Buffering Effects of Religiosity on Crime

Abstract

Studies examining the protective effect of religiosity on crime are frequently rooted in the assumption that the impact of religiosity is invariant across sociodemographic differences. This study systematically examines the validity of this assumption across gender and the developmental periods of adolescence and young adulthood. Using a nationally representative sample of adolescents (n = 90,202) and young adults (n = 93,710), negative binomial regression (NBR) is employed to examine the associations between religiosity and criminal behaviors (e.g., drug selling, theft) among male and female adolescents and young adults. Results indicate that the protective relationship between religiosity and criminal behaviors such as drug selling and theft is consistent across gender as well as across the developmental periods of adolescence and young adulthood. This study provides support for the validity of the invariance hypothesis as the protective effect of religiosity on criminal behavior was consistently observed across important sociodemographic differences.

Keywords

religiosity drug selling theft adolescence young adulthood

It is well established that religious engagement serves to deter youth from involvement in antisocial and criminal behavior (B. R. Johnson, 2011; H. G. Koenig, King, & Carson, 2012). Indeed, systematic reviews of the religiosity literature point to the consistent identification of the protective effect of religiosity on crime (B. R. Johnson & Jang, 2010) and suggest that this deterrent effect is moderate in size (r = −.12 to −.24; Baier & Wright, 2001). Moreover, an extensive body of scholarship indicates that such effects can be identified both in terms of public participation in religious services and privately held religious beliefs (Benda, Pope, & Kelleher, 2006; Good & Willoughby, 2006; B. R. Johnson, Larson, De Li, & Jang, 2000; Ryan, Testa, & Zhai, 2008; Salas-Wright, Olate, & Vaughn, 2013; Salas-Wright, Olate, Vaughn, & Tran, 2013; Salas-Wright, Vaughn, Hodge, & Perron, 2012; Smith & Faris, 2002). Beyond direct effects, research also suggests that religious engagement serves to moderate the deleterious impact of a variety of key psychosocial risk factors (e.g., peer attitudes, pressure, and behavior; stress and strain, sensation seeking, and exposure to community violence) on problem behavior (Bahr & Hoffman, 2008; Desmond, Soper, & Kraus, 2011; Fowler, Ahmed, Tompsett, Jozefowicz-Simbeni, & Toro, 2008; Jang & Johnson, 2005). Simply, the body of evidence seems to point to a straightforward empirical conclusion that “more religion” is associated with “less crime” (DiIulio, 2009).

Importantly, however, most studies examining these relationships are rooted in the assumption that the impact of religiosity on problem behavior is invariant across sociodemographic differences. As such, the majority of studies have omitted the examination of factors such as gender or developmental age group (e.g., adolescence, young adulthood) via sample stratification or interaction effects. This approach raises important questions as evidence suggests that differences in religious engagement can be identified across gender and between adolescents and young adults. Indeed, in terms of age differences, adolescents have been found to report substantially greater rates of religious service attendance as well as slightly higher levels of private religiosity and theistic beliefs compared with young adults (Smith & Snell, 2009). Similarly, female youth consistently report greater religious service attendance, youth group participation, and private religiosity compared with their male counterparts (Kerestes, Youniss, & Metz, 2004; Smith, Faris, Denton, & Regnerus, 2003; Smith & Snell, 2009). In addition, a small but noteworthy collection of studies have systematically examined and identified some differences in the effect of religiosity on problem behavior across sociodemographic differences (Brown, Parks, Zimmerman, & Phillips, 2001; Button, Hewitt, Rhee, Corley, & Stallings, 2010; Caputo, 2008; M. C. Johnson & Morris, 2008; Kim-Spoon, Longo, & McCullough, 2012; Milot & Ludden, 2009; Salas-Wright, Vaughn, & Maynard, 2013). In the face of these demographic trends and preliminary empirical evidence suggesting variability in religiosity by age and gender, the assumption that the relationship between religiosity and criminal behavior is invariant warrants further investigation.

Drawing from a nationally representative sample of adolescents (ages 12-17) and young adults (ages 18-25), the aim of this study is to test the invariance hypothesis regarding the effect of religiosity on crime. The invariance hypothesis asserts that the protective effects of religious service attendance and religious beliefs on criminal behavior are general and therefore consistent across gender and the developmental periods of adolescence and young adulthood. To examine this hypothesis, two principal questions guide this study: First, does the effect of religiosity on criminal behavior vary by gender and between adolescents and young adults? And, second, does the moderating effect of religiosity on key psychosocial risk factors for criminal behavior vary across these categories as well? Two criminal behaviors were selected for examination in this study: drug selling and theft. Drug selling and theft were selected as they are substantively appropriate for examination as they serve as “mid-range” criminal behaviors that are both serious yet sufficiently common during the periods of adolescence and young adulthood. Furthermore, these criminal behaviors were identified as the only two criminal outcome variables that were common across the adolescent and young adult data sets examined in this study. In all, research has clearly demonstrated that religiosity exerts a protective effect on antisocial and criminal behavior, but far less is understood in terms of the effect of this relationship across key sociodemographic differences. Taking advantage of a large, nationally representative data set, this study examines the invariance hypothesis by systematically comparing the influence of religiosity on two major crimes (i.e., illicit drug selling and theft) in the lives of male and female adolescents and young adults.

Religious Engagement and Criminal Behavior

Research has consistently suggested that religious engagement exerts a protective, or deterrent, effect on a variety of criminal behaviors. Baier and Wright (2001) demonstrated convincingly in their meta-analysis of 60 articles examining the relationship between religiosity and crime that public religious involvement and private religious beliefs have a consistently inverse (r = −.12) deterrent effect on criminal behavior. Notably, Baier and Wright found this deterrent effect to be significantly more robust (r = −.24) in reference to nonviolent crimes such as drug related and “non-victim” criminal behaviors. B. R. Johnson and Jang (2010) found similar results in a comprehensive assessment of 270 articles published between 1944 and 2010 examining the relationship between religiosity and crime—more than 90% of the articles found religiosity to be protective for antisocial behavior. Recent studies using multidimensional measures of religiosity have identified similar deterrent effects in terms of crime in general (Benda, 2002; Jang & Johnson, 2011; B. R. Johnson, Jang, Larson, & De Li, 2001; Pickering & Vazsonyi, 2010; Schreck, Burek, & Clark-Miller, 2007) and minor forms of delinquency such as status offenses and property crimes (Benda, 1995). Notably, evidence suggests that the protective effect of religiosity may relate more to the initiation of criminal or problem behavior than to the desistence from crime (Ulmer, Desmond, Jang, & Johnson, 2010, 2012).

