Economic Opportunities and Immigrant Youth Violence

Abstract

This study examines self-reported delinquency among immigrant youth to determine the effects of both legitimate and illegitimate opportunities on youth violence. Data from the study show that parental employment reduces, but youth’s intensive work during the school year and involvement in property crime increase, the risk of youth violence. Although findings from the study indicate the importance of parental employment in preventing immigrant youth’s violence, the study suggests that the relationship between employment and violence can change over the life course, and youth’s intensive work while attending school may not be beneficial for preventing youth violence. The effect of economic opportunities on the risk of youth violence is also implied in the relationship between negative neighborhood characteristics and youth violence and helps explain higher level of violence among Latino immigrant youths.

Keywords

immigration adaptation immigration and crime employment and crime youth violence

Background

The growth of the immigrant population in the recent time has facilitated research interests in the link between immigration and crime. The immigrant youth population in the United States has steadily increased since the 1960s after the United States changed its immigration policy and opened the door for immigration from non-European countries (Passel, 2011). Since 1990, roughly 1 million immigrants have been added to the U.S. population each year, and their children, both foreign born and U.S. born, have been the fastest growing segment of the U.S. child population (King & Harris, 2007). While in the 1960s immigrant youth accounted for only 6.4% of children under 18 years old in the United States, the immigrant youth population reached more than 17 million or 23.2% of the U.S. child population in 2009; a substantial majority of these immigrant youth (84%) were U.S.-born children of immigrants (Passel, 2011).

Understanding immigrant youth’s behaviors is important because risk behaviors and violence during adolescence tend to persist into adulthood and because the accumulation of disadvantages that affect antisocial behavior tends to stabilize over time (Loeber, 1982; Turney & Kao, 2011). A large body of research showed that as a group foreign-born youth were less likely than their native-born peers to engage in risk behaviors (Bui, 2011; Hussey et al., 2007; King & Harris, 2007; Morenoff & Astor, 2006; Pena et al., 2008), but high levels of youth crime among certain immigrant groups in different locations and times were also documented. For example, immigrant youth from Southeast Asia living in California were found to be more likely than Whites and members of more established Asian groups to engage in criminal activity (Kposowa & Tsunokai, 2003). Among Puerto Rican newcomers, those living in New York City had higher rates of homicide, while those living elsewhere had homicide rates comparable to native Whites’ homicide rates (Rosenwaike & Hempstead, 1990). An increase in violent youth gangs and juvenile crime was also observed in certain neighborhoods in the urban areas where recent immigrants and their children concentrated (Bankston, 1998). As past research on immigrant adaptation tended to rely on acculturation, assimilation, and ecological perspectives for understanding crime and violence among immigrants (Alaniz, Cartmill, & Parker, 1998; Bui, 2009, 2011; Desmond & Kubrin, 2009; Kulis, Marsiglia, Sicotte, & Nieri, 2007; Morenoff & Astor, 2006; Unger, Ritt-Olson, Soto, & Baezconde-Garbanati, 2009), the roles of economic opportunity and employment in immigrant youth’s adaptation outcomes have not been well understood.

Research in the general population has suggested different effects of unemployment on youth crime. On one hand, a lack of legitimate opportunities could facilitate economic crime and utilitarian violence such as robbery among unemployed youths (Bankston, 1998; Baron & Hartnagel, 1998; Desroches, 1997; Sullivan, 1989; Williams & Kornblum, 1985). In addition, family financial problems that resulted from parental unemployment could facilitate delinquency because they were associated with children’s psychological and behavioral problems as well as school performance, the two factors that were associated with delinquency (Baker, Hishinuma, Chang, & Nixon, 2009; Harland, Reijneveld, Brugman, Verloove-Vanhorick, & Verhulst, 2002; Madge, 1983; Mazerolle, 1998). On the other hand, literature also suggests that youth’s employment may have negative impacts on youth’s behavior (Lageson & Uggen, 2013). Although employment provides a means to satisfy financial needs and may help youth avoid illegal means to earn money, youth’s employment, especially while attending school, can be detrimental because intensive paid work has the potential to disrupt youth’s academic performance and promote problem behaviors that interfere with longer term educational attainment and development (Steinberg & Cauffman, 1995).

Employment can have important implications for immigrant youth’s behavior. Recent immigrants form a diverse group that includes both professionals with high levels of education and those who came to the United States with limited education and few occupational skills (Portes & Rumbaut, 2006). At the same time, the transition of the U.S. economy to a postindustrial condition since the 1970s has created a sharp decline in employment opportunities for the unskilled (Bankston, 1998; Portes & Zhou, 1993). For many immigrants, a lack of language and occupational skills, cultural isolation, social alienation, and labor market discriminations often serve as barriers to employment (Rumbaut, 1996; Wortley, 2008). Joblessness may facilitate immigrant youth’s involvement in illegal activity to meet their financial needs (Bankston, 1998). Parental unemployment may increase immigration stress and family conflicts that are often experienced by immigrants during the adaptation process, thus increasing the risk of problematic behaviors among immigrant youth by alienating them from families (Hallsten, Szulkin, & Sarnecki, 2013; Zhou & Banskton, 1998). In addition, because immigrants often earn low incomes, their children may need to work long hours to contribute to the family economy. Thus, it is important to explore the relationships between employment and immigrant youth’s behavior, particularly violence, to improve understanding of the effects of economic opportunities on immigrant youth’s adaptation outcomes.

Conceptual Framework

Youth’s Unemployment and Youth Crime

The relationship between employment and crime has been discussed under different perspectives. Based on the assumption of human rationality, economic perspective emphasizes economic incentives of crime compared to those from legitimate work (Becker, 1968; Freeman, 1996). Becker (1968) argued that people who committed crime weighed the benefits and costs of illegal behaviors and expected the utility of crime would exceed other utilities they could get by using their time and other resources. Similarly, Freeman (1996) suggested that market labor incentives influenced the supply of men to crime, and the collapse of the job market for less skilled men during the 1980s and 1990s might contribute to their increased rate of criminal activity. Focusing on social structure that shaped economic opportunity, Merton (1938) proposed that material success depended on economic opportunities, which were not equally available to everyone. When employment and opportunities for developing job skills were unavailable, individuals turned to crime as an innovative way to achieve material goals. Social control theory developed by Hirschi (2009) also considers the importance of work in preventing crime. Because “idle hands are the devil’s workshop,” involvement in conventional activities such as work can keep people too busy to engage in crime (Hirschi, 2009, p. 22). Research by Sampson and Laub (1993) suggested that attachment and commitment to work could reduce criminal behavior by increasing social control through mutual ties with employers and coworkers. Ethnographic studies conducted in areas with depressed economic conditions showed that youth, especially those who already left schools and could not find jobs, resorted to illegal means to meet their financial needs (Sullivan, 1989; Williams & Kornblum, 1985).

