The Influence of Religion on the Criminal Behavior of Emerging Adults

Recent generations of young adults are experiencing a new life course stage: emerging adulthood. During this ‘new’ stage of the life course, traditional social bonds and turning points may not be present, may be delayed, or may not operate in the same manner as they have for prior generations. One such bond, religion, is examined here. Focusing on the United States, emerging adulthood is investigated as a distinct stage of the life course. The criminality of emerging adults is presented, a theoretical examination of the relationship between religion and crime is provided, the role of religion in emerging adults’ lives is explored, research on the role of religion’s influence on criminal offending is presented, and theoretical and policy implications are offered.


Introduction
The role of religion as a key component of the lives of many American youth has been established by scholars, such as Salas-Wright et al. (2015) and , across the social science literature. Previous studies have identified several beneficial outcomes related to the role of religion in youth populations (Koenig et al. 2012). For example,  found that young people who are religious are less likely to take part in substance use or abuse. Other studies, such as Salas-Wright et al. (2014), revealed that youth who are more religious engage in less violence, and Baier and Wright (2001) showed youth who engage in higher levels of religious participation exhibit fewer antisocial behaviors, like crime. Much of the literature exploring the role of religion as a factor influencing the antisocial behaviors of youth populations has largely focused on adolescent populations. For example, Johnson et al. (2000) used data from the National Youth Survey's fifth wave, in which the sampled participants were between the ages of 11 and 17 years old, to study how religious involvement mediates the effect of neighborhood disorder on youth crime. However, an area that has not yet been sufficiently explored in the study of the antisocial behaviors of young people is the relationship between religion and the criminal behavior of emerging adults.
The path to adulthood has radically changed in the United States, as it has in many other high-income nations since the 1960s (Arnett 2015;Cote 2000;. Scholars have found that the period between adolescence and adulthood has been extended with many traditional markers of adulthood, like marriage and completion of higher education, being postponed, leading to delayed transition to adulthood (Arnett 2015;Cote 2000;. This prolonged period between adolescence and adulthood has been identified by Arnett as emerging adulthood (Arnett 1998). Originally conceptualized by Arnett to occur between 18 and 25 years of age, the age range has since been expanded out to the end of the twenties (Arnett 2015). Others, such as Salvatore, Taniguchi, and Welsh , have argued it may expand to older ages as well. Emerging adulthood has been identified as a time of exploration and experimentation for young people which may manifest in a variety of areas ).
Since being introduced by Arnett in the 1990s, emerging adulthood has been an area examined across the social science literature, including sociology (e.g., Salas-Wright et al. 2015), psychology (e.g., Arnett 2000), social work (Haffejee et al. 2013) and, in particular, criminology (e.g., Jang and Rhodes 2012;Marcus 2009;Piquero et al. 2002). Studies have found high rates of risky and dangerous behaviors, such as binge drinking and unsafe sex, in emerging adult populations . For example, studies by (Arnett 1998(Arnett , 2005, Chassin et al. (2002), and White et al. (2005) found that risky driving and substance use are common in samples of emerging adults. Prior research has also found that emerging adulthood is a fertile stage for criminal offending and substance use (Haffejee et al. 2013;Piquero et al. 2002). Many emerging adults offend in a similar manner to adolescents, committing lowlevel, non-violent crimes Salvatore 2017).
Many studies have focused on the role of religion as a factor that may influence risky behaviors in emerging adults (e.g., Salas-Wright et al. 2015) or have utilized religion as a variable to predict a decrease in offending during emerging adulthood. For example, in their 2012 study, Salvatore and Taniguchi examined the role of turning points and social bonds on offending during emerging adulthood. Data from the National Longitudinal Study of Adolescent Health revealed that turning points, like marriage and parenthood, as well as social bonds, like religious participation, reduced offending during emerging adulthood.