Similar effects have been identified when examining the two primary subcomponents of religiosity: religious service attendance and private religiosity. Frequent attendance at religious services has been found to be inversely associated with minor crimes such as petty theft and vandalism (Good & Willoughby, 2006; Petts, 2009), severe criminal behaviors such as felony theft and drug selling (B. R. Johnson, Jang, De Li, & Larson, 2000; B. R. Johnson, Larson, et al., 2000), as well as domestic violence (Ellison, Trinitapoli, Anderson, & Johnson, 2007) and involvement in the criminal justice system (Ryan et al., 2008). Private religiosity, which refers to the importance ascribed by individuals to their religious beliefs, has been found to exert a similar deterrent effect on crime. Elevated levels of private religiosity are associated with the decreased likelihood of participation in minor crimes such as status and property offenses (Benda & Corwyn, 1997, 2001), more severe criminal behaviors such as larceny and armed robbery (Smith & Faris, 2002), and the justification of criminal involvement (Koster, Goudriaan, & van der Schans, 2009). In all, religious engagement, be it public religious service attendance or private religiosity, has been found to be protective against both minor and severe forms of criminal behavior.

Religiosity as a Moderator of Psychosocial Risk and Problem Behavior

In addition to direct effects, religious engagement has also been identified as an important moderating factor that can serve to buffer the relationship between psychosocial risk factors and involvement in problem behavior. The list of psychosocial risk factors for crime and antisocial behavior is far-reaching, including factors across the domains of genetic, dispositional, psychoemotional, peer, and community risk (DeLisi, Beaver, Vaughn, & Wright, 2009; Hittner & Swickert, 2006; Loeber & Farrington, 1998). In reference to religiosity as a moderating factor, Jang and Johnson (2003) found in a study of a nationally representative sample of African American adults that religiosity moderated the relationship between negative emotions, such as depressed affect and temper, and nonviolent and violent deviant behavior. Additional studies have examined the moderating effect of religiosity across gender, but the results are somewhat inconclusive. In a national sample of African American adults, Jang and Johnson (2005) found that the moderating effect of organizational religiosity on the relationship between emotional distress and aggressive behavior was significant among females, but not males. Button and colleagues (2010), however, in a study of monozygotic and dizygotic twin pairs, examined the moderating effect of religiosity on the genetic variance of deviant behavior among both adolescent boys and girls. In the study, religiosity was found to significantly moderate the relationship across gender.

A variety of studies have also examined the moderating effect of religiosity on the relationship between psychosocial risk factors and the use and abuse of drugs and alcohol (Bahr & Hoffman, 2008; Desmond et al., 2011; Fowler et al., 2008). For instance, Desmond and colleagues found that religiosity significantly buffered the positive relationship between peer factors and substance use in a national sample of adolescents. Bahr and Hoffman identified a similar relationship in terms of religiosity moderating the positive relationship between peer substance use and adolescent substance use. However, additional studies examining the moderating effect of religiosity have been designed to separately tap public and private religiosity and have found more nuanced results. Fowler and colleagues identified public but not private religiosity as a significant moderator of the relationship between exposure to community violence and substance use among a diverse sample of young adults. Author and colleagues found contradictory findings in examining the moderating effect of religiosity on the relationship between sensation seeking, substance use views, and adolescent substance use: Private religiosity significantly moderated the relationship, but no significant interaction effect was identified for public religiosity (Salas-Wright, C.P. et al., under review). Overall, while results are somewhat mixed and further research into the nuances of religiosity, gender, and risk factors is warranted, the overall pattern of findings suggests that religious engagement tends to buffer against the deleterious effects of a variety of key psychosocial risk factors.

Religiosity Across Gender and Developmental Age Groups

While a robust body of literature has established the relationship between religiosity and criminal and antisocial behavior, only a relatively limited number of studies have systematically examined these relationships across the lines of key sociodemographic factors such as gender and developmental age. Several studies suggest that the protective effect of religiosity is particularly robust among females. Caputo (2008), using a national sample of adolescents, found that the relationship between religiosity and risk behavior was moderated by gender, suggesting that the protective effect of religiosity was more robust among females than among males. Along the same lines, Brown and colleagues (2001) found similar results among African American and White adolescents as the protective effect of religiosity for problem drinking was greater among female adolescents than male adolescents. Still other studies have found the impact of religiosity not simply to be different in magnitude, but to be significant only among females and not males (M. C. Johnson & Morris, 2008). In contrast, some studies have found few differences in the relationship between religiosity and violence across gender (Salas-Wright et al., 2013), while others have found religiosity to be more strongly associated with positive mental health and educational outcomes among males than females (Kim-Spoon et al., 2012; Milot & Ludden, 2009). Finally, there exists limited evidence that religiosity may be more protective among adolescents than among young adults (Button et al., 2010). In sum, studies that have examined the invariance of the buffering effects of religiosity on crime across gender and developmental age differences are few in number and have generated somewhat mixed results. This suggests that further research into elucidating the role of gender and developmental stage in the relationship between religiosity and crime is needed.

Conceptual Framework

The conceptual framework for this study is rooted in the empirical research and theory on the links between self-control and problem behavior. Although numerous discipline-specific definitions exist, self-control can be understood as the capacity of individuals to intentionally refrain from involvement in immediately gratifying behaviors on the basis of either subsequent expected benefit or conformity with social or moral expectations (Strayhorn, 2002). As noted by Gottfredson and Hirschi (1990), the capacity of individuals to exhibit, and potentially enhance, such self-regulation has far-reaching implications for involvement in deviant, antisocial, and criminal behavior.

Self-control has been found to be relatively stable over time and is therefore typically understood as an enduring trait or skill (de Ridder, Lensvelt-Mulders, Finkenauer, Stok, & Baumeister, 2012); however, despite the general tendency of self-control to remain stable among individuals, research suggests that an important degree of plasticity nevertheless exists (Vohs & Baumeister, 2011). In fact, scholars have increasingly come to conceptualize the development of self-control as analogous to the development of skeletal muscle which, although fatigued in the short-run by exertion, can be strengthened by sustained training or exercise (Baumeister, Vohs, & Tice, 2007; Hagger, Wood, Stiff, & Chatzisarantis, 2010). To this end, the enhancement of self-control has become an increasingly important approach to prevention and intervention efforts designed to target delinquency and crime (Piquero, Jennings, & Farrington, 2010).

Recent scholarship has highlighted the role that religiosity and religious engagement can play in the promotion of self-control, self-regulation, and, consequently, the likelihood of the involvement of individuals in antisocial and criminal behavior (McCullough & Willoughby, 2009). From this vantage point, religion may be conceptualized as exerting a “treatment effect” on criminal behavior by means of enhancing the self-regulatory capacity of individuals. The enhancement of self-control by means of religious involvement might take place via cognitive-based and behavioral driven pathways. In terms of a cognitive pathway, religious traditions often include behavioral proscriptions as well as communal narratives that speak to the importance of self-discipline, moral behavior, and the capacity to control one’s behavior (Smith, 2003). With respect to a behavioral pathway, religious communities typically encourage the involvement of individuals in regular, disciplined practices such as private prayer and meditation, involvement in service and charitable giving, and participation in frequent public religious services (Smith & Denton, 2005). If practiced routinely, both these cognitive and behavioral components may assist individuals in strengthening their capacity for self-control and self-regulation, thereby decreasing the likelihood of involvement in impulsive, risky, or criminal behavior.

