The Effects of Self-Control,Gang Membership,and Parental Attachment/Identification on Police Contacts Among Latino and African American Youths

Abstract

This study assessed the correlates of self-control and police contact in a sample of Chicago public high school students. The investigation examined the effects of parental attachment/identification, family structure, and peer association on self-control and the effects of parental attachment/identification, family structure, peer association, and self-control on police contact. Differences between African American and Latino youth on the predictors of the two dependent measures were tested in separate regression models. Weak parental attachment/identification and gang affiliation (peer association) predicted low self-control among all students. Among African American youth, only weak maternal attachment/identification predicted low self-control; both weak maternal attachment/identification and gang affiliation predicted low self-control among Latino youth. Gang affiliation predicted police stops (delinquency) among African Americans but not among Latinos. However, both African American and Latino students with lower self-control were more likely to be stopped by the police than those with higher self-control.

Keywords

self-control gangs police contacts minority youth

Gottfredson and Hirschi’s (1990) General Theory of Crime (GTC) posits that people commit crimes because of weakly developed or low self-control, which, the researchers claim, explains all delinquent, criminal, and analogous behaviors and is the single most important cause of crime at the individual level. People with greater self-control tend to consider the long-term consequences of their behavior and are careful and deliberate in their decision making. Such individuals self-monitor effectively and conform in socially desirable ways. In contrast, people with low self-control tend to be present-oriented, impulsive, nonempathic, and reckless; they prefer physical over mental activities and engage in antisocial and other related behaviors in order to meet their selfish needs. Self-control is learned in childhood and, once learned, is highly resistant to change (Gottfredson & Hirschi, 1990).

Low self-control is related to poor parent–child bonding and ineffective parenting strategies, including inadequate monitoring of the child and the failure to recognize and correct the child’s misbehavior (Gottfredson & Hirschi, 1990, 2003). Gottfredson and Hirschi (2003) augmented their GTC by asserting that the child’s attachment to the parent is an essential component of effective disciplinary practices (Cullen, Unnever, Wright, & Beaver, 2008). Children with weak attachments to their parents tend to be unresponsive to punishment and corrective interventions and often fail to become fully socialized. As Cullen et al. (2008) aptly noted, “The broader implication . . . is that for any factor—whether family, community, or societal—to influence social control and thus offending and analogous behaviors, it must have an impact on the effectiveness of the parenting that occurs early in life” (p. 65). These influences directly affect parents’ ability to manage their children. In short, children’s attachment to their parents is a prerequisite to effective parenting and the formation of sufficient self-control. Thus, the relationship between parent and child (i.e., attachment) and the effects of that relationship on the development of self-control and antisocial behavior constitute an aspect of the GTC that demands further examination.

The current study focuses on self-control and parental attachment/identification. With respect to the former, we initially explore self-control as a dependent measure in order to enhance our understanding of its correlates and mediators, and subsequently, as an independent variable in order to isolate its effects as a predictor of delinquency in a multivariate analytic framework. With respect to the latter, we use a measure of parental attachment/identification that is modeled after Hirschi’s (2004) most recent redefinition of self-control as well as a measure of family structure that captures the extent to which youths are being supervised in the home. The present study rests on the “parental management thesis” of GTC, which assumes that effective parenting is related to high levels of self-control and moderates the effects of negative peer influences, thereby diminishing the likelihood of delinquent and analogous behaviors (Cullen et al., 2008).

Minorities, Self-Control, and Parental Attachment/Identification

Perhaps most important to the testing of Gottfredson and Hirschi’s GTC is whether its propositions are generalizable. In the 20 years since its inception, the GTC has been examined repeatedly across different settings, times, and populations in an effort to gauge the power of self-control to explain delinquency broadly (e.g., Lynskey, Winfree, Esbensen, & Clason, 2000; Miller, Jennings, Alvarez-Rivera, & Lanza-Kaduce, 2009; Simons, Simons, Chen, Brody, & Lin, 2007; Tittle & Botchkovar, 2005; Vazsonyi & Crosswhite, 2004). Past studies of minorities and self-control have focused on African Americans in the rural South (Vazsonyi & Crosswhite, 2004), Mexican American women (Tacon & Caldera, 2001), and samples of Puerto Ricans (Miller et al., 2009), Native Americans (Morris, Wood, & Dunaway, 2007), and Russian youth (Tittle & Botchkovar, 2005).

Latinos and Self-Control

In an exploration of the cultural elasticity of the GTC, Miller et al. (2009) studied the relationship between self-control and delinquency among Latino adolescents in Puerto Rico. Their results showed that maternal attachment and self-control each independently predicted deviant behaviors (e.g., cigarette smoking, shoplifting, skipping school; cf. Hope & Chapple, 2005). Miller et al. (2009) called for an exploration of the GTC with Latinos in the United States to assess the effects of parental attachment and self-control among various cultural groups. In a study involving Latino youth, using Miller et al.’s data set, Alvarez-Rivera and Fox (2010) found that weak attachments to parents, religion, and school were related to deviant behavior but low self-control was not. Alvarez-Rivera and Fox noted that their findings had limited generalizability. Finally, Maldonado-Molina, Piquero, Jennings, Bird, and Canino (2009) found that sensation-seeking (an element of self-control) was positively related to delinquency in a study of youth from New York City and Puerto Rico.

African Americans and Self-Control

The generality of GTC also has been explored among African American adolescents (e.g., Simons et al. 2007; Vazsonyi & Crosswhite, 2004; see also Lynskey et al., 2000). For example, Simons et al. (2007) found that parental regulation and monitoring decreased antisocial behavior among African American youth, whereas parental rejection and resentment increased the likelihood that these youths would associate with delinquent peers. The effects of both of these variables were mediated by sociocognitive and emotional factors, such as low self-control, and generally supported the tenets of the GTC.

Mason, Cauce, Gonzales, and Hiraga (1996) found a curvilinear relationship between parental control and delinquency among African American youth. Both highly authoritarian and permissive parenting styles were found to be equally ineffective techniques for preventing misbehavior among youth. Vazsonyi and Crosswhite (2004) found that self-control explained delinquency among African American adolescents living in a poor, rural community, and recommended that their study be replicated with a sample of urban African American youth.