Scholarship on immigration and crime suggests the role of economic opportunities in low crime rates among immigrants. Studies using aggregated data showed that disadvantaged metropolitan areas that experienced an increase in immigrant population had lower crime rates than areas that had similar socioeconomic conditions but small immigrant populations (Martinez, 2006; Martinez, Stowell, & Lee, 2010; Ramey, 2013; Sampson, 2008; Stowell, Messner, McGeever, & Raffalovick, 2009; Velez, 2009). Immigrant revitalization whereby large and growing immigrant populations improve local economic structure and strengthen social ties has been given credits for lowering crime rates, particularly in disadvantaged areas (Ramey, 2013). It has been argued that ethnic enclaves could serve as economic niches to provide economic opportunities for low-skill immigrants, and economic diversification of immigrant communities could allow new immigrants to access employment in businesses established by earlier immigrants, thus helping new immigrants avoid discrimination by native-owned businesses and job competitions outside ethnic enclaves (Portes & Zhou, 1993; Reid, Weiss, Aldelman, & Jaret, 2005). Research indicated that ethnic groups in disadvantaged areas that did not participate in the enclave economy tended to experience higher levels of crime (Martinez, Lee, & Nielsen, 2004). Research also found that in some situations, low-skilled immigrants could have a better job opportunity than their low-skilled, native-born counterparts because employers preferred hiring immigrants over African Americans (Reid et al., 2005). These economic opportunities may contribute to low crime rates among immigrants.

Not all immigrants live and work in ethnic enclaves, however, and not all immigrants who live in ethnic enclaves can benefit from ethnic economies. In fact, only ethnic enclaves with economic resources can provide employment opportunities for new immigrants, while many impoverished and disadvantaged minority communities lack economic opportunities even for their long-time residents (Frank, Cerdá, & Rendón, 2007). In response to limited legitimate labor market, criminal immigrant subcultures can provide opportunities for immigrants to engage in property crime (Bankston, 1998). A lack of legitimate means for earning incomes can also facilitate violence. Literature suggests that living in impoverished communities can increase the likelihood of involvement in criminal activities and violence (Martinez, 2006; Martinez et al., 2004). Unemployed youth can engage in utilitarian violence such as robbery as a way to obtain financial or material resources (Baron & Hartnagel, 1998; Desroches, 1997). Research showed that violence was often associated with drug trades among youth who lacked legitimate work (Williams & Kornblum, 1985). In addition, limited economic opportunities can facilitate youth violence because of frustration or a need for retaliation (Reid et al., 2005).

Youth’s Intensive Employment and Youth Crime

Although employment can provide means to satisfy financial needs and help immigrant youth avoid illegal activity, working long hours while attending school can negatively affect youth’s behavior (Lageson & Uggen, 2013). As immigrants tend to earn low incomes, immigrant youth may need to work while attending school to help the family economy. However, engaging in intensive paid work while attending school can be problematic because intensive paid work has the potential to disrupt academic performance and promote problem behaviors that may interfere with longer term educational attainment (Steinberg & Cauffman, 1995). When adolescents engage in intensive work role, spending long hours on the job, they may invest less time and attention in their school roles and responsibilities; consequently, they may be likely to experience decreased educational performance and lower aspirations for further schooling, which in turn can facilitate delinquency (Lageson & Uggen, 2013). In addition, entering the workforce can substantially alter the way youth use their time. Those who work long hours tend to spend more time in unstructured socializing with peers in the absence of authority figures, such as teachers and parents who otherwise can manage youth’s behavior by using informal social control that emphasizes conventional values (Osgood, 1999). Youth who work also have more disposable incomes and may spend more time outside the home in unstructured activities, such as shopping, cruising in a car, going on dates as well as going to bars and parties (Felson, 1986). Thus, youth’s intensive work can erode emotional ties to parents, reduce commitment to school, and scale back educational aspirations (Hirschi, 1983). Each of these possible consequences may serve to weaken the restraints that can prevent delinquent conduct. Studies found that youth’s intensive work, especially more than 20 hours per week in paid employment during the school year, was associated with delinquency and various problem behaviors, including higher rates of delinquency and substance use (Bachman, Staff, O’Malley, Schulenberg, & Freedman-Doan, 2011; Hirschi, 2009; Paternoster, Bushway, Apel, & Brame, 2003; Wright, Cullen, & Williams, 1997).

Parental Unemployment and Youth Crime

Not only a lack of economic opportunity can facilitate criminal tendency among unemployed individuals, but parental unemployment can also affect children’s behavior through its effects on family economy, family relationships, and children’s psychological well-beings. Parental unemployment can deprive children of necessary material support and serve as family disruptor that adversely influences family relationships and parental monitoring, which in turn can facilitate children’s risk behavior (Baker et al., 2009; Larzelere & Patterson, 1990). Unemployment can result in more severe stress for immigrants who often lack coping resources from relatives and friends in their country of origin (Hovey & Magan, 2000). Parents who respond to economic loss with increased irritability and pessimism are less nurturing but more punitive and arbitrary in their interactions with their children; these parenting behaviors can increase children’s risk of emotional problems and deviant behavior (McLoyd, 1989). Exposure to various negative life events, such as parental unemployment, losing a family member and negative relations with adults can be criminogenic for children (Mazerolle, 1998). The stigma of unemployment can affect children who feel embarrassed when parents are out of work (Madge, 1983). Emotional stress resulting from parent’s unemployment can negatively affect children’s school performance, which is often related to delinquency (Madge, 1983). Studies showed that parental employment and high levels of income were associated with lower levels of delinquency, but persistent parental unemployment was associated with higher rates of cumulative delinquency (Farnworth, Thornberry, Krohn, & Lizotte, 1994; Hallsten et al., 2013; Paternoster & Brame, 1997; Ritakallio, Kaltiala-Heino, Kivivuori, & Rimpelä, 2005). In addition, unemployment stress experienced by immigrants can increase family conflicts, which often exist in immigrant families due to different levels of acculturation between parents and children (Zhou & Bankston, 1998). Severe strains in family relationships can facilitate risk behavior among youth by alienating them from families.

Current Study

The present study examines self-reported delinquency among immigrant youth to determine the relationships of legal and illegal opportunities with violence among immigrant youth. Specifically, the study will test four hypotheses based on the conceptual framework presented in the previous sections. First, because crime, including utilitarian violence, can provide economic incentives for unemployed youth, and because unemployed youth may engage in violence out of frustration, it is expected that youth’s unemployment is associated with higher risk of youth violence (Hypothesis 1). Although work provides legal means to achieve financial goals, youth’s long-hour work while attending school may increase the likelihood of risk behaviors, including violence. Thus, it is expected that youth’s intensive work during the school year is associated with higher level of youth violence (Hypothesis 2). Because youth’s property crime is often organized in the context of youth gangs and involves violence, it is expected that youth’s property crime is associated with higher level of youth violence (Hypothesis 3). In addition, due to negative impacts of parental unemployment on family relationships and youth’s emotional well-beings, which in turn can facilitate youth’s problematic behaviors, it is expected that parental unemployment is associated with higher level of youth violence (Hypothesis 4).