In all, religion has been found to have important implications for the enhancement of self-control and, in turn, the propensity of individuals to take part in crime. While it is plausible that such a treatment effect might vary by gender and developmental period, we hypothesize that the links between religiosity, self-control, and criminal behavior are consistent across these sociodemographic differences. To that end, we aim to test the invariance hypothesis which states that the link between religion and criminal behavior is stable for male and female adolescents and young adults.

The Present Study

The present study, employing 5 years of cross-sectional data from a large national data set (i.e., the National Survey on Drug Use and Health [NSDUH] 2006-2010), tests the invariance hypothesis by examining the relationship between religiosity and crime across samples stratified by gender and developmental age. The NSDUH is advantageous because it is nationally representative of adolescents and young adults over a 5-year period and also because the size of the data source allows for stratification by various categories without significant reductions in statistical power. As such, this study can explore the relationship of religiosity and drug selling and theft among adolescent males and females as well as among young adult males and females. To this end, two fundamental questions guide this study: First, does the effect of religiosity on criminal behavior vary by gender and between adolescents and young adults? And, second, does the moderating effect of religiosity on key psychosocial risk factors for criminal behavior vary across these categories as well? These questions add further nuance to the examination of religiosity as a protective factor among young people in the United States. Given the differential levels of religious engagement across the lines of gender and age, and the relatively limited and ambiguous body of research on the systematic examination of these differences, a more in-depth understanding of these relationships can enhance theorizing in this area and potentially advance the development and implementation of programs designed to draw upon religiosity in the prevention of crime. In all, while research has demonstrated that religious engagement is of relevance to crime, this study aims to test the invariance hypothesis by examining the relationship between these factors in a more nuanced fashion.

Method

Sample and Procedures

This study is based on public-use data collected between 2006 and 2010 as part of the NSDUH (Substance Abuse and Mental Health Services Administration [SAMHSA], 2011). The NSDUH is designed to provide cross-sectional population estimates of substance use and health-related behaviors in the U.S. general population. It utilizes multistage area probability sampling methods to select a representative sample of the U.S. civilian, non-institutionalized population aged 12 years or older for participation in the study. Multistage sampling designs commonly are used when attempting to provide nationally representative estimates. This is because interviewing all participants is not feasible so larger units are the first stage selected from which subsequent levels of strata are partitioned until individuals from households are selected. With respect to the NSDUH, all 50 states and the District of Columbia were employed. Study participants include household residents; residents of shelters, rooming houses, and group homes; and civilians residing on military bases. To improve the precision of drug use estimates for subgroups, adolescents aged 12 to 17 years and young adults aged 18 to 25 years were oversampled.

NSDUH study participants were interviewed in private at their places of residence. Potential participants were assured that their names would not be recorded and that their responses would be kept strictly confidential. Participants were paid US$30 for their participation. The NSDUH interview utilizes a computer-assisted interviewing (CAI) methodology to increase the likelihood of valid respondent reports of illicit drug use and high-risk behaviors (SAMHSA, 2011). The CAI methodology includes a combination of computer-assisted personal interviewing (CAPI) and audio computer-assisted self-interviewing (ACASI) methodologies. A more detailed description of the NSDUH sampling and data collection procedures are documented in greater detail elsewhere (SAMHSA, 2011).

A total of 280,098 respondents aged 12 years or older completed the survey between 2006 and 2010. Weighted response rates for these years were approximately 90% for household screening and 75% for interviewing (SAMHSA, 2011). Each independent, cross-sectional NSDUH sample was considered representative of the U.S. general population aged 12 years or older. The current study restricted analyses to adolescents aged 12-17 years (n = 90,202) and young adults aged 18-25 years (n = 93,710). The mean age of the subsample of adolescents is 14.58 (SD = 1.69) and that of young adults is 20.90 (SD =1.76). In terms of gender, the adolescent (male = 51.05%, female = 48.95%) and young adult (male = 47.80%, female = 52.20%) respondents are evenly distributed, but both subgroups are unevenly distributed in terms of race/ethnicity. Overall, more than half of the respondents are White (60.2%), 18.5% are Hispanic, and 14.5% are African American. The annual family income of 24.85% of the sample is less than US$20,000; 33.01% have income between US$20,000 and US$49,999; 15.93% have income between US$50,000 and US$74,999; and 26.22% have more than US$75,000 annual family income.

Measures

Criminal Behavior

The frequency of adolescent (n = 3,080; 3.12%) and young adult (n = 5,373; 5.63%) involvement in drug selling was determined based on responses to the following question: “During the past 12 months, how many times have you sold illegal drugs?” Similarly, the frequency of adolescent (n = 4,302; 4.49%) and young adult (n = 3,130; 3.36%) involvement in theft was identified based on responses to the following question: “During the past 12 months, how many times have you stolen something worth more than US$50?” Although the NSDUH measured various delinquency and crime variables, different variables were measured with adolescents and young adults. We selected the two delinquent/criminal behavior variables that were common across the adolescent and young adult data sets to provide comparisons across these two age groups.

Religious Service Attendance and Private Religiosity

Two measures of religiosity were examined in this study: religious service attendance and private religiosity. Religious service attendance was measured by asking respondents: “During the past 12 months, how many times did you attend religious services (excluding special occasions such as weddings, funerals, etc.)?” Consistent with the original NSDUH coding scheme, respondents were categorized into six ordinal groups ranging from no religious service attendance to attendance at more than 52 religious services in the previous year. Private religiosity was assessed by summing two items measuring the degree to which respondents considered religious beliefs to be important to their life and decision making (Cronbach’s α for adolescents = .87; Cronbach’s α for young adults = .89). More precisely, these two items were measured by asking respondents the extent to which they agreed with the following statements: “Your religious beliefs are a very important part of your life” and “Your religious beliefs influence how you make decisions in your life.” Both items had the response format of “strongly disagree” (1), “disagree” (2), “agree” (3), and “strongly agree” (4).

Risk Propensity

Risk propensity was assessed by summing two items related to the frequency of enjoyment of participation in dangerous or risky behaviors (Cronbach’s α for adolescents = .81; Cronbach’s α for young adults = .85). These items measured the frequency of respondent agreement with the following statements, “How often do you get a real kick out of doing things that are a little dangerous?” and “How often do you like to test yourself by doing something a little risky?” Both items had the response format of “never” (0), “seldom” (1), or “sometimes/always” (3) with higher scores indicating greater risk propensity.