Racial Comparisons and GTC

Few studies of the GTC have compared racial or ethnic minorities; however, there have been some recent exceptions. For example, in a study of self-control that included gender and race, Turner and Piquero (2002) reported that on measures of self-control assessed during childhood and adolescence, non-Whites scored consistently lower than Whites; they also found that low self-control in youth was a predictor of adult criminality (see also Gibson, Ward, Wright, Beaver, & Delisi, 2010, for an examination of the role of gender and self-control). Turner and Piquero (2002) and others suggested that future research should explore differences between African Americans and Latinos on self-control and criminal justice contact, which is the focus of the present investigation.

Lynskey et al. (2000) examined the association between parental effectiveness and self-control in a national survey of nearly 6,000 African Americans, Whites, Latinos, and Asians in 11 cities. In this highly diverse sample, the researchers found that weak parental attachment and poor parental monitoring predicted low self-control among youth. Lynskey et al. also reported a link between gang involvement and family structure (i.e., whether a child was reared by one or both biological parents; see also Brannigan, Gemmell, Pevalin, & Wade, 2002; Phythian, Keane, & Krull, 2008). In addition, their findings indicated that African American males exhibited greater self-control than adolescents in other racial groups yet they were more likely to join a gang than non–African American youth. Thus, Lynskey et al. (2000) concluded that “self-control theory cannot provide an explanation for gang membership across racial or ethnic groups” (p. 15).

Arbona and Power (2003) compared African Americans, Whites, and Latinos on parental attachment, self-esteem, and antisocial behavior. After controlling for socioeconomic variables, the researchers found that higher levels of parental attachment were related to higher levels of self-esteem and to lower levels of antisocial behavior. The relationships among these variables were similar for all three groups of adolescents.

Exploring the relationships among parenting styles, self-control, and delinquency in a study of high school students from diverse ethnic and racial backgrounds, Hay (2001) found that noncorporeal and fair disciplinary practices increased self-control in all racial groups. Applying Baumrind’s (1996) theory of parenting styles, Hay found that children of more demanding, but responsive, parents (authoritative) exhibited greater self-control than children of authoritarian, permissive, or rejecting–neglecting parents. The race of the children was unrelated to self-control. Hay (2001) asserted that the GTC overstates the influence of parental control and underemphasizes the context or manner in which it is exercised. He concluded that low self-control is related to delinquency by mediating the effects of discipline and parental monitoring. Based on calls for research from Miller et al. (2009), Turner and Piquero (2002), Alvarez-Rivera and Fox (2010), and others, the present study was designed to further illuminate the correlates of self-control and police contact among the two most populous minority groups in the United States: African Americans and Latinos.

Self-Control and Police Contact

A handful of studies have examined the relationship between self-control and police contact (DeLisi, 2001). For example, one study showed that people with low self-control are likely to experience more frequent and serious encounters with the police and courts than those with high self-control (DeLisi & Berg, 2006). Based on a sample of 7th- through 12th-graders, Beaver, DeLisi, Mears, and Stewart (2009) reported that boys with low self-control had more, and earlier, contact with the police and were more likely to be arrested and convicted than boys with high self-control. Other researchers have found a relationship between self-control and driving under the influence of alcohol. Furthermore, individuals with low self-control are less likely to wear seat-belts and more likely to drive under the influence of alcohol (Keane, Maxim, & Teevan, 1993). Thus, people with low self-control are more likely to have contact with the police because of their tendency to engage in imprudent or reckless actions (Forde & Kennedy, 1997).

Peer Relations and Delinquency

Regardless of positive parental practices, contact with delinquent peers exerts a powerful influence over youths’ behaviors; such influences should be investigated in explorations of the GTC (Miller et al., 2009; Perrone, Sullivan, Pratt, & Margaryan, 2004). Although Gottfredson and Hirschi (1990) argued that the relationship between delinquent peer association and deviance is spurious, numerous studies have found otherwise (Akers, 2008; Miller et al., 2009; Perrone et al., 2004) and point to the relevance of differential association as a factor in the etiology of antisocial behavior. Hirschi and Gottfredson (2000) have acknowledged that their GTC is commonly criticized for ignoring the effects of gang membership on youths’ involvement in delinquent and criminal behavior. Indeed, Hirschi’s (2004) redefinition of self-control includes peer delinquency as an important element in the measurement of the construct.

The Current Study

The current study assesses the correlates of self-control and delinquency in a sample of Chicago public high school students. The investigation examined the effects of parental attachment/identification, family structure, and peer association on self-control, which was the study’s first dependent measure, as well as the effects of parental attachment/identification, family structure, peer association, and self-control on police contact, which was the study’s second dependent measure. Differences between African American and Latino youth on the predictors of the two dependent measures were tested in separate regression models.

As we suggested above, the current research contributes to the literature in three ways. First, it compares African American and Latino youths on self-control, exploring its correlates and its effects. Second, it advances knowledge about self-control in minority youth populations by directly comparing samples of African American and Latino adolescents in Chicago who live in criminogenic environments and have considerable contact with the police (Friedman, Lurigio, Greenleaf, & Albertson, 2004). Third, it incorporates into one analytic framework several variables that have been recently identified as highly germane to the study of the GTC: parental attachment/identification, family structure, and peer associations.

Method

Procedures

Questionnaire data were obtained from a purposive sample of Chicago public high school students enrolled in 18 schools throughout the city. The sampling process was undertaken with the guidance of a Chicago Public School System (CPS) administrator, who served on the study’s advisory board. To select a representative sample of students from Chicago’s highly sizable (third largest in the country) and widely dispersed public high school system, two variables guided the selection of the sample: location and school size. Although students were our unit of analysis, the selection of a random sample of youth from the sampling frame of all of the city’s currently enrolled high school students would have been prohibitively costly as well as highly impractical, infeasible, and wasteful of time and resources. Hence, a set of three high schools was sampled from each of the major geographic zones of the city, which the CPS uses for administrative and record-keeping purposes: North, Near North, West, Central, South, and Far South. In each zone, a small-, medium-, and large-sized school was chosen to match the following CPS categories of school size: <500 students, 500 to 1,500 students, and ≥1,500 students (cps.edu/Schools/Find_a_school/Pages/Findaschool.aspx).