In this paper, we examine the role that religion plays on the criminal behavior of emerging adults in the United States of America. Since a large portion of the research in the area of emerging adulthood has been conducted in the context of the United States (e.g., Arnett 2015) or has used samples from the United States (e.g., Marcus 2009;Piquero et al. 2002;, it was deemed fitting to keep the discussion of this article within that context. We will begin by explaining what emerging adulthood is and how it evolved as a distinct stage of the life course. Next, we will explore prior studies examining criminal offending during emerging adulthood. Third, we will present the theoretical role of religion on influencing antisocial behaviors, like crime. Fourth, we will briefly examine the religious beliefs and practice of emerging adults. Fifth, we will summarize the existing research dealing with religion, crime, and emerging adulthood. We will conclude with implications for theory and policy as well as a look at potential areas of future scholarship.

What is Emerging Adulthood?
In order to understand the role that religion plays in the offending of young adults in the modern era, we need first to define emerging adulthood. Originally presented by Arnett in 1994, emerging adulthood is a re-conceptualization of the lives of young people aged between 18 and 25, later expanded into the late 20s (Arnett 2015). The concept of emerging adulthood signifies that traditional paths to adulthood, like completing higher education, getting married, becoming a parent, and so forth are now postponed for many young adults (Arnett 2015). Due to these delays, Arnett argued that instead of just a transition from adolescence to adulthood, young adults now experience a completely new stage of the life course, which Arnett termed emerging adulthood (Arnett 2015).
According to Arnett, emerging adulthood is grounded in cultural changes that have occurred since the 1960s, which have resulted in changes in behaviors at a societal and an individual level (Arnett 1998(Arnett , 2000(Arnett , 2005(Arnett , 2015. In his work, Arnett identified four key areas of change which fostered the development of emerging adulthood as a unique stage of the life course. First, was the shift from an economy rooted in manufacturing that generally did not require advanced education, to a more information-based culture which requires higher levels of education. As such, youth, who in prior generations could have lived a middle class lifestyle with just a high school education, were now expected to continue their education beyond high school, and seek employment in areas more geared toward service, such as health care and education. Next, was increased freedom for women in society, in particular in the areas of education and employment. These changes allowed more women to delay marriage and parenting, and pursue careers. Third, there we increased sexual freedom and tolerance of behaviors such as premarital sex. These changes led to the normalization of youth of having sex before marriage. The final area of change was the youth movement which demonized the maturation of adulthood and lionized the unfettered freedom of adolescence. As such, many, who in prior generations would have viewed themselves as young adults, now saw themselves in a state of extended adolescence (Arnett 2005(Arnett , 2015Tanner and Arnett 2009;). Today's emerging adults speak of "adulting" as a foreign practice, complete with unfulfilling errands, "scary" encounters with older generations and dreaded responsibilities (Lierz 2017).
Even the popular press is examining emerging adulthood, as behaviors typically reflected in younger adolescents are now seen more in those in emerging adulthood. As Time Magazine noted, "Today, the researchers say, 18-year-olds act more like 15-year-olds from previous decades. That was true across all demographic groups in the study" (MacMillan 2017). This not only means that emerging adulthood continues to be an important phenomenon, but also that criminal behavior may be shifting from the teen years to the twenties and perhaps even the thirties. This is troubling given how criminal records are treated for minors versus adults according to the law. While minors can get their records sealed or expunged, adults typically cannot-and this will surely hurt the employment prospects, financial stability and family lives of future emerging adults.

Emerging Adults and Crime
One of the most important findings in criminology is the peak years of offending (Ulmer and Steffensmeier 2014). Quetelet (1984) observed that for those involved in crime, offending tends to peak in late adolescence and early adulthood, and then declines as people age. In the modern era, the Uniform Crime Report found that the peak crime years in all crimes reported are those younger than 25 (Ulmer and Steffensmeier 2014). The peak years of offending, as noted by criminologists, as well as national crime data, encompass not only adolescence, but emerging adulthood as well. Across the social science literature, a consistent finding is that emerging adults are prone to risky and dangerous behaviors, including criminal offending and illegal substance use . Prominent scholars in the area of life course criminology have argued that emerging adulthood is a stage of the life course ripe for criminal offending (Moffitt et al. 2002). For example, in a study using a sample of male parolees released from the California Youth Authority, Piquero et al. (2002) found that arrest rates for non-violent and violent offenders reached a high point in the early 20s, during the emerging adulthood stage of the life course. These findings support the notion that emerging adulthood is a stage of the life course where criminality may be expressed. Data from national studies of crime further support these conclusions. For example, data from the 2016 Uniform Crime Report (UCR) found that people aged 18 through 25 had higher rates of arrest for violent and non-violent crimes relative to other groups in the population. For example, in 2016, there were 7860 individuals aged 15 arrested for violent crimes, compared to 14,040 aged 20 (Uniform Crime Report UCR). We found a similar pattern with non-violent arrests as well, with those aged 15 having 29,385 arrests for property crimes, and those aged 20 having 37,463 arrests for property crimes (Uniform Crime Report UCR).