Sociodemographic and Mental Health Covariates

The following demographic variables were used: age, race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and Other—American Indian or Alaska Native, Asian, other Pacific Islander or Native Hawaiian, and persons reporting more than one race), and total annual family income (less than US$20,000, US$20,000-US$49,999, US$50,000-US$74,999, and US$75,000 or more). Family income was ascertained by asking respondents: “Of these income groups, which category best represents your total combined family income during the previous calendar year?” Because adolescents are often unable to provide accurate estimates about family household income, responses from an adult or other household member were provided. In addition, we also examined lifetime history of depression and anxiety. This was based on whether respondents were ever told by a doctor or medical professional that they met criteria for either of these disorders.

Statistical Analysis

For all analyses, weighted prevalence estimates and standard errors were computed using Stata 12.1SE (StataCorp, 2013). This system implements a Taylor series linearization to adjust standard errors of estimates for complex survey sampling design effects including clustered data. Only a very insignificant amount of data were missing among key independent (<3%) and dependent variables (<0.50%), therefore listwise deletion was selected as the strategy for addressing missing data.

Negative Binomial Regression (NBR)

A series of eight NBR models were examined to analyze the direct effects and interaction effects between the sociodemographic factors, risk and protective factors, and drug selling and theft across developmental and gender subgroups. NBR was selected because preliminary data analyses revealed that the data for the frequency of drug selling and theft were non-normally distributed with an extreme positive skew. While linear regression strategies can yield acceptable results with such distributions, a more conservative strategy is to use a data analytic methodology designed for count outcomes with Poisson or NBR (Long & Freese, 2006). NBR was selected over Poisson regression as preliminary diagnostic analysis yielded evidence of overdispersion, which suggests that NBR models were preferable to Poisson regression models (Hilbe, 2011). Despite the over-inflation of zeros for the dependent variables, NBR was selected over zero-inflated negative binomial regression (ZINB) due to the lack of a clearly conceptual justification for the assumption that some respondents have absolutely no chance of scoring a non-zero on the dependent variable (Allison, 2012; Long & Freese, 2006).

To examine the moderating effect of religiosity, several sequential steps were taken. To begin, all continuous independent and moderating variables were standardized so as to avoid issues with multicollinearity (Cohen, Cohen, West, & Aiken, 2003). Next, the product (or interaction) terms were created by multiplying together the standardized independent and moderator variables. Finally, two sets of generalized linear regression analyses were conducted for each of the dependent variables. First, NBR was conducted for the sociodemographic and risk/protective factors only; next, an additional NBR was conducted with the sociodemographic, risk/protective factors, and the four interaction terms as well. Incidence rate ratios (IRR) and accompanying confidence intervals (CIs) were estimated for both NBR analyses. IRR, conceptually similar to odds ratios, refer to the ratio of the enactment of a particular behavior, in this case substance use, given distinct conditions (Hilbe, 2011). This multistep regression process ensures that accurate and readily interpretable estimates are provided for both the main effects and the interaction effects (Frazier, Tix, & Barron, 2004).

Results

Trends in Religiosity by Developmental Age and Gender

The mean values for public and private religiosity among adolescents and young adults by gender between 2006 and 2010 are shown in Figures 1 and 2. As can be readily deciphered, the stability of these measures across developmental age and gender groups is high. Nevertheless, differences between these subgroups can be identified. In terms of public religious participation, the mean scores for adolescent females are the highest followed by those of adolescent males. Young adult females scored slightly lower than adolescent males, but—consistent with adolescent gender differences—young adult females scored higher than young adult males. Virtually the same pattern can be identified in terms of the differences between adolescent and young adult males and females in reference to private religiosity: Adolescent females stand out with the highest levels, adolescent males’ mean values are very similar to those of young adult females, and young adult male mean values are noticeably lower than those of the other age/gender groups.

Figure 1:

Adolescent and Young Adult Religious Service Attendance Between 2006 and 2010 Stratified by Gender

Figure 2:

Adolescent and Young Adult Private Religiosity Between 2006 and 2010 Stratified by Gender

Drug Selling

Table 1 displays results of NBR models examining the associations between risk correlates, religiosity, and drug selling. Results at Step 1 indicate a uniform pattern of protection as religious service attendance and private religiosity were both found to be significantly associated with the decreased likelihood of drug selling across all gender and developmental age group permutations. Effects were particularly robust in terms of private religiosity among female adolescents (IRR = 0.61, 95% CI = [0.53, 0.70]) and, to a lesser extent male adolescents (IRR = 0.73, 95% CI = [0.68, 0.79]). The magnitude of the association for public religiosity among female young adults (IRR = 0.67, 95% CI = [0.60, 0.76]) was also noteworthy. Results also revealed that risk propensity is significantly associated with drug selling across all gender/age subgroups. The only significant interaction effect identified was between risk propensity and private religiosity among females (IRR = 1.22, 95% CI = [1.06, 1.41]). As illustrated in Figure 3, among adolescent females with low risk propensity, drug selling is only slightly greater among youth with low private religiosity than among youth with high private religiosity. However, when risk propensity is high, drug selling is disproportionately elevated among adolescents with low private religiosity as compared with those with high private religiosity.

Table 1:

Main and Moderating Effects of Religiosity on Adolescent and Young Adult Drug Selling by Gender