Approval for data collection was obtained from the Chicago Board of Education’s Legal Department, which is responsible for protecting student confidentiality. In accordance with CPS and IRB confidentiality requirements, the surveys were anonymous. Data collection procedures followed the directives of each of the participating high school’s principals.

Questionnaires were administered by the project manager during regular school hours, in class periods that were set aside by each high school for standardized test administration. The selection of this class period ensured that an academically heterogeneous group of students (e.g., honors and nonhonors students) would be included in the study. The teacher left the room while the survey was being conducted. Only students giving their informed consent participated in the survey, which took approximately 25 minutes to complete. The data were aggregated across the 18 schools. At the request of CPS administrators, no data from any of the individual schools were analyzed.

Sample Characteristics

The completion rate for the survey was 94% (n = 891). Of the 943 students asked to complete the survey, only 5 students refused to participate; 47 surveys were unusable. Half of the sample consisted of freshman, and the other half consisted of juniors. The mean age of the students was 16 years. Approximately 55% of the respondents were African American, 27% were Latino, 7% were White, and 3% were Asian (Friedman et al., 2004). The sample was 46% male and 54% female. These demographic characteristics closely reflected those of the population of students attending Chicago Public High Schools. Specifically, at the time of the study, 9% of Chicago’s public high school students were White, 86% were African American or Latino, and 53% were female (cps.edu/About_CPS/Ataglance/Pages/Stats_and_facts.aspx). The study’s subsample comparisons included only African American and Latino students; the former constituted two thirds (n = 525) and the latter one third (n = 263) of the combined racial/ethnic subsamples.

Survey Content

The questionnaire consisted of 131 items in open- and closed-ended response formats. The instrument contained a variety of rating scales that measured variables in several content domains, including delinquent attitudes, gang involvement, and perceptions of law enforcement and experiences with the police, as well as attitudes toward school and other social institutions. Table 1 presents descriptive statistics for the study’s measures.

Table 1.

Variables Used in the Equations

Name	Description	Min	Max	M	SD	N
Dependent variable
Low self-control	Composite measure for Gottfredson & Hirschi’s (1990) concept, using four indicators for three elements	4	20	9.88	2.90	826
Stopped by police^a	Self reported police stops; stopped = 1, not stopped = 0	0	1	0.58	0.50	880
Independent, control, and selection variables
LikeMom	Respondent indicated that they want to be like their mother when they grow up	1	5	3.41	1.27	847
LikeDad	Respondent indicated that they want to be like their father when they grow up	1	5	2.58	1.37	846
Age	Age of respondent	13	20	15.79	1.20	885
Male	Sex of respondent; male = 1, female = 0	0	1	0.46	0.50	891
African American	African American youth; African American = 1, all else = 0	0	1	0.55	0.50	896
Latino	Latino youth; Latino = 1, all else = 0	0	1	0.27	0.44	896
LivesMom^b	Dummy variable. Respondent lives with mother or mother and stepparent	0	1	0.44	0.50	880
LivesDad^b	Dummy variable. Respondent lives with father or father and stepparent	0	1	0.05	0.22	880
LivesOther^b	Dummy variable. Respondent lives in other guardian arrangement	0	1	0.12	0.32	880
GangMember	Measure representing differential association; gang member = 1, non–gang member = 0	0	1	0.09	0.28	869

Note: Figures are rounded.

Official reasons given for the stop included truancy (being late, skipping school), curfew violations, disturbing the peace (loud music, making too much noise), driving violations among those of driving age (dangerous driving ), mob action, and suspicion, among other reasons. Answers from the kids about why they believed they were stopped included harassment, illegal possession, appearance and suspicion, and mob action.

Respondent lives with both parents is excluded as the comparison group.

Dependent Measures

Self-control

The study focused on two dependent measures: self-control and police contact. Self-control is the essence of Gottfredson and Hirschi’s (1990) GTC. In its original formulation, the theory suggested that self-control could be measured on several dimensions, including impulsivity, shortsightedness, and preference for physical over mental activities, risk-seeking, self-centeredness, and temper dysregulation. Grounded in the GTC, Grasmick, Tittle, Bursik, and Arneklev (1993) developed a cognitive framework for further conceptualizing and operationalizing self-control. This cognitive conceptualization of the construct, also adopted by Miller et al. (2009), is one of the most common attitudinal measures of self-control used to test the GTC (Cochran, Aleksa, & Chamlin, 2006). Grasmick et al.’s (1993) framework provided the basis for the self-control measure included in the current investigation as both a dependent and independent variable.

Several researchers have suggested that the six original dimensions of Gottfredson and Hirschi’s (1990) measure of self-control might actually comprise more than one construct (Longshore, Turner, & Stein, 1996; Piquero & Bouffard, 2007). The present study’s self-control measure aligns with three of Gottfredson and Hirschi’s dimensions of self-control, which are more cognitive- than behavioral-based in nature (i.e., impulsivity, self-centeredness, and short-sightedness). In short, our measure of self-control was designed to be nontautological—avoiding the measurement of self-control by including its consequences as defining elements (Akers, 1991)—and to capture the cognitive dimensions of the construct (Miller et al., 2009).

The current measure of self-control was derived from four questions in the youth survey, which were answered on a 5-point Likert-type scale, ranging from strongly agree to strongly disagree. To measure impulsivity, respondents were asked to indicate the degree to which they agreed or disagreed with the statement, “I can’t seem to stay out of trouble no matter how I try.” Self-centeredness was measured by two questions. The first asked respondents to indicate the degree to which they agreed or disagreed with the statement “People who leave their keys in their cars are as much to blame if it’s stolen as the people who steal it.” The second statement asked respondents to indicate the degree to which they agreed or disagreed with the statement “Most acts people call ‘delinquent’ don’t really hurt anyone.” Shortsightedness was measured by a question asking respondents to indicate the degree to which they agreed or disagreed with the statement “It’s OK to cut school.”

The three dimensions (four questions) of self-control were highly intercorrelated (ps < .000). Principal components factor analyses were conducted to determine whether these variables could be combined into a single self-control measure. The questions produced a single factor with all loadings higher than .550; the loadings ranged from .581 to .698. Hence, the questions were merged to create an additive index of self-control. The composite measure scores ranged from 4 to 20; the mean and median score on the index was approximately 10. High values on the measure indicated low self-control.