The potential for emerging adults to engage in criminal behavior is perhaps most reflected in studies examining illegal drug use. As illegal drug usage is prominently examined in the literature dealing with emerging adulthood, this section begins with a look at illegal drug usage in emerging adults, followed by a discussion of studies examining offender typologies and types of offending prominent in emerging adults, and finally, criminological theories that can be applied to explain crime in emerging adults.
The emerging adulthood stage of the life course is ripe for experimentation and identity exploration through substance use, including the use of illegal drugs . Walters (2014) stated that those in late adolescence and early adulthood (the emerging adulthood age range) have some of the highest rates of illegal substance use and abuse, compared with younger and older segments of the population. For example, using a nationwide sample, the Monitoring the Future Study, which followed a sample of high school graduates each year until adulthood, thus providing data through emerging adulthood, found that emerging adults have the highest prevalence of past-30-day illicit drug use, ranging from 22 to 28% in those 18 to 24 years of age (Schulenberg et al. 2016). The use of illicit drugs in 2016 increased in the senior year of high school (age 18) from 24 to 28% percent in those aged 21 to 22 (Schulenberg et al. 2016), reflecting the high rates of substance use as discussed by Arnett (2005). Another national survey, the National Survey on Drug Use and Health, a large state-based study which contains both historical and current data about the prevalence of drug use across emerging adulthood has yielded similar findings, with those in the 18 to 25 year age range having significantly increased levels of illegal substance use relative to younger members of the population as well as older members of the population (Center for Behavioral Health Statistics and Quality 2017). For example, in 2015, there were 4346 respondents in the 12 to 17 year age group who stated they had used illicit drugs in the past year, compared to 13,102 of the respondents in the 18 to 25 year old age group (Center for Behavioral Health Statistics and Quality 2017).
Previous studies suggest that experimental drug use is highest in emerging adulthood, and then tends to decrease starting at around age 25 as individuals assume adult social roles and build the bonds and attachments that act to inhibit crime and deviance in adults (Bachman et al. 1997). As discussed above, emerging adulthood is often a period of newfound freedom from the controls of parents and teachers as youth graduate high school and leave their parents' homes. With this new freedom may come experimentation with substance use, or increased substance use, especially in those who have established a pattern of substance use in adolescence (Arnett 2000(Arnett , 2015Rohrbach et al. 2005).
Illicit drug use is particularly problematic for emerging adults given their high rates of substance abuse disorders, as nationwide, emerging adults are three times more likely to report illicit drug dependence and misuse compared to the general population (Jones et al. 2015). This is problematic as drug crime often correlates with other types of crime (Craddock et al. 1997) as well as public health concerns, such as sexually transmitted infections (Dembo et al. 2009). These potentially negative outcomes could follow an emerging adult for the rest of their life, impacting their health, relationships, and status in society, as many of these issues, such as having a criminal record, are stigmatized in society.
In addition to illicit drug use, studies have explored the other types of crime prevalent in emerging adults. For example, Salvatore, Taniguchi, and Welsh ) used data from the publicly accessible version of the National Longitudinal Study of Adolescent Health (Add Health) to examine the types of offenses emerging adults were committing.  created index variables to place offenses into two categories: (1) Life Course Persistent Offending (more serious, predatory offenses, such as burglary) and (2) Prolonged Adolescent Offending (less serious, non-predatory crimes, such as stealing items worth less than $50). The findings revealed that the bulk of emerging adults were engaging in less serious "Prolonged Adolescent" offenses; conversely, only a small percentage (around 6%) had committed the more predatory "Life Course Persistent" crimes  In another study using a sample of male parolees released from the California Youth Authority, Piquero et al. (2002) found that arrest rates for non-violent and violent offenders reached a high point in the early 20s, during the emerging adulthood stage of the life course. These findings largely mirror the work of Moffitt (1993) and Salvatore (2017) that suggests that emerging adults will engage in offending in similar patterns to adolescents, with most engaging in lower-level, non-predatory offenses, and a smaller percentage committing more serious offenses.