	Adolescents (12-17 Years)				Young Adults (18-25 Years)
	Male (n = 41,624)		Female (n = 40,637)		Male (n = 42,600)		Female (n = 47,369)
	IRR	95% CI	IRR	95% CI	IRR	95% CI	IRR	95% CI
Step 1
Sociodemographic factors
Age	6.18***	[5.09, 7.51]	4.39***	[3.30, 5.83]	0.64***	[0.58, 0.71]	0.58***	[0.49, 0.69]
Race/ethnicity
African American	1.74***	[1.40, 2.17]	1.82**	[1.23, 2.69]	1.84***	[1.55, 2.18]	1.61**	[1.19, 2.19]
Hispanic	1.39*	[1.06, 1.82]	0.75	[0.52, 1.07]	0.82*	[0.69, 0.98]	0.89	[0.67, 1.18]
Other	1.17	[0.79, 1.74]	1.13	[0.66, 1.95]	0.91	[0.74, 1.13]	0.81	[0.55, 1.17]
Family income (in US$)
20,000-49,000	0.77*	[0.61, 0.97]	0.83	[0.58, 1.19]	0.97	[0.84, 1.11]	1.07	[0.87, 1.31]
50,000-74,999	0.62**	[0.45, 0.85]	0.52**	[0.36, 0.76]	0.86	[0.73, 1.02]	0.76	[0.56, 1.04]
>75,000	0.72*	[0.55, 0.94]	0.44***	[0.30, 0.63]	0.93	[0.80, 1.08]	0.78	[0.59, 1.04]
Mental health
Depression	1.62**	[1.23, 2.13]	2.78***	[2.12, 3.63]	1.37**	[1.14, 1.63]	1.98***	[1.56, 2.52]
Anxiety	2.20***	[1.78, 2.22]	2.17***	[1.52, 3.09]	2.19***	[1.80, 2.67]	1.94***	[1.40, 2.69]
Risk propensity	1.99***	[1.78, 2.22]	3.15***	[2.69, 3.69]	1.86***	[1.74, 1.99]	2.48***	[2.23, 2.75]
Religious attendance	0.83**	[0.75, 0.93]	0.82**	[0.71, 0.95]	0.81***	[0.76, 0.87]	0.67***	[0.60, 0.76]
Private religiosity	0.73***	[0.68, 0.79]	0.61***	[0.53, 0.70]	0.79***	[0.75, 0.84]	0.81***	[0.73, 0.90]
Step 2
Interaction effects
Risk propensity × Public	0.98	[0.85, 1.14]	1.05	[0.84, 1.30]	1.03	[0.95, 1.10]	1.07	[0.96, 1.19]
Risk propensity × Private	1.02	[0.92, 1.13]	1.22**	[1.06, 1.41]	1.03	[0.97, 1.09]	1.01	[0.91, 1.12]

Note. Incidence rate ratios in bold are significant at p < .05 or lower. IRR = incidence rate ratios; CI = confidence interval.

* = p < .05; ** = p < .01; *** = p < .001

Figure 3:

Private Religiosity as a Moderator of Risk Propensity and Drug Selling Among Female Young Adults

Theft

Table 2 displays results of NBR models examining the associations between risk correlates, religiosity, and theft. Similar to drug selling, results at Step 1 indicate a nearly uniform pattern of protection. With the exception of public religiosity among adolescent females, public and private religiosity was found to be significantly associated with the decreased likelihood of drug selling across all gender and developmental age stratifications. Effects were especially noteworthy in terms of private religiosity among female (IRR = 0.73, 95% CI = [0.66, 0.80]) and male adolescents (IRR = 0.77, 95% CI = [0.72, 0.83]). Also consistent with drug selling, results indicated that risk propensity is significantly associated with theft across all gender/age subgroups. The only significant interaction effect identified was between risk propensity and public religiosity among males (IRR = 1.13, 95% CI = [1.03, 1.24]). As illustrated in Figure 4, among adolescent males with low risk propensity, theft is virtually identical among youth with low private religiosity and those with high private religiosity. However, when risk propensity is high, theft is markedly greater among adolescents with low private religiosity as compared with those with high private religiosity.

Table 2:

Main and Moderating Effects of Religiosity for Adolescent and Young Adult Theft by Gender

	Adolescent (12-17 Years)				Young Adults (18-25 Years)
	Male (n = 41,632)		Female (n = 40,628)		Male (n = 42,699)		Female (n = 47,395)
	IRR	95% CI	IRR	95% CI	IRR	95% CI	IRR	95% CI
Step 1
Sociodemographic factors
Age	2.00***	[1.72, 2.32]	2.15***	[1.74, 2.67]	0.46***	[0.40, 0.54]	0.47***	[0.39, 0.56]
Race/ethnicity
African American	2.03***	[1.70, 2.43]	2.39***	[1.86, 3.08]	2.03***	[1.61, 2.54]	3.32***	[2.52, 4.37]
Hispanic	1.64***	[1.37, 1.96]	1.25	[0.97, 1.61]	1.38**	[1.10, 1.73]	1.74***	[1.32, 2.31]
Other	1.23	[0.95, 1.61]	1.57*	[1.05, 2.34]	1.53**	[1.17, 2.00]	1.59*	[1.06, 2.39]
Family income (in US$)
20,000-49,000	0.87	[0.72, 1.04]	0.90	[0.69, 1.18]	1.02	[0.85, 1.23]	0.91	[0.72, 1.14]
50,000-74,999	0.65***	[0.53, 0.80]	0.57***	[0.42, 0.77]	1.07	[0.87, 1.33]	0.94	[0.69, 1.28]
>75,000	0.62***	[0.51, 0.76]	0.53***	[0.40, 0.70]	1.17	[0.94, 1.45]	0.77	[0.59, 1.01]
Mental health
Depression	1.69***	[1.34, 2.12]	2.33***	[1.82, 3.00]	1.72***	[1.34, 2.21]	1.80***	[1.41, 2.29]
Anxiety	1.61**	[1.18, 2.18]	1.74**	[1.21, 2.51]	2.46***	[1.78, 3.41]	1.77***	[1.32, 2.39]
Risk propensity	2.29***	[2.10, 2.49]	2.42***	[2.17, 2.71]	1.90***	[1.73, 2.10]	2.24***	[2.02, 2.49]
Religious attendance	0.90**	[0.83, 0.97]	0.93	[0.83, 1.04]	0.94*	[0.86, 1.02]	0.87*	[0.77, 0.98]
Private religiosity	0.77***	[0.72, 0.83]	0.73***	[0.66, 0.80]	0.78***	[0.72, 0.85]	0.83**	[0.74, 0.94]
Step 2
Interaction effects
Risk propensity × Public	1.13**	[1.03, 1.24]	1.01	[0.88, 1.15]	1.04	[0.95, 1.14]	0.98	[0.87, 1.10]
Risk propensity × Private	1.03	[0.95, 1.10]	0.95	[0.86, 1.06]	1.02	[0.93, 1.11]	0.95	[0.86, 1.06]

Note. Incidence rate ratios in bold are significant at p < .05 or lower. IRR = incidence rate ratios; CI = confidence interval.

* = p < .05; ** = p < .01; *** = p < .001

Figure 4:

Private Religiosity as a Moderator of Risk Propensity and Drug Selling Among Female Young Adults

Discussion

Empirical research has demonstrated that religiosity is inversely associated with the participation of young people in criminal and antisocial behavior. However, less is understood in terms of the variability of this relationship across gender and developmental age groups. By and large, findings from this study support the invariance hypothesis in as much as they suggest that the protective relationship between religiosity and criminal behaviors such as drug selling and theft is consistent across gender as well as across the developmental periods of adolescence and young adulthood. Importantly, the invariance of the protective effect of religiosity was identified for both religious service attendance and private religiosity, indicating that the protective effect of both major forms of religious participation was consistent across sociodemographic differences. That said, while the invariance hypothesis was strongly supported in terms of the identification of statistically significant associations, a limited degree of nuance was nevertheless observed in terms of difference in effect sizes. For instance, while the invariance hypothesis held for both religious service attendance and private religiosity, an examination of effect sizes indicates that the protective effect of private religiosity tended to be slightly stronger than that of religious service attendance. As for developmental age, with only one notable exception, the protective effect of religious service attendance and private religiosity tended to be slightly greater among adolescents than among young adults. The aforementioned exception was that the protective effect of religious service attendance for both criminal behaviors examined was more robust among female young adults than among their female adolescent counterparts. With respect to gender, no clear pattern of effect sizes could be observed; however, in reference to drug selling, more robust relationships were identified for private religiosity among adolescent females and religious service attendance among young adult females than among males. Also, in terms of the type of criminal behavior, the effects of religiosity on drug selling tended to be slightly greater than the effects on theft. In all, while several patterns of differences in effect sizes could be observed, such differences were relatively small, suggesting that the protective effect of religiosity on crime across gender and developmental age is effectively uniform.