Stopped by the police

According to DeLisi and Berg (2006), people with low self-control are inclined to engage in public incivilities, such as drinking, playing loud music, or driving recklessly—behaviors that invariably increase the likelihood of police contact. Other studies have found that police contact predicts subsequent reoffending among juveniles and is a valid proxy for delinquency (Lim & Hellenga, 2007). Thus, “stopped by the police” was used in this study as a direct measure of youth contact with the criminal justice system and an indirect measure of delinquency (for comment, see Shedd & Hagan, 2006). The exact question asked, “Have you ever been stopped by the police?” The variable was coded as a dichotomous variable (1 = stopped, 0 = not stopped).

More than half of the youth in the survey indicated that the police had stopped them (58%). Most reported being stopped for a status or criminal offense. The official reasons that students reported for being stopped, which they indicated had been explained to them by the police, included truancy (skipping school, being late), curfew violations, disturbing the peace, driving violations, mob action (hanging out on street corners), and suspicion of criminal activity. During some of the stops, youth were found to be in possession of a firearm or other dangerous weapon. A total of 93 youths provided no reason for the stop because they reported that either the police gave them no reason or the youths did not ask for or remember the reason. Of note, a follow-up question asked youth how many times in the past 12 months had they been stopped. Of the youth in the sample that had been stopped by the police, approximately 31% indicated that they had only been stopped once in the past 12 months. A majority of those that had been stopped in the preceding year, however, reported multiple encounters with the police. For example, about 42% of the youth reported being stopped between 2 and 5 times, and about 14% reported between 5 and 7 encounters. The remainder of these youth reported beyond 7 stops.

Independent Measures

Demographic characteristics

Age, gender, and race were included in the analyses. Age was measured as a continuous variable. Gender has been one of the most robust predictors of both delinquent and criminal activity throughout the history of criminological research (Jennings et al., 2010; Steffensmeier & Streifel, 1991). Gottfredson and Hirschi (1990) have noted that boys and girls differ significantly on self-control in terms of levels but not in terms of causality. Thus, youth with lower levels of self-control are more prone to delinquency regardless of gender (e.g., Paternoster & Triplett, 1988). For example, Burton, Cullen, Evans, Fiftal-Alarid, and Dunaway (1998) reported that the relationship between gender and crime was nonsignificant when self-control was taken into account statistically. Burton et al. (1998) also found that the specific nature of the relationship between self-control and crime differed for men and women (see also Chapple & Johnson, 2009). Therefore, gender was included in the current statistical models and measured as a dichotomous variable (1= male, 0 = female).

In the overall sample, a set of dummy variables was created to measure race as the variable is categorical in nature. The groupings for African American youth and Latino youth were entered into the equation separately for assessment of racial or ethnic effects. All other self-identified racial and ethnic groups were used as a single category, “Other.” The Other category was left out of the equation to create a comparison group and to preclude multicollinearity.

Parental attachment/identification

In addition to self-control (see above), parental attachment/identification was also measured as an independent variable. Hirschi (1969) argued that parental attachment is a fundamental element in delinquency prevention and later identified attachment as a critical aspect of effective parenting (Gottfredson & Hirschi, 2003). In his discussion of the relationship between parental attachment and delinquency, Hirschi (1969) stated that “thus affectional identification, love, or respect is taken as the crucial element of the bond to the parent. Even if the child in effect considers the opinion of his parents, he may conclude that parental reaction is not sufficiently important to deter him from the act. . . . Perhaps the best single item in the present [Hirschi’s] data is: ‘Would you like to be the kind of person your mother (father) is?’” (p. 92). Thirty years later, Miller et al. (2009) concluded that “individuals with strong affection and respect for significant others are less likely to be delinquent because they do not wish to incur harm or disapproval from those important to them” (p. 477). Hirschi’s (2004) most recent formulation of the GTC includes a reconstituted measure of self-control that asks the following question: “Would you like to be the kind of person your mother is?”

Based on the theories and research findings of both Hirschi (1969, 2004) and Miller et al. (2009), two questions were used to measure parental attachment/identification. Students were asked to indicate the degree to which they agreed or disagreed with the statements “I want to be the kind of person my mother is” and “I want to be the kind of person my father is.” Students who reported higher levels of agreement on either of these statements were considered to be more attached to their parents. A significantly higher percentage of respondents in the sample indicated that they wanted to be like their mothers than like their fathers.

Family structure

Children from a two-parent household are more likely to be closely monitored than those from a single-parent household and therefore less likely to have low self-control and to engage in delinquent acts (Demuth & Brown, 2004). Thus, we used family structure as a rough measure of parental supervision (Miller et al., 2009). A set of dummy variables was created for this measure; each category (lives with mom, lives with dad, and lives with other) was compared with students living in a biological, two-parent family.

Gang membership

Delinquent peer association also was included in the analysis (Miller et al., 2009) and measured by asking respondents whether they were gang members; an affirmative response betokened a direct association with delinquent peers. Although other measures of delinquent peer association are more common (Meldrum, Young, & Weerman, 2009), research supports the notion that gang membership is a powerful vehicle for delinquent peers to affiliate with one another and to become more involved in the commission of delinquent, antisocial, and criminal acts. For example, Battin, Hill, Abbott, Catalano, and Hawkins (1998) suggested that gang membership increases delinquency through each individual member’s frequent interactions with delinquent peers who encourage a variety of antisocial behaviors. Therefore, in recognition of the influence of differential peer associations on antisocial behavior, the present research controls for the effects of delinquent peers in an analysis of the relationship between self-control and delinquency.

Analytic Approach

A two-stage analytic process was used, resulting in the estimation of six models. The first three models (i.e., all respondents, African American, and Latino student models) examined the predictors of self-control and were estimated using ordinary least squares regression analyses. The second three models, which also included all respondents as well as separate models for African American and Latino youth, used logistic regression analyses. Similar to Gibbs, Giever, and Martin (1998), the second set of models explored the predictors of police contact. The analyses tested the effects of parental attachment/identification and family structure as well as gang membership on self-control (dependent measure 1) and police contacts (dependent measure 2) (Miller et al., 2009).