Other studies have examined specific types of crime more closely. For example, Loeber et al. (2008) used data from the Pittsburgh Youth Study (PYS) to examine violence and theft. The results of their study found that the annual rate of violent offending peaked at four offenses per year in the late teens and then dropped as the sample moved through emerging adulthood. The findings were similar regarding theft, with a peak of offending in the late teens, between 3.5 and 5.4 thefts annually, before dropping to about two thefts per year annually as the sample aged through emerging adulthood. This pattern reflects the relationship between age and crime, as presented by scholars like Gottfredson and Hirschi (1990), with crime peaking in the late teens and tapering off as people age. In another study, Marcus (2009) looked at self-reported violence, including armed robbery, gang fighting, using a weapon in a fight, pulling a knife on someone, or shooting and stabbing someone, using several of the Add Health data to cross sectionally analyze violence, including when the sample was in emerging adulthood. The presence of violence at wave 3 (during emerging adulthood) was predicted by having a history of violence as well as a lack of social bonds/attachments, such as marriage, and being at economic risk.
Several theories have been used to explain the prevalence of crime in emerging adult populations. In 2017, Salvatore presented the "emerging adult gap" thesis which argued that due to the delayed turning points and attenuated social bonds prevalent during emerging adulthood, many youths fall into an "emerging adulthood gap" similar to the "maturity" gap that Moffitt (1993) presented. According to Salvatore (2017), those caught in the "emerging adulthood gap" may be prone to committing offenses in patterns reflective of Moffitt (1993) developmental taxonomy of life course persistent offenders (a smaller, more predatory group of offenders who start offending earlier in the life course) and adolescent limited offenders (a larger group of youth who typically engage in less predatory crimes such as underage drinking and being loud and rowdy), who typically start offending later and desist sooner than the aforementioned life course persistent offenders.
Social control theory, as presented by Hirschi (1969), could be applied to the study of crime in emerging adults. Hirschi's theory argues that more strongly an individual is bonded to society, the less likely they are to commit crime. According to Hirschi (1969), there are four elements in a social bond: (1) attachment, which refers to the emotional bond one has with prosocial individuals like teachers, parents, and friends. Individuals would not want to engage in crime, for to do so would risk this emotional bond; (2) involvement-referring to an individual's level of engagement with prosocial activities, like school, sports, and religious activities. If an individual is engaged in these prosocial pursuits, they will not have time to engage in antisocial behavior like crime; (3) belief-the notion that if an individual believes in conventional society (e.g., laws), they will be less likely to engage in behaviors that go against convention; and (4) commitment, the idea that an individual will fear the consequences of committing crime that could involve punishment since, in such a case, the individual could lose social bonds and face sanctions. Hirschi (1969) theory can be used to explain crime in emerging adults by incorporating the changes and experimental nature of this stage of the life course. Emerging adults tend to have different relationships with their parents, often moving out of the home and no longer being subjected to their control, and teachers, who (if they are attending college) do not have the power over them as they did in high school. Professors in college are typically occupied with other roles in their positions, such as research and services, and as such, they typically do not operate as "police" in the classroom who might call parents in, for instance, when students do not attend class or miss assignments.
Another theory that could be applied to the criminality of emerging adults is strain theory. Agnew (2001) general strain theory, argues that there are three types of strain: (1) strain arising from the difference between an individual's expected goals and actual results; (2) strain arising from the removal of positively valued stimuli (like a romantic relationship or job); and (3) strain arising from the introduction of negative stimuli. If an individual experiences any of these types of strain, they may experience negative emotions like anger and rage, which could lead to various types of behavior, depending upon the level of support the individual has in their social network. Agnew posited that the emotion of anger could be correlated with crime, especially violence crime. Criminal behavior could take the shape of attacking a source of strain (such as fighting someone who insults them) or coping with negative emotions/thoughts through illegal drug use. Agnew (2001) theory could be used to explain the high rates of crimes found in emerging adults. For example, as emerging adulthood has been identified as a period of change and instability (Arnett 2005), emerging adults could react to the negative stimuli of uncertainty in areas such as relationships, employment, and housing using illegal drugs to self medicate and manage stress.