In addition to the examination of the direct effect associations between religiosity and crime across gender and developmental age, this study also examined the moderating effect of religiosity on the relationship between risk propensity and crime. Although no significant interaction effects were identified among young adults, two significant relationships were identified among adolescents. First, religious service attendance buffered the positive relationship between risk propensity and theft among adolescent males. Second, private religiosity moderated the relationship between risk propensity and drug selling among females, but not males. In both cases, religious engagement was found to be particularly important in terms of deterring youth with high propensity for risk-taking from involvement in criminal behavior. Simply, religious engagement not only has important direct effects but also functions to impede youth crime by means of weakening the deleterious effect of key risk factors.

The findings of this study on the uniformity of the deterrent effect of religious engagement on crime are, to the degree they have been examined in the empirical literature, somewhat novel. Indeed, while previous findings on the differential relationship across gender are mixed (Brown et al., 2001; Caputo, 2008; M. C. Johnson & Morris, 2008; Kim-Spoon et al., 2012; Milot & Ludden, 2009), few studies have systematically examined the role of gender in the religiosity-crime relationship and found no stable pattern of differences in terms of significance. That said, while the significance of the relationship between religious engagement and criminal behavior in the current study was consistent across sociodemographic differences, minor variations in effect sizes were nevertheless observed in terms of developmental age, gender, and type of criminal behavior. In terms of developmental age variation, the identification of a slightly stronger deterrent effect among adolescents than among young adults is consistent with previous empirical research on the effect of religiosity on youth problem behavior (Button et al., 2010). Albeit minor in magnitude, such effect size differences are perhaps unsurprising as studies have suggested that the relationship between religious engagement and social control may be greater during adolescence than during subsequent developmental stages (L. B. Koenig, McGue, Krueger, & Bouchard, 2005). Consequently, it is theoretically coherent that religiosity would have a stronger effect on criminal behavior during adolescence than during the period of young adulthood. As for gender, no pattern of effect size differences was identified in this study. However, as noted above, an effect size difference was observed between male and female young adults in the realm of drug selling; while religious service attendance was significant for both genders, the effects for females who routinely attended religious services were slightly greater than those of males. While evidence suggests that young adult females tend to be slightly more religious than their male counterparts (Smith & Snell, 2009), scant evidence can be identified that provides a compelling explanation for this difference.

In terms of the slight effect size differences observed for the relationship between religiosity and type of criminal behavior, these results are in keeping with the research literature: the effect size for the deterrent effect of religiosity was slightly greater for drug selling than for theft. While ample evidence suggests that religiosity protects against both minor (Benda & Corwyn, 2001; Petts, 2009) and severe (B. R. Johnson, Jang, et al., 2000; B. R. Johnson, Larson, et al., 2000) criminal behavior, studies that have systematically examined the differences in effect size by type of behavior typically have found a stronger effect among minor, rather than more severe criminal behaviors (Baier & Wright, 2001; Good & Willoughby, 2006). For instance, Baier and Wright’s meta-analysis found that the effect of religiosity on crime was stronger in studies of nonviolent crime (r = −.24) as compared with criminality in general (r = −.12). Good and Willoughby found a similar yet slightly more nuanced pattern in which higher levels of religiosity and spirituality were protective for minor delinquency and direct aggression, but not major delinquency. While the typology of minor versus severe crimes may not be applicable to the two criminal behavior variables examined in the study, recent studies on the severity of criminal behavior have nevertheless placed theft slightly ahead of drug selling in terms of severity (Ramchand, MacDonald, Haviland, & Morral, 2009). As such, the slight yet consistently greater effect size for drug selling as compared with theft is in keeping with previous research. Simply, religiosity appears to be more protective among criminal behaviors that are less severe in nature with the size of effects steadily decreasing as behaviors increase in severity.

While the national representativeness and multiyear design of the sample provide strong generalizability, findings from this investigation should be interpreted in light of several important limitations. First, although data were pooled from the NSDUH over a 5-year period (2006-2010), all data were cross-sectional and, consequently, conclusions about causality simply cannot be drawn. Second, the NSDUH data are entirely self-report and, as such, are prone to reporting errors and social desirability bias (Holden, 2010). Third, religiosity variables were measured on the basis of only a single item, and in the case of private religiosity, a two-item, nonstandardized measure. Given that scholars have pointed to the importance of multidimensional and standardized measures of religiosity (Rew & Wong, 2006), the reliance upon the religiosity variables available in the NSDUH is not without limitations. Finally, the NSDUH does not include information on the contextual, situational, or precipitating factors for criminal behavior, which are undoubtedly necessary for a fuller understanding of the relationship between religiosity and crime.

Overall, the findings of the present study support the invariance hypothesis, suggesting that religiosity is uniformly protective for crime across gender and the developmental age groups of adolescence and young adulthood. As such, the study provides an important contribution to our understanding of the deterrent effect of religiosity across key sociodemographic factors, an understudied facet of the religiosity-crime literature. While the pattern of significance for religiosity was consistent across the board, patterns of noteworthy yet relatively minor differences in effect size were nevertheless identified in terms of the type of religiosity, developmental age, and the type of criminal behavior. Private religiosity tended to provide slightly greater protection than religious service attendance across both gender and developmental age groups. In addition, religious engagement in general provided slightly more robust protection among adolescents than among young adults, and the effect of religiosity on crime was weaker for theft than for drug selling. In conclusion, religiosity has often been studied as a protective factor for criminal behavior among youth, but few studies have systematically examined the nature of this relationship across the lines of gender and developmental age groups. The study supports the invariance hypothesis which suggests that, while slight differences in effect size can be identified across certain facets of the religiosity-crime relationship, by and large, the protective effect of religiosity on crime is uniform.

Footnotes

Authors’ Note:

This research was supported in part by Grant Numbers T32 DA016184 and R25 DA026401 from the National Institute on Drug Abuse at the National Institutes of Health.