Method

Sample and Data

Data for the present study were collected by the National Longitudinal Study of Adolescent Health (Add Health). The Add Health Project used a complex sampling design to select a nationally representative and probability-based sample of students from grades 7 through 12 in the United States. A school-based design was used to sample high schools and the feeder middle-high schools across the country. Seventy nine percent of the schools contacted by the Project agreed to participate (a total of 132 schools). In-school questionnaires were administered to more than 90,000 students in these schools in a single day within one class period (45–60 min). A sample of students was selected from the union of students who had completed In-School questionnaires to participate in the In-Home data collection phase conducted between September 1994 and December 1995 (Wave I).Seventy-nine percent of selected students participated in face-to-face interviews in their private homes, resulting in a sample with 20,745 cases. A substantial majority of the student interviews (97%) was conducted in English, 444 interviews in Spanish, and the rest in other languages (Harris, 2011). The students’ parents (one parent or a parent figure for each student) also participated in In-Home interviews. Contextual data, which were created with information from the Census 1990 and other data sources, included statistics on characteristics of residential locations assigned to each case. The present study used In-Home data (Wave I) combined with contextual data (Wave I). Only cases of immigrant youth were included in the present study. After cases with missing values on weights, violent behavior, and racial/ethnic identification were eliminated, 4,201cases involving immigrant youth remained in the study for data analysis.

Measures

Sixteen variables were included in the present study, including the outcome variable (self-reported violence), three independent variables as well as eight sociodemographic and four contextual variables used as control variables. Most of the variables were created with data collected from interviews with students, except indicated otherwise.

Outcome variable

Self-reported violence is a dichotomous variable asking a student if he or she engaged in any of the following acts within 12 months prior to the interview: using or threatening with a weapon, taking part in a group fight, serious physical fighting, serious injuring someone, getting into a physical fight, pulling a knife or gun on someone, and shooting or stabbing someone.

Independent variables

Based on the research hypotheses, three independent variables include youth’s employment, parental employment, and youth’s self-reported property crime. Literature suggests that youth’s unemployment is associated with crime (Becker 1968; Freeman, 1996), but youth’s intensive work (more than 20 hours per week) during the school year increases the risk of negative behavior (Bachman et al., 2011; Paternoster et al., 2003; Wright et al., 1997). Thus, youth’s employment is a categorical variable measuring the extent of youth’s work with three categories: unemployment (not working at all), moderate employment (working during summer or working no more than 20 hr per week during school year), and intensive employment (working more than 20 hr per week during school year). Parental employment (parent’s interview) is a categorical variable with three categories that capture the extent of parent’s work and reflect parent’s financial contribution to the family: at least one parent working full time, parent working part time only, and parent’s unemployment. Youth’s self-reported property crime is a dichotomous variable asking a student if he or she engaged in any of the following acts within 12 months prior to the interview: burglarizing a building, stealing something worth less than US$50, stealing something worth more than US$50, shoplifting, stealing a car, and selling drugs.

Contextual variables

Unemployment rate, neighborhood poverty, immigrant density, and exposure to violence were included in the present study as control variables to isolate the sole effects of independent variables. Studies indicated the relationships between unemployment rates and crime rates (Allan & Steffensmeier, 1998; Freeman, 1996). Research showed that youth who experienced stressful life events or little parental supervision and lived in areas with high levels of male joblessness were especially more likely than those resided in other areas to be involved in delinquency (Hoffmann, 2002). Research also found that community disadvantages and neighborhood poverty not only had main effects on delinquency rates but also increased the effect of family poverty on delinquency (Hay, Fortson, Hollist, Altheimer, & Schaible, 2007). Literature indicates that exposure to violence is one important aspect of community disadvantage that can increase youth’s violence tendency through learning (Sampson & Wilson, 1995). On the other hand, the protective effect of immigrant community can be found in a low level of delinquency involvement among youth who live in the areas with high levels of immigrant density (Morenoff & Astor, 2006). For the present study, unemployment rate is a continuous variable measuring the rates of unemployment (percentage) in the census blocks where student respondents resided. Because a census block is smaller than a census track in size, having around 500 household units, the characteristics of a census block can have more immediate impacts on its residents. Neighborhood poverty is a continuous variable measuring the percentage of households with incomes below poverty level in the census blocks where student respondents resided. Exposure to violence is a continuous variable measuring how often student respondents observed violence (shooting or stabbing of persons) within 12 months prior to the interview (0 = never; 1 = once; 2 = more than once), and immigrant density is a continuous variable measuring the percentage of foreign born in the population of the census blocks where student respondents resided.

Sociodemographic variables

Age of the students is measured by years, ranging from 12 to 21. Sex is a dichotomous variable with two categories: male and female. Race/ethnicity is classified into five categories: non-Hispanic White (thereafter White), non-Hispanic Black (thereafter Black), Asian, Hispanic, and other race and ethnicity. Statistics of the “other race and ethnicity” category are not reported because they are not of the study interest. The present study includes first- and second-generation youth. Literature indicates that second-generation youth account for the substantial majority of immigrant youth and have higher levels of delinquency involvement than those in the first generation (Bankston, 1998; Passel, 2011). Thus, immigrant status is used to assess differences in violence involvement among immigrant youth from two generations. Consistent with immigration literature (Portes, 1996) and the U.S. Census Bureau (2001), immigrant status is defined by the birthplace of students and those of their parents. The first generation includes foreign-born students with both foreign-born parents and the second-generation, U.S.-born students with at least one foreign-born parent. Residential location is classified into three categories: urban, suburban, and rural. Family structure is a dichotomous variable measuring whether student respondents lived with two biological parents. Parent’s education is a dichotomous variable measuring whether both parents had college degrees (parent’s interview). Family’s income is a continuous variable measuring family’s total annual incomes in thousand dollars (parent’s interview).

Analytic Strategy

Data analysis involved univariate, bivariate, and multivariate analyses. Univariate analysis was used to describe the study sample characteristics. Bivarate analysis was performed to determine the relationships of self-reported violence with independent, sociodemographic, and contextual variables without control. Because it was difficult to observe the effects of continuous variables with a large value range (from 0 to more than 100) distributed across a nationally representative sample, four continuous variables, including family income, unemployment rate, immigrant density and neighborhood poverty, were recoded into categorical variables representing three values based on quartile distributions: low (first quartile), middle (second and third quartiles), and high (fourth quartile). The recoding allowed the comparison of the effects of three levels of incomes and neighborhood characteristics on youth violence. For categorical variables, adjusted F test (thereafter adj. F) was used instead of chi-square test to adjust sampling errors caused by the complex sampling design (Lee & Forthofer, 2006); t-test was used for continuous variables.

For multivariate analysis, three models of hierarchical logistic regression were used to estimate the sole effects of youth’s employment, parental employment, and youth’s property crime on youth’s self-reported violence. The initial or control model (Model 1) included the outcome variable and sociodemographic variables. Next, three independent variables were added to the analysis to assess the effects of both legitimate and illegitimate economic opportunities as well as the extent of youth’s work and parental employment on youth’s self-reported violence while sociodemographic variables were controlled. Finally, contextual variables were added to the analysis to assess the sole effects of economic opportunities by examining whether contextual variables mediated the effects of economic opportunities on youth violence.