Cross-coefficient analysis was used in both stages of analyses, yielding meaningful comparisons between the primary racial groups in the study. In summary, the models for each dependent measure were estimated for all respondents and for African American and Latino students separately in order to investigate differences between the groups. An estimation of Variation Inflation Factors (VIF) and other tests were performed to assess the correlations among the independent variables. Results demonstrated that none of the models were affected by multicollinearity. In other words, none of the independent variables used in this study were highly intercorrelated and, thus, the data met the independence requirement for conducting multivariate analyses. Because of the exploratory nature of the investigation the significance level was set at p ≤.10. However, as shown in the results section below, only two of our major findings on parental attachment/identification failed to reach the standard level of significance (p ≤.05).

Results

Predicting Self-Control

All respondents (Model 1)

The results of the analysis indicated that male students scored lower than female students on the self-control measure (Table 2). Race was also a significant predictor of self-control. Latino students scored lower on self-control than the other groups of students. Students who did not want to be like their mothers or fathers also scored lower on self-control. Similarly, students who reported that they lived with their mothers (or mothers and stepparents) scored lower on self-control, compared with those who lived with both biological parents—a finding that is consistent with previous research on the relationship between single-parent households and delinquency (e.g., Dornbusch et al., 1985; also see Demuth & Brown, 2004, for discussion of the relationship between family structure and delinquency).

Table 2.

Regression Equations Explaining Self-Control

	Low self- control
	All respondents (Model 1)		AA respondents (Model 2)		Latino respondents (Model 3)
	b	p	b	p	b	p
Age	−0.121	.154	−0.134	.235	−0.037	.827
Male^a b	0.850	.000	1.100	.000	0.230	.569
AA^c	0.051	.856	–	–	–	–
Latino^c	0.659	.030	–	–	–	–
LikeMom	−0.233	.006	−0.235	.027	−0.318	.072
LikeDad	−0.174	.033	−0.138	.201	−0.164	.317
LivesMom^d	−0.597	.015	−0.571	.100	−0.333	.465
LivesDad^d	0.289	.545	0.069	.912	−1.527	.250
LivesOther^d	−0.527	.142	−0.292	.510	−1.665	.106
GangMember^a	1.01	.005	0.582	.214	2.396	.002
Constant	12.62	.000	12.69	.000	12.44	.000
N	788		431		213
F	6.668	.000	4.318	.000	2.490	.000
R ²	.079		.076		.089

Note: Figures are rounded. Unstandardized regression coefficients are reported. AA = African American.

Cross-coefficient analysis revealed statistically significant differences between AA and Latinos on this measure.

Comparison group is Female.

Comparison group is Other race/ethnicity.

Comparison group is Lives With Two (Biological) Parent Household.

African American and Latino youth (Models 2 and 3)

The model for African Americans showed that those with weak maternal attachment/identification were more likely to score low on self-control than those with strong maternal attachment/identification. Latinos were similar to African American students on the maternal attachment/ identification measure; in particular, those not wanting to be like their mothers were more likely to score low on self-control (p ≤ .07). However, when comparing the two groups in separate models, a relationship was found between self-control and family structure among African American but not among Latino students. These relationships were examined in a cross-coefficient analysis using the following equation (see Paternoster, Brame, Mazerolle, & Piquero, 1998):

Z = \frac{b_{1} - b_{2}}{\sqrt{S E {b_{1}}^{2} + S E {b_{2}}^{2}}} .

Controlling for other factors, gang membership was found to be a significant predictor of self-control in the Latino model but not in the African American model. African American and Latino youth varied somewhat on the low self-control indicators; however, for both groups, low maternal attachment/identification was a significant predictor. Furthermore, gender was associated with self-control among African American but not among Latino students (i.e., being male was related to low self-control in the former but not in the latter). The cross-coefficient analysis confirmed these differences.

Predicting Police Stops

All respondents (Model 4)

Our presentation of the logistic regression results focuses on the percentage change in the odds of the dependent variable, which is based on the following formula: percentage change in the odds ratio = (e^β -1)*100. The exponential values of beta are presented in Table 3 [see exp(B)]. The model for all students showed that male students were 3 times more likely than female students to be stopped by the police and thus defined as delinquent (Table 3). This result is consistent with the findings of a substantial body of previous research (e.g., Hayslett-McCall & Bernard, 2002). The model also indicated that gang-affiliated youth were 6 times more likely to be stopped by the police than non–gang-affiliated youth. The model showed no direct effect of parental attachment/identification on delinquency. However, a significant relationship was found between self-control and police stops, the proxy for delinquency; that is, lower levels of self-control predicted being stopped by the police. Students with lower levels of self-control were 17% more likely to be stopped by the police than those with higher levels of self-control.

Table 3.

Logistic Regression Equations Explaining Stopped by the Police

	Stopped by police
	All respondents (Model 4)		AA respondents (Model 5)		Latino respondents (Model 6)
	Exp(B)	p	Exp(B)	p	Exp(B)	p
Age	1.098	.169	1.191	.060	1.044	.750
Male^a	3.081	.000	2.859	.000	3.691	.000
AA^b	0.948	.810	–	–	–	–
Latino^b	0.842	.479	–	–	–	–
LikeMom	0.922	.232	0.919	.338	1.011	.941
LikeDad	0.939	.337	0.931	.421	0.943	.659
LivesMom^c	1.347	.132	0.864	.609	1.605	.193
LivesDad^c	1.553	.260	0.947	.915	1.316	.999
LivesOther^c	1.406	.231	1.029	.937	1.295	.742
GangMember^d	6.385	.000	9.072	.001	1.935	.369
Self-Control	1.170	.000	1.162	.000	1.212	.001
Constant	0.049	.011	0.020	.018	0.041	.162
N	773		421		242
χ²	138.337	.000	74.849	.000	42.549	.000

Note: Figures are rounded. AA = African American.

Comparison group is Female.

Comparison group is Other race/ethnicity.

Comparison group is Lives With Two (Biological) Parent Household.

Cross-coefficient analysis revealed statistically significant differences between AA and Latino respondents on this measure.

Consistent with the findings of Gibbs et al. (1998), these results suggest that maternal attachment/identification indirectly affects police contacts through low-self control. An indirect relationship between low self-control and police stops was also mediated by family structure among Latino youth. Maternal attachment/identification, family structure, and race (Latino) were all related to low self-control, which, in turn, was related to police stops. Therefore, the relationships between family structure and parental attachment/identification and police stops appear to be mediated by self-control.