There have been several studies which have used the abovementioned theories to study their role in understanding crime in emerging adults. Using Add Health data, Jang and Rhodes (2012) explored the role of strain on crime and drug use for emerging adults. The unstable and exploratory nature of emerging adulthood, fraught with changes in residence, romantic relationships and employment, provides ample forms of strain to examine. The results of Jang and Rhodes's study revealed that social bonds and self-control mediate the effect of strain on offending, supporting Arnett's theory and work in addition to studies that have found that social bonds (e.g., religion and education) can help to reduce offending in emerging adults. In another study,  used Add Health data to test the effects of social bonds and turning points, as theorized by Hirschi (1969) social control theory and Laub and Sampson (2003) age-graded theory of informal social control. The results of Salvatore and Taniguchi's study found that traditional turning points, like marriage and having children, as well as social bonds/controls, like religion, are effective in reducing crime in emerging adults.
The above studies explain and verify the criminality present in emerging adult populations. As discussed, emerging adults typically engage in offending due to negative stimuli or the lack of social bonds and controls that were prevalent in prior generations and have prevented many young adults from engaging in crime. Due to being occupied with spouses, children, full-time employment, and military service, as well as religious, community, and familial attachments, young adults of the past did not experience emerging adulthood, and therefore, were less apt to participate in offending . As this area of research continues to grow, more trends and patterns of emerging adult offending will surely be revealed.

Theoretical Role of Religion during Emerging Adulthood
Sociological inquiry regarding the effect of religion on crime began in the 20th century and continues to the present day (Bainbridge 1989;Lombroso 1911;Schur 1969). Organized religion has traditionally served as a method for bonding individuals with their families, communities, and societies, conveying societal goals and instilling morals in young and old alike. The religious bond not only holds families and communities together through common beliefs and rituals but provides a moral compass which can act as an internal social control, preventing individuals from engaging in antisocial behaviors, like crime. The value of religion is reflected in a response from an interviewee who explained to one of the authors, "Why do people send their kids to madrasa (religious school)? For the same reasons you might send your children to religious school: to teach them morals, values, discipline, and about their culture" (Mahmoud 2013).
Religious institutions provide youth with a normative set of guidelines through the creation of bonds to the religious group (Smith 2003). The effectiveness of religion as a social control is demonstrated by studies that have found that religion can act as an effective social control for adolescents and young adults (Johnson et al. 2000;Petts 2009;Salas-Wright et al. 2015;. We know the potential value of religion as a social control for adolescents and emerging adults, but through what theoretical mechanism does it work? There are several different theories of religion which may explain why it can act as a social control. Most of these theories are grounded in the notion of religious involvement, referring to the level to which a person is engaged in a religious organization and therefore connected to a social network in the context of that organization (Johnson et al. 2000). Religious involvement may encourage the development of social networks which decrease antisocial behaviors such as crime (Gottfredson and Hirschi 1990). As Krohn (1986) noted, social networks are key to providing not only social, but emotional support, which may help individuals resist or desist from engaging in crime. Through interactions with other members of the religious group and the controls established by being engaged in the religious organization, youth may be aided in avoiding antisocial behaviors which could damage their lives, instead seeking lives of greater meaning. As such, involvement in a religious community may prevent an individual from engaging in antisocial behaviors, like delinquency and substance use (Petts 2009). The idea of religious involvement as a factor that encourages social bonds and controls which prevent crime fits nicely into Hirschi (1969) social control theory where social institutions like the family and education act to build individual attachment, commitment and involvement with society, thereby acting to prevent crime-as the individual would not want to risk endangering these attachments. Hirschi (1969) social control theory is reflected in the work of Sampson and Laub (1993) age-graded life course theory of crime which posits that stronger attachments to social institutions (like religion) are connected to lower levels of delinquency and crime. As such, we would expect an individual with higher levels of engagement with religious institutions (e.g., going to religious services more often, participating in religious oriented community events) to be less likely to commit crime.