Christopher P. Salas-Wright, PhD, is an assistant professor in the School of Social Work at the University of Texas at Austin and a faculty affiliate at the Lozano Long Institute of Latin American Studies (LLILAS) at the University of Texas at Austin. His research interests include youth problem behavior, religiosity and spirituality, and adolescent substance abuse and HIV prevention, particularly among Latino youth in both the United States and in Latin America.

Michael G. Vaughn, PhD, is a professor in the School of Social Work at Saint Louis University. He has contributed more than 200 scholarly publications across a wide range of areas. His research interests include youth violence and drug abuse in relation to antisocial behavior over the life course.

Brandy R. Maynard, PhD, is an assistant professor in the School of Social Work at Saint Louis University and a fellow with the Meadows Center for Preventing Educational Risk at The University of Texas at Austin. Her research interests include academic risk and externalizing behavior problems, evidence-based practice, and meta-analytic methods.

References

Allison

P. D.

(2012). Logistic regression using SAS: Theory and application (2nd ed.). Cary, NC: SAS Institute.

Bahr

S. J.

Hoffman

J. P.

(2008). Religiosity, peers, and adolescent drug use. Journal of Drug Issues, 38, 743-769.

Baier

C. L.

Wright

B. R. E.

(2001). If you love me, keep my commandments: A meta-analysis of the effect of religion on crime. Journal of Research in Crime & Delinquency, 38, 3-21.

Baumeister

R. F.

Vohs

K. D.

Tice

D. M.

(2007). The strength model of self-control. Current Directions in Psychological Science, 16, 351-355.

Benda

B. B.

(1995). The effect of religion on adolescent delinquency revisited. Journal of Research in Crime & Delinquency, 32, 446-466.

Benda

B. B.

(2002). A test of three competing theoretical models of delinquency using structural equation modeling. Journal of Social Service Research, 29, 55-91.

Benda

B. B.

Corwyn

R. F.

(1997). Religiosity and delinquency: The relationship after considering family and peer influences. Journal for the Scientific Study of Religion, 36, 81-92.

Benda

B. B.

Corwyn

R. F.

(2001). Are the effects of religion on crime mediated, moderated, and misrepresented by inappropriate measures? Journal of Social Service Research, 27, 57-86.

Benda

B. B.

Pope

S. K.

Kelleher

K. J.

(2006). Church attendance or religiousness. Alcohol Treatment Quarterly, 25, 75-87.

10.

Brown

T. L.

Parks

G. S.

Zimmerman

R. S.

Phillips

C. M.

(2001). The role of religion in predicting adolescent alcohol use and problem drinking. Journal of Studies on Alcohol, 62, 696-705.

11.

Button

T. M. M.

Hewitt

J. K.

Rhee

S. H.

Corley

R. P.

Stallings

M. C.

(2010). The moderating effect of religiosity on the genetic variants of problem alcohol use. Alcoholism: Clinical and Experimental Research, 34, 1619-1624.

12.

Caputo

R. K.

(2008). Religiousness and adolescent behaviors. Journal of Religion & Spirituality in Social Work: Social Thought, 24, 39-67.

13.

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.

14.

DeLisi

Beaver

K. M.

Vaughn

M. G.

Wright

J. P.

(2009). All in the family: Gene × Environment interaction between drd2 and criminal father is associated with five antisocial phenotypes. Criminal Justice and Behavior, 36, 1187-1197.

15.

de Ridder

D. T.

Lensvelt-Mulders

Finkenauer

Stok

F. M.

Baumeister

R. F.

(2012). Taking stock of self-control a meta-analysis of how trait self-control relates to a wide range of behaviors. Personality and Social Psychology Review, 16, 76-99.

16.

Desmond

S. A.

Soper

S. E.

Kraus

(2011). Religiosity, peers, and delinquency: Does religiosity reduce the effect of peers on delinquency? Sociological Spectrum: Mid-South Sociological Association, 31, 665-694.

17.

DiIulio

J. J.

(2009). More religion, less crime? Science, felonies, and the three faith factors. Annual Review of Law and Social Science, 5, 115-133.

18.

Ellison

C. G.

Trinitapoli

J. A.

Anderson

K. L.

Johnson

B. R.

(2007). Religion and domestic violence: An examination of variations by race and ethnicity. Violence Against Women, 13, 1094-1112.

19.

Fowler

P. J.

Ahmed

S. R.

Tompsett

C. J.

Jozefowicz-Simbeni

D. M. H.

Toro

P. A.

(2008). Community violence and externalizing problems: Moderating effects of race and religiosity in emerging adulthood. Journal of Community Psychology, 36, 835-850.

20.

Frazier

P. A.

Tix

A. P.

Barron

K. E.

(2004). Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology, 51, 115-134.

21.

Good

Willoughby

(2006). The role of spirituality versus religiosity and adolescent psychosocial adjustment. Journal of Youth and Adolescence, 35, 41-55.

22.

Gottfredson

M. R.

Hirschi

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

23.

Hagger

M. S.

Wood

Stiff

Chatzisarantis

N. L.

(2010). Ego depletion and the strength model of self-control: A meta-analysis. Psychological Bulletin, 136, 495.

24.

Hilbe

J. M.

(2011). Negative binomial regression (2nd ed.). New York, NY: Cambridge University Press.

25.

Hittner

J. B.

Swickert

(2006). Sensation seeking and alcohol use: A meta-analytic review. Addictive Behaviors, 31, 1383-1401.

26.

Holden

R. R.

(2010). Social desirability. In Corsini

R. J.

Craighead

W. E.

(Eds.), Encyclopedia of psychology (pp. 1628-1629). Hoboken, NJ: John Wiley.

27.

Jang

S. J.

Johnson

B. R.

(2003). Strain, negative emotions, and deviant coping among African Americans: A test of general strain theory. Journal of Quantitative Criminology, 19, 79-105.

28.

Jang

S. J.

Johnson

B. R.

(2005). Gender, religiosity, and reactions to strain among African Americans. Sociological Quarterly, 46, 323-357.

29.

Jang

S. J.

Johnson

B. R.

(2011). The effects of childhood exposure to drug users and religion on drug use in adolescence and young adulthood. Youth & Society, 43, 1220-1245.

30.

Johnson

B. R.

(2011). More God, less crime: Why religion matters and how it could matter more. Conshohocken, PA: Templeton Press.

31.

Johnson

B. R.

Jang

S. J.

(2010). Religion and crime: Assessing the role of the faith factor. In Rosenfeld

Quinet

Garcia

(Eds.), Contemporary issues in criminological theory and research: The role of social institutions (pp. 117-150). Belmont, CA: Wadsworth.

32.

Johnson

B. R.

Jang

S. J.

De Li

Larson

(2000). The “invisible institution” and black youth crime: The church as an agency of local social control. Journal of Youth and Adolescence, 29, 479-498.