African American and Latino youth (Models 5 and 6)

Both Latino and African American teenage boys were significantly more likely to be stopped by the police than their female counterparts. Gang membership was a significant predictor for African Americans; however, it was not for Latinos. African American students who were in a gang were 9 times more likely to be stopped by the police than those who were not in a gang. African Americans with lower levels of self-control were 16% more likely to be stopped by the police than African Americans with higher levels of self-control. Moreover, Latinos with lower levels of self-control were 21% more likely to be stopped by the police than Latinos with higher levels of self-control. Thus, self-control was indirectly related to police stops in Models 5 and 6. Maternal attachment/identification was not directly associated with police stops. Thus, the relationship between maternal attachment/identification and police stops (delinquency) appears to be mediated by self-control (Table 3).

Discussion

Correlates of Self-Control

Weak parental attachment/identification, living in a single-parent household, and gang affiliation (peer influence) predicted low self-control among all students (Model 1). After low self-control becomes entrenched, it can result in a greater and more sustained likelihood of delinquency and criminality. People with low self-control crave instant gratification and eschew hard work. Individuals with low self-control are more likely than those with high self-control to engage in criminal or delinquent behavior; nevertheless, such activity is neither inexorable nor persistent (Gottfredson & Hirschi, 1990). The overwhelming vicissitudes of life and the power of continued socialization explain why people with low self-control commit fewer crimes as they age (Curran & Renzetti, 2001), challenging the notion that low self-control affects crime throughout the life span (Gottfredson & Hirschi, 1990).

Correlates of Police Contact

Contact with the police was predicted by gender, gang affiliation, and low self-control for all students (Model 4). These same variables predicted police contacts in the African American model (Model 5); however, in the Latino model (Model 6), only gender and self-control were significant. Therefore, results suggest that parental attachment/identification is an important protective factor against low self-control directly and police contacts indirectly through its relationship with self-control. Similarly, Unnever, Cullen, and Agnew (2006) and Gibbs et al. (1998) found that the relationship between parenting and delinquency was mediated by low self-control. Findings from our model that predicted police contact among all the students support those reported by Beaver et al. (2009), who found that people with low self-control are more likely to have contact with the police than those with high self-control. These results are also consistent with those reported in several other investigations as well (Beaver et al., 2009; Gibbs et al., 1998; Simons et al., 2007; Unnever et al., 2006). In line with numerous studies of Hirschi’s Social Control Theory, the present research was unable to test the directionality of the relationship between self-control and delinquency. Nonetheless, although we used a proxy for delinquency (i.e., police contacts), our results support those reported in several studies that used more traditional or direct measures of delinquency, underscoring the robustness of Gottfredson and Hirschi’s (1990) GTC.

Future research should also explore the effects of unsolicited contact with the police, particularly when youth construe police stops as unreasonable or unprovoked. Studies have shown that these experiences could have damaging effects on adolescents, particularly when such encounters are perceived as disrespectful or demeaning (e.g., Friedman et al., 2004). In fact, societal reaction theorists suggest that deviance amplification rather than deterrence is likely to result from harsh encounters with criminal justice authorities (Smith & Brame, 1994).

Racial Differences Among Youth

This study found that African American and Latino students differed on the predictors of self-control and police contact. Specifically, among African American youth (Model 2), gender and maternal attachment/identification predicted self-control; both maternal attachment/identification and gang affiliation were significant predictors among Latino youth but not gender (Model 3). Thus, weak maternal attachment/identification was associated with low self-control among both African Americans and Latinos. Gang membership was related to low self-control among Latinos but not among African Americans. Among African Americans, gang membership was a significant predictor of police stops but not for Latinos.

Miller et al. (2009) found that low self-control and weak maternal attachment/ identification predicted delinquency in a sample of Puerto Rican youth, whereas we found that low self-control, but not maternal attachment, predicted delinquency (i.e., being stopped by the police) in an overall sample of Latino youth. (Chicago’s population is nearly 30% Latino, and a substantial majority is of Mexican American or Puerto Rican descent; Johnson, 2002). This difference might be explained by cultural variations in child rearing practices, as well as variations in youths’ relationships with their parents in the United States, compared with parental practices and parent–child relationships in other countries with sizable Latino populations. Perhaps the present study illustrates that among Latino adolescents enrolled in Chicago’s public school system (mostly second- and third-generation Latinos; see Ready & Brown-Gort, 2006), the effects of parental attachment/identification are outweighed by the effects of other factors on delinquency.

Our results regarding African American youth living in an urban setting are consistent with Vazsonyi and Crosswhite’s (2004) findings regarding African American youth living in a rural setting. In Vazsonyi and Crosswhite’s research, low self-control was associated with delinquency. In the present study, low self-control was a strong predictor of police stops among African American youth.

Parental Attachment/Identification

The results of the current investigation indicated that paternal attachment/identification was a direct predictor of self-control and a mediating factor in the prediction of police contact. We also found in the overall model and in the African American model that students who lived with both parents scored higher on self-control than students who lived with only one parent or the other—a finding that has implications for understanding the effects of parental control and family structure on delinquency. According to Gottfredson and Hirschi (1990), parental monitoring is essential to the successful development of self-control. Some parents fail to supervise their children or lack the time or desire to do so. Even if parents monitor their children faithfully, they might tacitly condone or inadvertently ignore their disruptive behaviors. The family is the principal vehicle for socialization. When it fails to perform this function, it can be a wellspring for juvenile delinquency and adult criminality. A physician and leading expert on child-rearing practices has recommended that parents and caregivers use more affection and less coercion to discipline their children (Prothrow-Stith, 1991). Hayslett-McCall and Bernard (2002) reported that less affluent parents of male children are more likely than more affluent parents of male children to use threats, yelling, and power-assertive disciplinary tactics, which can diminish the bond between parents and children. More research is needed to elucidate the direct and indirect effects of parental attachment and family structure on delinquency. Although we had no measures of parenting, our results are suggestive of the influence of parental attachment/identification on self-control directly and police contact indirectly.