The next major theory recognized for explaining the deterrence effect of religion is the so-called hellfire hypothesis. The hellfire hypothesis, most widely known from its presentation by Hirschi and Stark (1969), as well as a series of follow up studies, including Burkett and White (1974) and Stark (1996), proposes that religion deters individual level crime by using the threat of supernatural punishment and promotes normative behavior through the promise of a supernatural reward. For example, using the Christian Faith, if an individual engages in a crime such as murder, thus breaking one of the Ten Commandments, they will be subject to punishment in hell. On the other hand, if an individual does not commit crimes, they will be rewarded with an eternity in the paradise of heaven.
Rational choice theory is used to explain how crime connects an individual's self-imposed punishment with their level of religious commitment (Grasmick et al. 1991). According to Grasmick et al. (1991), people who strongly identify as religious have a greater likelihood of experiencing shame from committing deviant acts, and individuals saliently participating in religion-based social networks have a greater likelihood of experiencing embarrassment as a result of deviant acts (p. 253). As such, these processes would influence religious individuals to avoid committing crime by increasing the perceived chances of, and severity of, informal punishments (Baier and Wright 2001).
Another pair of theories explaining the role of religion in offending are reference group theory and differential association, both of which incorporate the role of the social group(s) on individual behavior. Reference group theory argues that individuals existing within reference groups in which they have backgrounds and beliefs that strongly resemble each other, "decisively" shape the behavior and attitudes of one another (Bock et al. 1987, pp. 91-92). Since each person juxtaposes and subsequently controls their behavior in a way that is reflective of the behavior and attitudes of the other members in their reference groups, an increase in morality within these groups will increase the level of morality for all members of the groups. Therefore, if reference groups become more centered on religion, religion will act to deter crime through the adoption and increased level of group level morality (Bock et al. 1987).
Using a differential association perspective, religion prevents crime by the processes of socialization and social selection (Burkett 1993). Through socialization, religious peer influence can change an individual's level of religious commitment though positive reinforcement, thereby deterring criminal acts (Baier and Wright 2001;Burkett and Warren 1987). In regard to social selection, religion shapes peer selection since individuals with a stronger commitment to religion will select peers with a similar level of (conventional religious) beliefs (Burkett and Warren 1987).
To take a somewhat different approach, the sociobiological perspective uses arousal theory to explain criminal behavior as being the result of an individual's need for neurological stimulation, with those prone to crime having a greater need for neural arousal than those who are not crime prone (Eysenck 1964). In other words, for individuals who need greater levels of neurological stimulation, normative social activities, such as participating in religious activities, will not provide enough stimulation; as such, they will seek out more stimulating activities, such as crime.
Based upon the abovementioned theories, we would expect religion to operate as an effective social control for emerging adults. Despite the somewhat different social circumstances, such as changes in the economy and nature of employment to which emerging adults are subjected, emerging adults are still connected to each other, their family, and their community through religious institutions. Being part of religious groups and institutions will shape the normative morality of emerging adults (as it has with prior generations), provide prosocial activities and role models, and help to instill internal social controls, preventing the individual from engaging in criminal behaviors.
As we will see in the next section, previous studies have suggested that recent cohorts of emerging adults may be less engaged with religion. It should be noted a decline in religion does not need to come with a decline in social or other values. Individuals can adopt their own moral systems, take to humanistic ideologies or philosophies, or simply indulge in altruistic behavior due to empathic impulses (Pinker 2012). In Homo Deus, Yuval Harari defines religion broadly, as any ideology that believes in a superhuman order. He thus includes Communism and Liberalism as "religions" (Harari 2017). However, most who speak of "religion" are referring to traditional, institutionalized religions, such as Islam, Buddhism, Hinduism, Judaism, Zoroastrianism, and Christianity. These institutionalized religions serve as "on the ground" moralizers that link the individual to the culture and the state. This is why Jean-Jacques Rousseau, among others, proposed condoning a civil religion that would tie the populace together (Rousseau 1762).

