-Control and Criminal Opportunity: A Prospective Test of the General Theory of Crime.” Social Problems 45:102–82.

Longshore, Douglas, Eunice Change, and Nena Messina. 2005. “Self-Control and Social Bonds: A Com-bined Control Perspective on Juvenile Offending.” Journal of Quantitative Criminology 21(4):419–37.

Longshore, Douglas and Susan Turner. 1998. “Self-Control and Criminal Opportunity: Cross-Sectional Test of the General Theory of Crime.” Criminal Justice and Behavior 25(1):81–98.

Longshore, Douglas, Susan Turner, and Judith A. Stein. 1996. “Self-Control in a Criminal Sample: An Examination of Construct Validity.” Criminology 34:209–28.

Lynam, Donald R., Avshalom Caspi, Terrie E. Moffitt, Per-Olof H. Wikstrom, Rolf Loeber, and Scott Novak. 2000. “The Interaction between Impulsivity and Neighborhood Context on Offending: The Effects of Impulsivity are Stronger in Poorer Neighborhoods.” Journal of Abnormal Psychology 100:563–74.

Maas, Cora J. M. and Joop J. Hox. 2005. “Sufficient Sample Sizes for Multilevel Models.” Methodology 1(3):86–92.

McCord, William and Joan McCord. 1959. Psychopathy and Delinquency. New York: Grune and Stratton.Messner, Steven F., Marvin D. Krohn, and Allen E. Liska, eds. 1989. Theoretical Integration in the Study of

Deviance and Crime. Albany: State University of New York Press.Miller, Joshua D. and Donald Lynam. 2001. “Structural Models of Personality and their Relation to Anti-

social Behavior: A Meta-Analytic Review.” Criminology 39:765–98. Morenoff, Jeffrey D., Robert J. Sampson, and Stephen W. Raudenbush. 2001. “Neighborhood Inequality,

Collective Efficacy, and the Spatial Dynamics of Urban Violence.” Criminology 39:517–60.Muthén, Linda K. and Bengt O. Muthén. 2006. Mplus User’s Guide: Statistical Analysis with Latent Variables.

4th ed. Los Angeles, CA: Muthén and Muthén.Nakhaie, M. Reza, Robert A. Silverman, and Theresa C. La Grange. 2000. “Self Control and Social Control:

An Examination of Gender, Ethnicity, Class, and Delinquency.” Canadian Journal of Sociology 25:35–59.Newman, Katherine S. 1999. No Shame in My Game: The Working Poor in the Inner City. New York: Alfred

Knopf and The Russell Sage Foundation.Perrone, Dina, Christopher J. Sullivan, Travis C. Pratt, and Satenik Margaryan. 2004. “Parental Ef-

ficacy, Self-Control, and Delinquency: A Test of a General Theory of Crime on a Nationally Rep-resentative Sample of Youth.” International Journal of Offender Therapy and Comparative Criminology 48:298–312.

Polakowski, Michael. 1994. “Linking Self- and Social Control with Deviance: Illuminating the Structure Underlying a General Theory of Crime and its Relation to Deviant Activity.” Journal of Quantitative Criminology 38:897–930.

Pratt, Travis C. and Franics T. Cullen. 2000. “The Empirical Status of Gottfredson and Hirschi’s General Theory of Crime: A Meta-Analysis.” Criminology 38:931–64.

Pratt, Travis C., Michael G. Turner, and Alex R. Piquero. 2004. “Parental Socialization and Community Context: A Longitudinal Analysis of the Structural Sources of Low Self-Control.” Journal of Research in Crime and Delinquency 41:219–43.

Sampson, Robert J. 1992. “Family Management and Child Development: Insights from Social Disorgani-zation Theory.” Pp. 63–93 in Facts, Frameworks, and Forecasts, Volume 3, edited by Joan McCord. New Brunswick, CT: Transaction Publishers.

———. 1997. “The Embeddedness of Child and Adolescent Development: A Community-Level Perspec-tive on Urban Violence.” Pp. 31–77 in Childhood and Violence in the Inner City, edited by Joan McCord. New York: Cambridge University Press.

Sampson, Robert J. and Byron W. Groves. 1989. “Community Structure and Crime: Testing Social-Disor-ganization Theory.” American Journal of Sociology 94:774–802.

Sampson, Robert J., Jeffrey D. Morenoff, and Felton Earls. 1999. “Beyond Social Capital: Spatial Dynam-ics of Collective Efficacy for Children.” American Sociological Review 64:633–60.

Sampson, Robert J. and John H. Laub. 1993. Crime in the Making: Pathways and Turning Points Through Life. Cambridge, MA: Harvard University Press.

SP5601_11.indd 221 1/8/09 1:58:34 PM

Page 18

222 Teasdale/silver

Sampson, Robert J., Stephen W. Raudenbush, and Felton Earls. 1997. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy.” Science 277:918–24.

Shaw, Clifford R. and Henry D. McKay. [1942]1969. Juvenile Delinquency in Urban Areas: A Study of Rates of Delinquents in Relation to Differential Characteristics of Local Communities in American Cities. Chicago: University of Chicago Press.

Silver, Eric and Lisa Miller. 2004. “Sources of Informal Social Control in Chicago Neighborhoods.” Crimi-nology 42:551–83.

Stark, Rodney. 1987. “Deviant Places: A Theory of the Ecology of Crime.” Criminology 25:893–909.Thornberry, Terrence 1987. “Toward an Interactional Theory of Delinquency.” Criminology 25:863–87.Tittle, Charles R. 1995. Control Balance: Toward a General Theory of Deviance. Boulder, CO: Westview Press.Tittle, Charles R. and Harold R. Grasmick. 1997. “Criminal Behavior and Age: A Test of Three Provocative

Hypotheses.” The Journal of Criminal Law and Criminology 88:309–42.Vazsonyi, Alexander T., Janice E. C. Wittekind, Laura A. Belliston, and Timothy D. Van Loh. 2004. “Ex-

tending the General Theory of Crime to ‘The East:’ Low Self-Control in Japanese late adolescence.” Journal of Quantitative Criminology 20(3):189–216.

Vazsonyi, Alexander T. and Jennifer M. Crosswhite. 2004. “A Test of Gottfredson and Hirschi’s Gen-eral Theory of Crime in African American Adolescents.” Journal of Research in Crime and Delinquency 41(4):407–32.

Vazsonyi, Alexander T., Lloyd E. Pickering, Marianne Junger, and Dick Hessing. 2001. “An Empirical Test of a General Theory of Crime: A Four-Nation Comparative Study of Self-Control and the Prediction of Delinquency.” Journal of Research in Crime and Delinquency 38(2):91–131.

Wright, Bradley R. E., Avshalom Caspi, Terrie E. Moffitt, and Phil A. Silva. 1999. “Low Self-Control, Social Bonds, and Crime: Social Causation, Social Selection, or Both?” Criminology 37:479–514.

SP5601_11.indd 222 1/8/09 1:58:34 PM