33.

Johnson

B. R.

Jang

S. J.

Larson

D. B.

De Li

(2001). Does adolescent religious commitment matter? A reexamination of the effects of religiosity on delinquency. Journal of Research in Crime & Delinquency, 38, 22-44.

34.

Johnson

B. R.

Larson

D. B.

De Li

Jang

S. J.

(2000). Escaping from the crime of inner cities: Church attendance and religious salience among disadvantaged youth. Justice Quarterly, 17, 377-391.

35.

Johnson

M. C.

Morris

R. G.

(2008). The moderating effects of religiosity on the relationship between stressful life events and delinquent behavior. Journal of Criminal Justice, 36, 486-493.

36.

Kerestes

Youniss

Metz

(2004). Longitudinal patterns of religious perspective and civic integration. Applied Developmental Science, 8, 39-46.

37.

Kim-Spoon

Longo

G. S.

McCullough

M. E.

(2012). Parent-adolescent relationship quality as a moderator for the influences of parents’ religiousness on adolescents’ religiousness and adjustment. Journal of Youth and Adolescence. Advance online publication. doi:10.1007/s10964-012-9796-1

38.

Koenig

H. G.

King

Carson

V. B.

(2012). Handbook of religion and health (2nd ed.). New York, NY: Oxford University Press.

39.

Koenig

L. B.

McGue

Krueger

R. F.

Bouchard

T. J.

(2005). Genetic and environmental influences on religiousness: Findings for retrospective and current religiousness ratings. Journal of Personality, 73, 471-488.

40.

Koster

Goudriaan

van der Schans

(2009). Shame and punishment: An international comparative study on the effects of religious affiliation and religiosity on attitudes to offending. European Journal of Criminology, 6, 481-495.

41.

Loeber

Farrington

D. P.

(Eds.). (1998). Serious and violent juvenile offenders. Risk factors and successful interventions. London, England: SAGE.

42.

Long

J. S.

Freese

(2006). Regression models for categorical dependent variables using Stata (2nd ed.). College Station, TX: StataCorp Press.

43.

McCullough

M. E.

Willoughby

B. L.

(2009). Religion, self-regulation, and self-control: Associations, explanations, and implications. Psychological Bulletin, 135, 69-93.

44.

Milot

A. S.

Ludden

A. B.

(2009). The effects of religion and gender on well-being, substance use, and academic engagement among rural adolescents. Youth & Society, 40, 403-425.

45.

Petts

R. J.

(2009). Family and religious characteristics’ influence on delinquency trajectories from adolescence to young adulthood. American Sociological Review, 74, 465-483.

46.

Pickering

L. E.

Vazsonyi

A. T.

(2010). Does family process mediate the effect of religiosity on adolescent deviance? Revisiting the notion of spuriousness. Criminal Justice and Behavior, 37, 97-118.

47.

Piquero

A. R.

Jennings

W. G.

Farrington

D. P.

(2010). On the malleability of self-control: Theoretical and policy implications regarding a general theory of crime. Justice Quarterly, 27, 803-834.

48.

Ramchand

MacDonald

J. M.

Haviland

Morral

A. R.

(2009). A developmental approach for measuring the severity of crimes. Journal of Quantitative Criminology, 25, 129-153.

49.

Rew

Wong

Y. J.

(2006). A systematic review of associations among religiosity/spirituality and adolescent health attitudes and behaviors. Journal of Adolescent Health, 38, 433-442.

50.

Ryan

J. P.

Testa

M. F.

Zhai

(2008). African-American males in foster care and the risk of delinquency: The value of social bonds and permanence. Child Welfare, 87, 115-140.

51.

Salas-Wright

C. P.

Olate

Vaughn

M. G.

(2013). The protective effects of religious coping and spirituality on delinquency: Results among high-risk and gang-involved Salvadoran youth. Criminal Justice and Behavior, 40, 988-1008.

52.

Salas-Wright

C. P.

Olate

Vaughn

M. G.

Tran

T. V.

(2013). Direct and mediated associations between religious coping, spirituality, and youth violence in El Salvador. Pan American Journal of Public Health, 34, 183-189.

53.

Salas-Wright

C. P.

Vaughn

M. G.

Hodge

D. R.

Perron

B. E.

(2012). Religiosity profiles of American youth in relation to substance use, violence, and delinquency. Journal of Youth and Adolescence, 41, 1560-1575.

54.

Salas-Wright

C. P.

Vaughn

M. G.

Maynard

B. R.

(2013). Religiosity and violence among adolescents in the United States: Findings from the National Survey on Drug Use and Health 2006-2010. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260513506279

55.

Salas-Wright

C. P.

Vaughn

M. G.

Maynard

B. R.

Clark

Snyder

(under review). Public or private religiosity: Which one is protective for adolescent substance use and by what pathways?

56.

Schreck

C. J.

Burek

M. W.

Clark-Miller

(2007). He sends rain upon the wicked: A panel study of the influence of religiosity on violent victimization. Journal of Interpersonal Violence, 22, 872-893.

57.

Smith

(2003). Theorizing religious effects among American adolescents. Journal for the Scientific Study of Religion, 42, 17-30.

58.

Smith

Denton

M. L.

(2005). Soul-searching: The religious and spiritual lives of American teenagers. New York, NY: Oxford University Press.

59.

Smith

Faris

(2002). Religion and American adolescent delinquency, risk behaviors, and constructive social activities. Chapel Hill, NC: National Study of Youth and Religion.

60.

Smith

Faris

Denton

M. L.

Regnerus

(2003). Mapping American adolescent subjective religiosity and attitudes of alienation toward religion: A research report. Sociology of Religion, 64, 111-133.

61.

Smith

Snell

(2009). Souls in transition: The religious and spiritual lives of emerging adults. New York, NY: Oxford University Press.

62.

StataCorp. (2013). Stata Statistical Software: Release 13. College Station, TX: Author.

63.

Strayhorn

J. M.

(2002). Self-control: Theory and research. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 7-16.

64.

Substance Abuse and Mental Health Services Administration (SAHMSA). (2011). Results from the 2010 National Survey on Drug Use and Health: Summary of national findings. Rockville, MD: Author.

65.

Ulmer

J. T.

Desmond

Jang

S. J.

Johnson

B. R.

(2010). Teenage religiosity and changes in marijuana use during the transition to adulthood. Interdisciplinary Journal of Research on Religion, 6, 1-19.

66.

Ulmer

J. T.

Desmond

Jang

S. J.

Johnson

B. R.

(2012). Religiosity and dynamics of marijuana use: Initiation, persistence, and desistence. Deviant Behavior, 33, 448-468.

67.

Vohs

K. D.

Baumeister

R. F.

(Eds.). (2011). Handbook of self-regulation: Research, theory, and applications. New York, NY: Guilford.