Emerging Adults and Religious Beliefs
As noted above, religion is an important social control that can inhibit deviance and crime (Laub and Sampson 2001). Religion helps to guide and shape our view of the world, providing values that direct our behavior. In addition, developing religious beliefs may be an essential part of our identities (Arnett 2015, p. 211). Prior studies have found that youth with higher levels of religious involvement are less likely to engage in substance use and abuse ) and have lower rates of offending Salas-Wright et al. 2015). Despite this, scholars like Arnett (1998) have found that emerging adults in college are less likely to attend religious services on a regular basis. However, even though lower levels of religious participation have been reported in college aged students, many still state that religion is important to them (Arnett and Jensen 2002). To put it another way, some evidence suggests that emerging adults may be less likely to participate in religious services, but they still report that it is an important factor in their lives. The role of religion in the lives of emerging adults is an important relationship for us to examine as there is a growing body of literature examining the relationship between variance in religious expression and high risk and antisocial behaviors in populations of emerging adults (Salas-Wright et al. 2015). The bulk of studies exploring the relationship between religion and antisocial behaviors, like drug use and crime, have centered on the belief that religiosity is a homogeneous factor (Salas-Wright et al. 2015, p. 67). Studies have tended to focus on the protective role of aspects of religiosity, like attendance at religious services, participation in religious groups or activities, or created scaled measures of items pertaining to religiosity (Salas-Wright et al. 2015, p. 67). Several studies, such as Salvatore and Taniguchi (2012), have taken one or both of these approaches, which certainly can measure the influence of religion on the aforementioned antisocial behaviors. However, studies in recent years have argued that the different aspects of religiosity may not always act independently; instead, they may reflect a distinct, consistent religiosity profile (Pearce and Denton 2011;Smith and Denton 2005;Smith and Snell 2009).
So, if religion is important to emerging adults, and religion is also a key social control, with prior studies finding higher levels of religious participation being related to decreased levels of offending and substance use, what does the literature say about the relationship between the two? As emerging adulthood is a relatively new area of research, there have been few expansive looks at the role of religion during emerging adulthood. However, there is an expansive body of literature exploring the religiosity of adolescents which provides us with a starting point and context to understand religiosity in youth populations. Below, we will briefly examine adolescent religiosity profiles, and then turn our attention to the research examining religiosity in emerging adults.
To date, one of the most comprehensive sources for data dealing with adolescents and emerging adults and religions is the National Study of Youth and Religion (NSYR). The NSYR started in 2003 at the University of Notre Dame. Using a National Sample of teenagers who were followed through emerging adulthood, the study provides a detailed look at the religious practices and beliefs of a recent cohort of emerging adults. In 2008, Denton, Pearce, and Smith used data from two time points (wave 1, aged 13-17, and wave 2, aged 16 to 21 years) to report on several aspects of religious practices and beliefs from emerging adults, including religious affiliation, beliefs, practices, religious choice, and personal religiosity and spirituality. Previous studies have examined religiosity profiles and their relationships with behavior in adolescents in the United States and other high income nations (Pearce and Denton 2011;Smith and Denton 2005) using NSYR data. For example, Smith and Denton (2005) constructed the "Ideal Types." These "Ideal Types" include a group identified by Smith and Denton (2005) as "the devoted", describing adolescents who go religious services once a week or more, state that their faith is very or extremely important, express a belief in God, and state they pray a few times or more per week. This 'devoted' type was 8% of the sample. There were three additional "Ideal Types" identified by the study: (1) "the regulars" (27%); (2) "the sporadic" (17%); and (3) "the disengaged" (12%). These groups were then used to study the relationship between being in one of these classes and several life outcomes, including antisocial behaviors, like substance use. Looking at the results from an adolescent-based portion of the NSYR gives us some context regarding the level of religious engagement in youth populations, as well as how the level of religious engagement may influence antisocial outcomes. As Hirschi (1969) social control theory suggests, being in a more engaged religious group (like "the devoted") builds stronger (prosocial) attachments to the community and society, thereby inhibiting delinquency in adolescent populations. We focus specifically on the role of religiosity on antisocial outcomes in the next section.