Three weights were available in Add Health data, including stratification weight based on regions (West, Midwest, South and Northeast), cluster weight based on schools, and grand sample weight. All three weights were included in data analysis to adjust sampling errors caused by the complex sampling design. A special statistical software package, SPSS version 21 for complex samples, was used for statistical analysis. Due to the complex sampling design involved in data collections, means were estimated, and standard errors (SE) were used instead of standard deviations.

Missing values

For categorical variables, a missing category was created to represent missing-value cases. As the missing-value category is of little interest for the study, statistics for the missing category are not reported. Missing values for continuous variables were handled in two ways, depending on the extent of missing values. Because family income had the greatest percentage of cases with missing values (27%) and because incomes tend to vary with immigration status as well as race and ethnicity, missing values in family incomes were imputed with predicted marginal means from regression for 10 categories of immigration status by race and ethnicity (two immigration generations by five racial and ethnic categories). Predicted marginal income means were estimated by regressing observed family income on family structure, parent’s birth place (foreign born vs. native born), parent’s education, parent’s employment status, and residential locations. A correspondent dichotomous variable (missing vs. nonmissing) was created and included in data analysis to check the bias that might be caused by regression imputations. The results of data analysis (not reported) indicated that the coefficients for the missing versus nonmissing variables were not significant, meaning that mean replacement did not bias the estimates. For other continuous variables (with less than 5% missing-value cases), mean replacement was applied, and correspondent dichotomous variables (missing vs. nonmissing) were included in data analysis to check the bias of mean replacement. Results of data analysis (not reported) indicated that the coefficients for the missing/nonmissing variables were not significant, meaning that mean replacements did not bias the estimates.

Sample Characteristics

The age of student participants ranged from 12 to 21 years with an estimated mean (thereafter mean) of 16.1 years (SE = 0.18). Males accounted for 49.7% of the sample, Whites 14.5%, Blacks 7.4%, Asians 25.2%, and Hispanics 50.2%. The first generation accounted for 34.6% of the sample and the second generation 65.4%. Nearly 43% of the students resided in urban areas, 53.3% in suburban areas, and 3.9% in rural areas. Nearly 60% of the students reported living with two biological parents. More than 13% of the students reported having both parents with college degrees; the mean of students’ family annual incomes was 38.43 thousand dollars (SE = 1.76). More than 40% of the students (42.1%) reported that they did not work at all; nearly half of the students (44.9%) indicated moderate employment, and the rest of the students (13%) reported intensive employment. Two thirds of the students reported that their parents were employed, with 61.7% of the students having at least one parent working full time and 5.2% having parents who worked only part time; 8.4% of the students reported that their parents did not work. For neighborhood characteristics, the mean of unemployment rates was 7.71 (SE = 0.35), immigrant density 20.6 (SE = 3.29), and neighborhood poverty 25.28 (SE = 1.8). Finally, more than 40% of the students (41.1%) reported engaging in property crime and 44.6% in violent behaviors within 12 months prior to the interview. Table 1 summarizes characteristics of the study sample.

Table 1.

Sample Characteristics.

	N	%
Sample size	4,201	100.0
Male	2,089	49.7
Non-Hispanic White	608	14.5
Non-Hispanic Black	309	7.4
Asian	1,056	25.2
Hispanic	2,105	50.2
First generation	1,455	34.6
Second generation	2,746	65.4
Urban	1,797	42.8
Suburban	2,238	53.3
Rural	165	3.9
Two biological parents	2,495	59.4
Both parents with college degrees	552	13.1
Youth’s employment
Intensive employment	546	12.9
Moderate employment	1,890	44.9
Unemployed	1,765	42.1
Parental employment
Full time	2,591	61.7
Part time	217	5.2
Unemployed	354	8.4
Self-reported violence	1,875	44.6
Youth’s self-reported property crime	1,728	41.1
	Mean	SD
Age (in year)	16.10	0.18
Family income (in thousands)	38.43	1.76
Immigrant density	20.60	3.29
Neighborhood poverty	25.28	1.80
Unemployment rate	7.71	0.35
Exposure to violence	0.19	0.17

Findings

Bivariate Analysis

Table 2 shows significant bivariate relationships between most of the sociodemographic variables and youth’s self-reported violence. Students who reported violence were significantly younger than those who did not report; the mean age of students who reported violent is 15.89 years compared to 16.27 years among those who did not report violence (t = 3.28; p < .01). Males (55.5%) were significantly more likely than females (38.2%) to report violent behavior (adj. F = 47.19; p < .001). Among racial and ethnic groups, Hispanics (52.3%) and Blacks (44.5%) were more likely than Asians (40.1%) and Whites (41.9%) to report violent behavior (adj. F = 4.51; p < .05). Second-generation students (49.5%) were significantly more likely than their first-generation counterparts (41.3%) to report violence (adj. F = 15.4; p < .001). Students living with two biological parents (43.0%) were less likely than other students (53.8%) to report violent behavior (adj. F = 8.64; p < .001). Similarly, students whose both parents had college degrees (34.8%) were significantly less likely than other students (49.2%) to report violence (adj. F = 9.49; p < 0.001). In addition, students in families with the highest income level (40.5%) were less likely than those in families with lower income levels (49.3% and 49.5%) to report violence (adj. F = 5.73; p < 0.01). Residential locations are not significantly related to self-reported violence.

Table 2.

Distribution of Youth’s Self-Reported Violence by Sociodemographic, Independent, and Contextual Variables.

Estimates (%) N = 4,201
	% Yes (SE)	Adjusted F
Sex		47.19***
Male	55.5 (2.1)
Female	38.2 (1.9)
Race		4.51**
Non-Hispanic White	41.9 (2.9)
Non-Hispanic Black	44.5 (5.5)
Asian	40.1 (2.6)
Hispanic	52.3 (2.3)
Immigrant status		15.4***
First generation	41.3 (1.8)
Second generation	49.5 (1.9)
Residential location		5.74
Urban	48.4 (2.2)
Suburban	44.9 (2.3)
Rural	48.8 (1.6)
Family structure		8.64***
Two biological parents	43.0 (1.7)
Other types of parents	53.8 (2.6)
Parental education		9.46***
Both parents with college degrees	34.8 (2.6)
Other	49.2 (5.2)
Family income		5.73**
Low	49.5 (2.3)
Middle	49.3 (2.1)
High	40.5 (2.5)
Youth’s employment		3.08*
Intensive	54.5 (3.5)
Moderate	46.9 (2.0)
Unemployed	44.7 (2.2)
Parental employment		3.75*
Full time	47.2 (1.9)
Part time	45.9 (5.2)
Unemployed	57.3 (3.9)
Youth’s property crime		283.86***
Yes	68.7 (2.1)
No	31.5 (1.7)
Unemployment rates		0.91
Low	43.8 (3.3)
Middle	47.5 (1.9)
High	48.6 (2.6)
Neighborhood poverty		3.59*
Low	44.2 (3.6)
Middle	44.5 (1.9)
High	52.5 (2.3)
Immigrant density		0.40
Low	45.4 (2.6)
Middle	48.2 (2.3)
High	46.2 (3.6)

Estimates (Means) N = 4,201
	Yes (SE)	No (SE)	t test
Age	15.89 (0.20)	16.27 (0.12)	3.28**
Exposure to violence	0.33 (0.30)	0.07 (0.02)	11.03***

*Significant at .05 level; **Significant at .01 level; ***Significant at .001 level.