Limitations of Current Study

The present research has several limitations. For example, the use of police stops as a proxy for delinquency is problematic because it also measures police behavior. Nonetheless, police contact is highly correlated with delinquent activity, and most police officers stop youth with some degree of suspicion and justification (Friedman et al., 2004). Indeed, in the present study, many juveniles were stopped for a serious crime, such as a weapon offense. With respect to self-control, we assessed only half of the dimensions of Gottfredson and Hirschi’s (1990) construct. In light of the continued debate about the validity and coherence of these dimensions, we opted to measure self-control as a more unitary concept by focusing on its cognitive aspects—a decision that also avoided the tautological debate as noted in a paraphrasing of Akers’s (1991) critique of the GTC: “The only way to determine if people have low self-control is to measure if they engage in behavior that suggests low self-control.” Moreover, our measure of parental attachment/identification was a soft indicator of parental control but is nonetheless an important factor in the development of antisocial and criminal behavior and most certainly a cognate of parental discipline and involvement (Hirschi, 2004).

The current study included only youth who were attending public school. The dropout rate in Chicago is higher than the national average, and the most seriously delinquent youth are the least likely to attend school (Greene, 2002). Although we examined single-parent households, structural factors, such as poverty or the social environment, were not included in our study. We encourage researchers to control for income and other socioeconomic indicators in future studies of self-control and delinquency. Finally, the optimal analytic approach for this study would involve the creation of causal pathways (e.g., path analysis and structural equation modeling). However, our data violated important statistical assumptions that precluded the use of such methods. Nevertheless, we believe that our results support the link between weak parental attachment/identification, low self-control, and delinquent behavior as measured by police contacts.

Footnotes

Acknowledgements

Considerable thanks are owed to the three reviewers for their helpful comments and suggestions.

The author(s) declared no conflicts of interests with respect to the authorship and/or publication of this article.

The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This study was funded by a grant from the United States Department of Justice (Grant 98-IJ-CX-007’7). The opinions expressed herein are solely those of the authors and do not reflect the official opinions of the National Institute of Justice.

References

Akers

R. L.

(1991). Self-control as a general theory of crime. Journal of Quantitative Criminology, 7, 200-211.

Akers

R. L.

(2008). Self-control and social learning theory. In Goode

(Ed.), Out of control: Assessing the general theory of crime (pp. 77-89). Stanford, CA: Stanford University Press.

Alvarez-Rivera

Fox

(2010). Institutional attachments and self-control: Understanding deviance among Hispanic adolescents. Journal of Criminal Justice, 38(4), 666-674.

Arbona

Power

(2003). Parental attachment, self-esteem, and antisocial behaviors among African American, European American, and Mexican American adolescents. Journal of Counseling Psychology, 50, 40-51.

Brannigan

Gemmell

Pevalin

D. J.

Wade

T. J.

(2002). Self-control and social control in childhood misconduct and aggression: The role of family structure, hyperactivity, and hostile parenting. Canadian Journal of Criminology, 44, 119-142.

Battin

S. R.

Hill

K. G.

Abbott

R. D.

Catalano

R. F.

Hawkins

J. D.

(1998). The contribution of gang membership to delinquency beyond delinquent friends. Criminology, 36, 93-115.

Baumrind

(1996). Parenting: The discipline controversy revisited. Family Relations, 45, 405-414.

Beaver

DeLisi

Mears

Stewart

(2009). Low self-control and contact with the criminal justice system in a nationally representative sample of males. Justice Quarterly, 26, 695-715.

Burton

V. S.

Cullen

F. T.

Evans

T. D.

Fiftal-Alarid

Dunaway

R. G.

(1998). Gender, self-control, and crime. Journal of Research in Crime and Delinquency, 35, 125-147.

10.

Chapple

Johnson

(2009, November). Gender and the acquisition of self-control. Paper presented at the annual meeting of the American Society of Criminology, Los Angeles, CA.

11.

Cochran

J. K.

Aleksa

Chamlin

M. B.

(2006). Self-restraint: A study on the capacity and desire for self-control. Western Criminology Review, 7, 27-40.

12.

Curran

Renzetti

(2001). Theories of crime (2nd ed.). Boston, MA: Allyn & Bacon.

13.

Cullen

F. T.

Unnever

J. D.

Wright

J. P.

Beaver

K. M.

(2008). Parenting and self-control. In Goode

(Ed.), Out of control: Assessing the general theory of crime (pp. 61-74). Stanford, CA: Stanford University Press.

14.

DeLisi

(2001). Designed to fail: Self-control and involvement in the criminal justice system. American Journal of Criminal Justice, 26, 131-148.

15.

DeLisi

Berg

(2006). Exploring theoretical linkages between self-control theory and criminal justice processing. Journal of Criminal Justice, 34, 153-163.

16.

Demuth

Brown

(2004). Family structure, family processes, and adolescent delinquency: The significance of parental absence versus parental gender. Journal of Research in Crime and Delinquency, 41, 58-81.

17.

Dornbusch

S. J.

Carlsmith

Bushwell

Ritter

Leiderman

Hastorf

A. H.

Gross

R. T.

(1985). Single parents, extended households, and the control of adolescents. Child Development, 56, 326-341.

18.

Forde

Kennedy

(1997). Risk lifestyles, routine activities, and the general theory of crime. Justice Quarterly, 14, 265-294.

19.

Friedman

Lurigio

A. J.

Greenleaf

R. G.

Albertson

(2004). Encounters between police and youth: The social costs of disrespect. Journal of Crime and Justice, 27, 1-25.

20.

Gibbs

J. J.

Giever

Martin

(1998). Parental management and self-control: An empirical test of Gottfredson and Hirschi’s general theory. Journal of Research in Crime and Delinquency, 35, 40-70.

21.

Gibson

Ward

Wright

Beaver

Delisi

(2010). Where does gender fit in the measurement of self control? Criminal Justice & Behavior, 37, 883-903.

22.

Gottfredson

M. R.

Hirschi

(1990). A general theory of crime. Stanford, CA: Stanford University Press.

23.

Gottfredson

M. R.

Hirschi

(2003). Social control and opportunity. In Britt

C. L.

Gottfredson

M. R.

(Eds.), Control theories of crime and delinquency (pp. 5-19). New Brunswick, NJ: Transaction Books.

24.

Grasmick

H. G.

Tittle

C. R.

Bursik

R. J.

Arneklev

B. J.

(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.