Results of bivariate analysis show significant relationship of youth’s employment, parental employment, and youth’s self-reported property crime with youth’s self-reported violence. Intensively employed students (54.5%) were significantly more likely than students with moderate employment (46.8%) and unemployed students (44.7%) to report violent behavior (adj. F = 3.08; p < .05). Students who reported property crime (68.7%) were significantly more likely than others (31.5%) to report violent behavior (adj. F = 283.86; p < .001). In addition, students whose parents were unemployed (57.3%) were more likely than those whose parents worked part time (45.9%) or full time (47.2%) to report violent behavior (adj. F = 3.75; p < .05). Of four contextual variables, only two (neighborhood poverty and exposure to violence) are significantly associated with self-reported delinquency. Students living in the neighborhood with the highest level of poverty (52.5%) were more likely than those who lived in areas with lower levels of poverty (44.5% and 44.2%) to report violent behavior (adj. F = 3.59; p < .05); students who reported violent behavior observed (or watched) more violent incidents (stabbing and shooting) than those who did not report violence (means = 0.33 and 0.07, respectively; t = 11.03; p < .001). Unemployment rates and immigrant density are not significantly associated with youth’s self-reported violence.

Multivariate Analysis

Table 3 presents results of logistic regressions. In Model 1, most of the sociodemographic variables, except residential location, were significantly associated with youth’s self-reported violence. Age is negatively related to youth’s self-reported violence in consistence with the results of bivariate analysis. As age increases by 1 year, the odds of violence is lower by 11% (b = −0.12; t = −3.54; p < .001; odds ratio [OR] = 0.89). The odds of violence among males is 121% higher than that among females (b = 0.79; t = 7.12; p < .001; OR = 2.21). There is no significant differences in the odds of violence among Blacks, Asians, and Whites, but the odds of violence among Hispanics is 66% higher than that among Whites (b = 0.51; t = 2.96; p < .01; OR = 1.66). The odds of violence among first-generation students is 31% lower than that among their second-generation counterparts (b = −0.37; t = −3.39; p = .001; OR = 0.69). Living with both biological parents and having both parents with college degree is associated with lower odds of violence. The odds of violence among students living with both biological parents is 35% lower than that among other students (b = −0.43; t = −3.39; p = 0.001; OR = 0.65), and the odds of violence among students having both parents with college degree is 29% lowers than that among other students (b = −0.34; t = −2.54; p = .012; OR = 0.71). The odds of violence among students living in families with the middle level of income is 34% higher than that among students living in families with the highest level of income distribution (b = 0.30; t = 2.20; p = .029; OR = 1.34), but there is no difference in the odds of violence between students living in families with the lowest and the highest levels of incomes.

Table 3.

Results of Hierarchical Logistic Regression.

Variables	Model 1 (N = 4,201)			Model 2 (N = 4,201)			Model 3 (N = 4,201)
Variables	B	SE	OR	B	SE	OR	B	SE	OR
Age	−0.12***	0.04	0.89	−0.13**	0.04	0.88	−0.15***	0.04	0.86
Male	0.79***	0.11	2.21	0.74***	0.13	2.10	0.70***	0.14	2.02
Race/ethnicity (ref: White)
Non-Hispanic Black	0.04	0.24	1.05	−0.01	0.25	0.99	−0.19	0.25	0.83
Asian	0.25	0.17	1.28	0.16	0.20	1.17	0.11	0.21	1.12
Hispanic	0.51**	0.17	1.66	0.42*	0.19	1.53	0.32	0.20	1.38
Residential location (ref: rural)
Urban	−0.14	0.26	0.87	−0.16	0.40	0.85	−0.26	0.36	0.77
Suburban	−0.05	0.27	0.95	−0.09	0.40	0.92	−0.10	0.37	0.90
First generation (ref: second generation)	−0.37***	0.11	0.69	−0.15	0.12	0.86	−0.08	0.13	0.92
Two bio parents (ref: other types of parent)	−0.43***	0.13	0.65	−0.31*	0.14	0.73	−0.22	0.14	0.81
Both parents with college education	−0.34*	0.13	0.71	−0.32*	0.14	0.72	−0.31*	0.15	0.74
Family income (ref: high)
Family income—Low	0.23	0.17	1.26	0.40*	0.19	1.48	0.29	0.19	1.33
Family income—Middle	0.30*	0.13	1.34	0.39*	0.15	1.48	0.29	0.15	1.34
Youth’s employment (ref: moderate)
Intensive				0.56**	0.18	1.76	0.53**	0.19	1.70
Unemployed				−0.14	0.12	0.87	−0.14	0.13	0.87
Parental employment (ref: full time)
Unemployed				0.45*	0.19	1.56	0.49*	0.20	1.63
Part time				−0.28	0.25	0.76	−0.21	0.25	0.82
Youth’s property crime				1.52***	0.10	4.58	1.41***	0.10	4.10
Immigrant density (ref: high)
Low							0.15	0.18	1.16
Middle							0.26	0.16	1.30
Neighborhood poverty (ref: high)
Low							−0.27	0.20	0.77
Middle							−0.31*	0.14	0.73
Unemployment rate (ref: high)
Low							0.16	0.19	1.17
Middle							0.18	0.15	1.20
Exposure to violence							1.13***	0.15	3.10
	Nagelkerke = 0.10			Nagelkerke = 0.25			Nagelkerke = 0.30

*Significant at .05 level; **Significant at .01 level; ***Significant at .001 level.

In Model 2 (Table 3), all three independent variables were significantly associated with youth’s self-reported violence but in different directions. Inconsistent with Hypothesis 1, unemployment is not associated with youth violence because there is no difference in the odds of self-reported violence between moderately employed students and unemployed students. On the other hand, consistent with Hypothesis 2, youth’s extensive employment is associated with higher odds of self-reported violence. The odds of violence among intensively employed students is 76% higher than that among moderately employed students (b = 0.56, t = 2.96, p < .01; OR = 1.76). Also consistent with Hypothesis 3, youth’s property crime is significantly associated with higher odds of youth’s self-reported violence. The odds of violence among students involved in property crime is more than 3 times higher than that among those who were not involved in property crime (b = 1.52; t = 15.82; p < .001; OR = 4.58). In addition, consistent with Hypothesis 4, parental unemployment is associated with higher odds of youth’s self-reported violence. The odds of violence among students whose parents did not work is 56% higher than that among students whose parents were employed full time (b = 0.45; t = 2.40, p < .05; OR = 1.56). However, there is no difference in the odds of violence between students whose parents worked only part time and those whose parents worked full time.

The mediating effects of independent variables on the relationship between demographic variables and youth violence are found in Model 2 where the coefficient for immigrant status becomes insignificant, suggesting that higher level of youth violence in the second generation is explained by these independent variables. Additional analysis indicates that second-generation students are more likely than their first-generation counterparts to be involved in property crime (46% and 32% respectively; adj. F = 25.83; p < .001) and intensive employment (13% and 10% respectively; adj. F = 11.42; p < .001). Other sociodemographic variables remain significantly associated with youth’s self-reported violence in the same directions as in the control model (Model 1).

Model 3 (Table 3) shows that two contextual variables (neighborhood poverty and exposure to violence) are significantly related to youth’s self-reported violence. The odds of violence among students living in neighborhoods with the middle level of poverty is 27% lower than the odds of violence among students living in neighborhoods with the highest level of poverty (b = −0.31, t = −2.22; p = .03; OR = 0.73). In addition, as exposure to violence increases by one level, the odds of violence increases by 205% (b = 1.12; t = 7.71; p < .001, OR = 3.05). All three independent variables remain significantly associated with youth’s self-reported violence independent of the effects of contextual variables. Three demographic variables race/ethnicity, family structure, and family income become insignificant in Model 3. This suggests that two significant contextual variables mediate the effects of race/ethnicity, family structure, and family incomes on youth violence. In other words, high level of neighborhood poverty and exposure violence explains high rates of violence among Hispanic students, students from low-income families, and students who did not live with two biological parents. Other sociodemographic variables (age, sex, and parent’s education) remain significantly associated with youth’s self-reported violence.

Discussions and Conclusion

The present study explores the effect of both legal and illegal economic opportunities on violence among immigrant youth. Most of the research hypotheses (Hypotheses 2, 3, and 4) are supported by data from the study. Youth’s intensive employment, parental unemployment, and youth’s involvement in property crime are associated with higher levels of youth’s self-reported violence. Hypothesis 1 is not supported by data from the study because there is no significant difference in the odds of violence between unemployed students and students with moderate employment.

There are several explanations for a lack of association between youth’s unemployment and youth violence. First, individual employment may not serve as a good measure for predicting individual criminal behavior. In fact, studies that found significant associations between crime and unemployment tended to use aggregated data, and the crime rates tended to be related to property crime rates (Allan & Steffensmeier, 1989; Britt, 1994; Chiricos, 1987; Crutchfield, 1989). On the other hand, research using individual-level data found no overall effect of employment on either individual criminal behavior or substance use (Apel et al., 2007). Second, the effects of employment on behavior may vary across the life course. Studies found that employment was effective in reducing crime only for adults, especially those who were 25 years old or older (Sampson & Laub, 1993; Uggen, 2000; Wright, Cullen, & Williams, 2002). In addition, ethnographic studies that showed the relationship between unemployment and violence tended to include youth who already dropped out from school (Baron & Hartnagel, 1998; Desroches, 1997; Sullivan, 1989; Williams & Kornblum, 1985). On the other hand, while intensive paid work at younger age has the potential to disrupt academic performance and promote problem behaviors that interfere with longer term educational attainment and adjustment (Lageson & Uggen, 2013; Steinberg & Cauffman, 1995), the literature also suggests that the association between intensive work and crime may be spurious because youth who work intensively may differ from others and may have characteristics that are associated with deviance and delinquency, including their lack of interest in school (Apel et al., 2007).

The significant relationships between two contextual variables (exposure to violence and neighborhood poverty) and youth’s violence as well as their mediating effects on the relationships between race/ethnicity and youth’s violence suggest the contributions of socioeconomic status, especially family income to racial differences in violence, particularly higher level of violence among Latino immigrant youth. Literature indicates that immigrants from Asia, Africa, Europe, and Canada are much more likely than those from Mexico and Latin America to arrive in the United States under the professional emigration category and earn higher incomes; nearly 80% of those immigrants have completed high school education and 40% college education (Portes & Rumbaut, 2006). On the contrary, immigrants from Mexico and Latin America often came to the United States as farm workers or manual laborers; only less than half (44%) of Latino immigrants have completed high school and 10% college education (Portes & Rumbaut, 2006). In addition, the immigration category of most Latino immigrants (temporary working visa, family reunification, or undocumented immigrants) often makes them ineligible for government assistance. Latino immigrants who live in impoverish neighborhoods may not benefit from immigrant revitalization often found in immigrant communities that have small businesses from earlier immigrants. Because socioeconomic status determines residential locations and school quality, Latino immigrants tend to reside in the worst neighborhoods and their children attend the most segregated schools under the worst conditions (Crosnoe, 2006). Negative neighborhood conditions may further contribute to delinquency because gangs in the slum tend to recruit underachievers and lure them into illegal activities that involve violence.

In conclusion, the study suggests important role of parental employment in reducing the risk of violence among immigrant youth as well as potential harms associated with youth’s intensive work while attending school. The role of economic opportunity for immigrant parents is also implied in the relationships between youth violence and neighborhood contexts (neighborhood poverty and exposure to violence), which help explain higher rate of violence among Latino immigrant youth. Findings from the study suggest that the ideal of equal opportunity is still not a reality for many immigrant children in the United States. Rather, their adaptation outcomes are more likely to be determined by their parents’ socioeconomic background and employment. Not only immigrant children from poor families experience economic disadvantage and may have to work at early age, they may be more likely than those from better-off families to face violence associated with intensive work while attending schools or to be exposed to violence commonly found in low income neighborhoods that serve as resettle locations for many immigrants.

Limitations

The study has several limitations that need to take into account when interpreting the study findings. First, because the study used the school-based sample, immigrant youth who already dropped out from schools were not included. The exclusion of the dropouts from the study sample inhibited a full assessment of the relationship between unemployment and youth violence because youth who already dropped out from school may have greater financial needs that can be addressed by illegal activities when legal means are not available. Second, the use of cross-sectional data to avoid bias caused by high attritions in the second wave of data collection also prevented the study from determining causal relationships of youth’s intensive employment and property crime with youth violence. In addition, due to a lack of data, the study was unable to explore differences in terms of personal characteristics between immigrant youth who worked intensively and those who did not work. These data may help determine whether the relationship between intensive employment and violence is spurious and is explained by these characteristics. Future research should address these shortcomings to improve understandings of the relationship between economic opportunities and immigrant youth’s violence.