25.

Greene

J. P.

(2002). High school graduation rates in the United States. New York, NY: The Manhattan Institute for Policy Research.

26.

Hay

(2001). Parenting, self-control, and delinquency: A test of self-control theory. Criminology, 39, 707-736.

27.

Hayslett-McCall

K. L.

Bernard

(2002). Attachment, masculinity, and self-control: A theory of male crime rates. Theoretical Criminology, 6, 5-33.

28.

Hirschi

(1969). Causes of delinquency. Berkeley: University of California Press.

29.

Hirschi

(2004). Self-control and crime. In Baumeister

R. F.

Vohs

K. D.

(Eds.), Handbook of self-regulation: Research, theory, and applications (pp. 537-552). New York, NY: Guilford.

30.

Hirschi

Gottfredson

(2000). In defense of self-control. Theoretical Criminology, 4, 55-69.

31.

Hope

Chapple

(2005). Maternal characteristics, parenting, and adolescent sexual behavior: The role of self-control. Deviant Behavior, 26, 25-45.

32.

Jennings

W. G.

Maldonado-Molina

Piquero

Odgers

Bird

Canino

(2010). Sex differences in trajectories of offending among Puerto Rican youth. Crime & Delinquency, 56, 327-357.

33.

Johnson

K. M.

(2002, April). The changing face of Chicago: Demographic trends in the 1990s. Chicago Fed Letter, no. 176.

34.

Keane

Maxim

Teevan

(1993). Drinking and driving, self-control, and gender: Testing a general theory of crime. Journal of Research in Crime & Delinquency, 30, 30-46.

35.

Lim

Hellenga

(2007, November). Examining racial differences in reoffending: Factors that predict recidivism in Anglo and African American youths. Paper presented at the annual meeting of the American Society of Criminology, Atlanta, GA.

36.

Longshore

Turner

Stein

J. A.

(1996). Self-control in criminal sample: An examination of construct validity. Criminology, 34, 209-228.

37.

Lynskey

D. P.

Winfree

L. T.

Esbensen

F. A.

Clason

(2000). Linking gender, minority group status and family matters to self-control theory: A multivariate analysis of key self-control concepts in a youth-gang context. Juvenile and Family Court Journal, 51, 1-19.

38.

Maldonado-Molina

Piquero

Jennings

Bird

Canino

(2009). Trajectories of delinquency among Puerto Rican children and adolescents at two sites. Journal of Research in Crime and Delinquency, 46, 144-181.

39.

Mason

Cauce

Gonzales

Hiraga

(1996). Neither too sweet nor too sour: Problem peers, maternal control, and problem behavior in African American adolescents. Child Development, 67, 212-213.

40.

Meldrum

R. C.

Young

J. T. N.

Weerman

F. M.

(2009). Reconsidering the effect of self-control and delinquent peers. Journal of Research in Crime and Delinquency, 46, 353-376.

41.

Miller

H. V

Jennings

W. G.

Alvarez-Rivera

L. L.

Lanza-Kaduce

(2009). Self-control, attachment, and deviance among Hispanic adolescents. Journal of Criminal Justice, 37, 469-478.

42.

Morris

G. D.

Wood

P. B.

Dunaway

R. G.

(2007). Testing the cultural invariance of parenting and self-control as predictors of American Indian delinquency. Western Criminology Review, 8, 32-47.

43.

Paternoster

Brame

Mazerolle

Piquero

(1998). Using the correct statistical test for the equality of regression coefficients. Criminology, 36, 859-866.

44.

Paternoster

Triplett

(1988). Disaggregating self-reported delinquency and its implications for theory. Criminology, 26, 591-625.

45.

Perrone

Sullivan

C. J.

Pratt

T. C.

Margaryan

(2004). Parental efficacy, self-control, and delinquency: A test of a general theory of crime on a nationally representative sample of youth. International Journal of Offender Therapy and Comparative Criminology, 48, 298-312.

46.

Phythian

Keane

Krull

(2008). Family structure and parental behavior: Identifying the sources of adolescent self-control. Western Criminology Review, 9, 73-87.

47.

Piquero

A. R.

Bouffard

J. A.

(2007). Something old, something new: A preliminary investigation of Hirschi’s redefined self-control. Justice Quarterly, 24, 1-27.

48.

Prothrow-Stith

(1991). Deadly consequences: How violence is destroying our teenage population and a plan to begin solving the problem. New York, NY: HarperCollins.

49.

Ready

Brown-Gort

(2006). The state of Latino Chicago: This is home now. Institute for Latino Studies, University of Notre Dame, Netherlands. Retrieved from http://latinostudies.nd.edu/pubs/pubs/StateofLatino-final.pdf

50.

Shedd

Hagan

(2006). Toward a developmental and comparative conflict theory of race, ethnicity, and perceptions of criminal injustice. In Peterson

R. D.

Krivo

L. J.

Hagan

(Eds.), The many colors of crime (pp. 313-333). New York: New York University Press.

51.

Simons

Chen

Brody

Lin

(2007). Identifying the psychological factors that mediate the association between parenting practices and delinquency. Criminology, 45, 451-518.

52.

Smith

Brame

(1994). On the initiation and contribution of delinquency. Criminology, 32, 607-629.

53.

Steffensmeier

Streifel

(1991). Age, gender, and crime across three historical periods: 1935, 1960, and 1985. Social Forces, 69, 869-882.

54.

Tacon

Caldera

(2001). Attachment and parental correlates in late adolescent Mexican American women. Hispanic Journal of Behavioral Sciences, 23, 71-82.

55.

Tittle

C. R.

Botchkovar

(2005). Self-control, criminal motivation and deterrence: An investigation using Russian respondents. Criminology, 43, 307-353.

56.

Turner

Piquero

(2002). The stability of self-control. Journal of Criminal Justice, 30, 457-471.

57.

Unnever

Cullen

Agnew

(2006). Why is “bad” parenting criminogenic? Implications from rival theories. Youth Violence and Juvenile Justice, 4, 3-33.

58.

Vazsonyi

Crosswhite

(2004). A test of Gottfredson and Hirschi’s General Theory of Crime in African American Adolescents. Journal of Research in Crime and Delinquency, 41, 407-432.