Policy Implications

Despite the study’s limitations, findings from the study have important implications for social policies aimed at preventing violence among immigrant youth. Support and assistance (e.g., education and vocational training) to improve economic opportunities for disadvantaged immigrant families are necessary to increase incomes for immigrant families and the ability of immigrant parents to address their children’s financial needs, thus decreasing financial stress and the risk of immigrant youth’s involvement in property crime and violence. Because first-generation immigrants (foreign born) tend to earn low incomes, improving economic opportunities for immigrant parents also help reduce intensive work among immigrant children, helping them spend more time for studying and avoid problematic behaviors that may be associated with youth’s intensive work. In addition, improving safety conditions in disadvantaged communities where new immigrants tend to reside can reduce the exposure of immigrant children to violence, thus decreasing the risk of violence among immigrant youth.

Footnotes

Acknowledgments

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website (). No direct support was received from grant P01-HD31921 for this analysis.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Alaniz

M. L.

Cartmill

R. S.

Parker

R. N.

(1998). Immigrants and violence: The importance of neighborhood context. Hispanic Journal of Behavioral Sciences, 20, 155–174.

Allan

Steffensmeier

(1989). Youth, underemployment, and property crime: Differential effects of job availability and job quality on juvenile and young adult arrest rates. American Sociological Review, 54, 107–123.

Apel

Bushway

Brame

Haviland

Nagin

Paternoster

(2007). Unpacking the relationship between adolescent employment and antisocial behavior: A matched samples comparison. Criminology, 45, 67–97.

Bachman

J. G.

Staff

O’Malley

Schulenberg

J. E.

Freedman-Doan

(2011). Twelfth-grade student work intensity linked to later educational attainment and substance use: New longitudinal evidence. Developmental Psychology, 47, 344–336.

Baker

C. K.

Hishinuma

E. S.

Chang

J. Y.

Nixon

D. C.

(2010). The relationship among exposure to stressful life events, drug use, and violence perpetration in a sample of Native Hawaiian, Samoan, and Filipino adolescents. Journal of Interpersonal Violence, 25, 379–399.

Bankston

C. L.

(1998). Youth gangs and the new second generation: A review essay. Aggression and Violent Behavior, 3, 35–45.

Baron

S. W.

Hartnagel

T. F.

(1998). Street youth and criminal violence. Journal of Research in Crime and Deliquency, 35, 166–192.

Becker

G. S.

(1968). Crime and punishment: An economic approach. Journal of Political Economy, 76, 169–217.

Britt

C. L.

(1994). Crime and unemployment among youths in the United States, 1958-1990. American Journal of Economics and Sociology, 53, 99–109.

10.

Bui

H. N.

(2009). Parent-child conflicts, school troubles, and delinquency among immigrants. Crime & Delinquency, 55, 412–441.

11.

Bui

H. N.

(2011). Segmented assimilation: Racial differences in the relationship between immigration status and delinquency. In Coll

C. G.

Marks

(Eds.), Is becoming an American a developmental risk? (pp. 135–158). Washington, DC: American Psychological Association.

12.

Chiricos

T. G.

(1987). Rates of crime and unemployment: An analysis of aggregate research evidence. Social Problems, 34, 187–212.

13.

Crosnoe

(2006). The diverse experiences of Hispanic students in American educational system. Sociological Forum, 20, 261–588.

14.

Crutchfield

R. D.

(1989). Labor stratification and violent crime. Social Forces, 68, 489–512.

15.

Desmond

S. A.

Kubrin

C. R.

(2009). The power of place: Immigrant communities and adolescent violence. Sociological Quarterly, 20, 581–607.

16.

Desroches

F. J.

(1997). Behind the bars: Experiences in crime. Toronto, Canada: Canadian Scholars Press.

17.

Farnworth

Thornberry

Krohn

Lizotte

A. J.

(1994). Measurement in the study of class and delinquency: Integrating theory and research: Journal of Research in Crime and Delinquency, 31, 32–61.

18.

Felson

(1986). Linking criminal choices, routine activities, informal control, and criminal outcomes. In Cornish

D. B.

Clarke

R. V.

(Eds.), The reasoning criminal: rational choice perspectives on offending (pp. 119–28). New York, NY: Springer-Verlag.

19.

Frank

Cerdá

Rendón

(2007). Barrios and burbs: Residential context and health risk behaviors among Angeleno adolescents. Journal of Health and Social Behavior, 48, 283–300.

20.

Freeman

R. B.

(1996). Why do so many young men commit crimes and what might we do about it? Journal of Economic Perspectives, 10, 22–45.

21.

Hallsten

Szulkin

Sarnecki

(2013). Crime as a price of inequality: The gap in registered crime between childhood immigrants, children of immigrants and children of native Swedes. British Journal of Criminology, 53, 456–481.

22.

Harland

Reijneveld

S. A.

Brugman

Verloove-Vanhorick

S. P.

Verhulst

F. C.

(2002). Family factors and life events as risk factors for behavioural and emotional problems in children. European Child & Adolescent Psychiatry, 11, 176–184.

23.

Harris

K. M.

(2011). Design features of add health. Carolina Population Center, University of North Carolina at Chapel Hill. Retrieved from http://www.cpc.unc.edu/projects/addhealth/data/guides/design%20paper%20WI-IV.pdf

24.

Hay

Fortson

E. N.

Hollist

D. R.

Altheimer

Schaible

L. M.

(2007). Compounded risk: The implications for delinquency coming from a poor family that lives in a poor community. Journal of Youth and Adolescence, 36, 593–605.

25.

Hirschi

(1983). Crime and the family. In Wilson

J. Q.

(Ed.), Crime and public policy (pp. 53–68). San Francisco, CA: Institute for Contemporary Studies.

26.

Hirschi

(2009). Cause of delinquency. Piscataway, NJ: Transaction.

27.

Hoffmann

(2002). The community context of family structure and adolescent drug use. Journal of Marriage and Family, 64, 314–330.

28.

Hovey

J. D.

Magan

(2000). Acculturative stress, anxiety and depression among Mexican immigrants in the Midwest United States. Journal of Immigrant Health, 2, 119–131.

29.

Hussey

J. M.

Hallfors

D. D.

Waller

Iritani

Halpern

C. T.

Bauer

(2007). Sexual behavior and drug use among Asian and Latino adolescents: Association with immigrant status. Journal of Immigrant Health, 9, 85–94.

30.

King

R. B.

Harris

K. M.

(2007). Romantic relationships among adolescents. International Migration Review, 41, 344–370.

31.

Kposowa

A. J.

Tsunokai

G. T.

(2003). Offending patterns among Southeast Asians in the State of California. Journal of Ethnicity in Criminal Justice, 1, 93–113.

32.

Kulis

Marsiglia

F. F.

Sicotte

Nieri

(2007). Neighborhood effects on youth substance use in a Southwestern city. Sociological Perspectives, 50, 273–301.

33.

Lageson

Uggen

(2013). How work affects crime and crime affects work over the life course. In Gibson

Krohn

(Eds.), Handbook of life course criminology (pp. 201–212). New York, NY: Springer.

34.

Larzelere

R. E.

Patterson

G. R.

(1990). Parental management: Mediator of the effect of socioeconomic status on early delinquency. Criminology, 28, 301–324.

35.

Lee

E. S.

Forthofer

R. N.

(2006). Analyzing complex survey data (2nd ed). Thousand Oaks, CA: Sage.

36.

Loeber

(1982). The stability of antisocial and delinquent child behavior: A review. Child Development, 53, 1431–1446.

37.

Madge

(1983). Unemployment and its effects on children. Journal of Child Psychology and Psychiatry, 24, 311–319.

38.

Martinez

Jr . (2006). Coming to America: The impact of the new immigration on crime. In Martínez

Jr Valenzuela

Jr . (Eds.), Immigration and crime: Race, ethnicity, and violence (pp. 1–19). New York: New York University Press.

39.

Martinez

Jr Lee

M. T.

Nielsen

A. L.

(2004). Segmented assimilation, local context, and determinants of drug violence in Miami and San Diego: Does ethnicity matter? International Migration Review, 38, 131–157.

40.

Martinez

Jr Stowell

J. I.

Lee

M. T.

(2010). Immigration and crime in an era of transformation: A longitudinal analysis of homicides in San Diego neighborhoods. Criminology, 48, 797–829.

41.

Mazerolle

(1998). Gender, general strain, and delinquency: An empirical examination. Justice Quarterly, 15, 65–91.

42.

McLoyd

V. C.

(1989). Socialization and development in a changing economy: The effects of paternal job and income loss on children. American Psychologist, 44, 293–302.

43.

Merton

(1938). Social structure and anomie. American Sociological Review, 3, 672–682.

44.

Morenoff

J. D.

Astor

(2006). Immigrant assimilation and crime: Generational differences in youth violence in Chicago. In Martinez

Jr Valenzuela

Jr . (Eds.), Immigration and crime: Race, ethnicity, and violence (pp. 36–63). New York: New York University Press.

45.

Osgood

D. W.

(1999). Having the time of their lives: All work and no play? In Booth

Crouter

A. C.

Shanahan

M. J.

(Eds.), Transitions to adulthood in a changing economy (pp. 176–186). Westport, CT: Praeger.

46.

Passel

(2011). Demography of immigrant youth: Past, present and future. Immigrant Youth, 21, 19–42. Retrieved from www.futureofchildren.org

47.

Paternoster

Brame

(1997). Multiple routes to delinquency? A test of developmental and general theories of crime. Criminology, 35, 49–84.

48.

Paternoster

Bushway

Apel

Brame

(2003). The effect of teenage employment on delinquency and problem behaviors. Social Forces, 82, 297–335.

49.

Pena

J. B.

Wyman

P. A.

Brown

H. C.

Matthieu

M. M.

Olivares

T. E.

Hartel

Zayas

L. H.

(2008). Immigration generation status and its association with suicide attempts, substance use, and depressive symptoms among Latino adolescents in the USA. Prevention Sciences, 9, 299–310.

50.

Portes

(1996). Preface. In Portes

(Ed.), The second generation (pp. 9–10). New York, NY: Russell Sage Foundation.

51.

Portes

Rumbaut

(2006). Immigrant America: A portrait. Berkeley: University of California Press.

52.

Portes

Zhou

(1993). The new second generation: Segmented assimilation and its variants. Annals of the American Academy of Political and Social Science, 530, 74–96.

53.

Ramey

(2013). Immigrant revitalization and neighborhood violent crimes in established and new destination cities. Social Force, 92, 579–629.

54.

Reid

L. W.

Weiss

H. E.

Adelman

R. M.

Jaret

(2005). The immigration–crime relationship: Evidence across US metropolitan areas. Social Science Research, 34, 757–780.

55.

Ritakallio

Kaltiala-Heino

Kivivuori

Rimpelä

(2005). Delinquent behaviour and depression in middle adolescence: A Finnish community sample. Journal of Adolescence, 28, 155–159.

56.

Rosenwaike

Hempstead

(1990). Mortality among three Puerto Rican populations: Residents of Puerto Ricans in New York and immigrants in New York City and in the balance of the United States, 1979-1981. International Migration Review, 24, 684–702.

57.

Rumbaut

(1996). Origins and destinies: Immigration, race, and ethnicity in contemporary America. In Pedraza

Rumbaut

(Eds.), Origins and destinies: Immigration, race, and ethnicity in America (pp. 21–42). Belmont, CA: Wadsworth.

58.

Sampson

R. J.

(2008). Rethinking crime and immigration. Contexts, 7, 28–33.

59.

Sampson

R. J.

Laub

J. H.

(1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

60.

Sampson

R. J.

Wilson

W. J.

(1995). Toward a theory of race, crime, and urban inequality. In Hagan

Peterson

R. D.

(Eds.), Crime and inequality (pp. 37–54). Stanford, CA: Stanford University Press.

61.

Stowell

J. I.

Messner

S. F.

McGeever

K. F.

Raffalovich

L. R.

(2009). Immigration and the recent violent crime drop in the United States: A pooled, cross-sectional time series analysis of metropolitan areas. Criminology, 47, 889–928.

62.

Steinberg

Cauffman

(1995). The impact of employment on adolescent development. In Vasta

(Ed.), Annals of child development (Vol 11, pp. 131–166). London, England: Kingsley.

63.

Sullivan

(1989). Getting paid: Youth crime and work in the inner city. Ithaca, NY: Cornell University Press.

64.

Turney

Kao

(2011). Behavioral outcomes in early childhood: Immigrant paradox or disadvantage. In Coll

C. G.

Marks

(Eds.), Is becoming an American a developmental risk? (pp. 79–108). Washington, DC: American Psychological Association.

65.

Uggen

(2000). Work as a turning point in the lives of criminals: A duration model of age, employment, and recidivism. American Sociological Review, 65, 529–546.

66.

Unger

Ritt-Olson

Soto

Baezconde-Garbanati

(2009). Parent-child acculturation discrepancies as a risk factor for substance use among Hispanic adolescents in Southern California. Journal of Immigrant Health, 11, 149–157.

67.

US Census Bureau (2001). Profile of the foreign-born population in the United States: 2000. Washington, DC: US Department of Commerce.

68.

Velez

(2009). Contextualization the immigration and crime effect: An analysis of homicides in Chicago neighborhood. Homicide Studies, 13, 325–336.

69.

Williams

Kornblum

(1985). Growing up poor. Laham, MD: Lexington Books.

70.

Wortley

(2008). A province at the crossroads: Statistics on youth violence in Ontario. In McMurtry

Curling

(Eds.), The review of the roots of youth violence (Vol 4, pp. 1–64). Toronto: Queens Printer for Ontario.

71.

Wright

J. P.

Cullen

F. T.

Williams

(1997). Working while in school and delinquent involvement: Implications for social policy. Crime and Delinquency, 43, 203–221.

72.

Wright

J. P.

Cullen

F. T.

Williams

(2002). Embededness of adolescent employment and participation in delinquency: A life course perspective. Western Criminology Review, 4, 1–19.

73.

Zhou

Bankston

C. L.

(1998). Growing up American: How Vietnamese children adapt to life in the United States. New York, NY: Russell Sage.