To better understand the role that religion plays in the lives of emerging adults, we present some of the key findings of the NSYR to help better understand how the different dimensions of religiosity are expressed in a population of emerging adults, followed by a brief discussion of a follow up study by Smith and Snell (2009) using the same "Ideal Types" as Smith and Denton (2005). The NSYR surveyed a scientific sample of Americans to see if they belonged to a variety of religions, including Catholic, Jewish, Muslim, Protestant, no religion, or other religion. Denton et al. (2008) findings revealed that religious affiliations were similar between wave 1 and wave 2. This finding may suggest that the exploratory nature of emerging adulthood does not influence the formal religious affiliation one has during this stage of the life course. On the other hand, Arnett (2015) conducted interviews with several subjects who stated that while they did identify with a religion, they did not necessarily believe in it, doubted the beliefs, or just did not participate in religious services. What we may see is that even though religious affiliation is consistent during emerging adulthood, it does not influence what an individual believes, or how they practice their beliefs, during emerging adulthood . It is worth noting that other studies, such as the 2012 Pew Foundation Survey, have found an increase in the number of 18 to 29 year olds who identify with religious affiliation (Hackett and Grimm 2012). This may suggest that emerging adults are less committed to the religious traditions they were raised with; however, it is possible that there may be a return to these affiliations as emerging adults age.
Turning our attention to Denton et al. (2008) findings regarding religious beliefs, the NSYR found minor changes in religious beliefs between waves one and two. For example, in the first wave, 13% of respondents reported that they did not know or were unsure if they believed in God, and in wave 2, the percentage rose to 18% (Denton et al. 2008, p. 7). Another question dealt with perceptions of God, ranging from seeing God as a personal being that influenced their lives, to a creator who is no longer active in the world, to a cosmic force. Results from wave 1 to wave 2 found a decrease from 67 to 63% of those who viewed God on a more personal level. This is a small but notable decrease, perhaps reflecting the uncertain nature of emerging adulthood. Next, those who viewed God as an uninvolved creator decreased by 1% between the two waves, and there was an increase of 2% in those who viewed God as a cosmic force. Denton et al. (2008) findings seem to suggest that while religion still matters to emerging adults, religiosity does decrease during this life stage.
The next area examined in the NSYR was public religious practice, which included attendance at religious services and at religious education classes, as well as youth group involvement. The results of the NSYR found that between waves 1 and 2, there was a marked decrease of 15% in the percentage of youth who attending religious services weekly (Denton et al. 2008, p. 14). This finding could be explained by the changes experienced in adolescence (in wave 1) and emerging adulthood (wave 2).
During adolescence, young people are typically subjected to household rules set by parents, one of which may be attendance at religious services. During emerging adulthood, youth may move into their own apartment, live with roommates, or be away at college, and as such, their parents' rules regarding participation in religious services may no longer influence their behaviors. Another factor offering a possible explanation for the drop in weekly attendance found by the NSYR was the level of individualism during emerging adulthood, which Arnett noted may make emerging adults more suspicious of organized religions (Arnett 2015). It is also possible that emerging adults may be living busier lives relative to their high school years, being occupied with work, school and day to day activities during the week, and their social lives on the weekends. As a result, they simply may not have the free time to go to religious services (Arnett 2015, p. 218).
The follow up study using Smith and Denton (2005) "Ideal Types" by Smith and Snell (2009) used data from the NSYR to apply the same "Ideal Types" employed to create religious profiles to the sample as adolescents, during emerging adulthood (when the sample was between the ages of 18 and 23). Reflective of the abovementioned findings from the NSYR, Smith and Snell found that there were changes in the groups, with the devoted class dropping from 8% during adolescence, to 5% during emerging adulthood. Not surprisingly (given the exploratory nature of emerging adulthood), the disengagement class more than doubled in size, from 12% in adolescence to 25% during emerging adulthood. There was little change in the "sporadic" type, with a 1% increase from adolescence (17%) to emerging adulthood (18%). There was a more marked decrease in the "regulars" class, dropping from 27% in adolescence to 15% in emerging adulthood. The findings of Smith and Snell (2009) follow up study are reflective of what is expected given the differences described above in the level of religious participation and religious experimentation. Interestingly, Smith and Snell found that those in the "devoted" and "regular" groups were less likely to have antisocial outcomes like binge drinking, marijuana use, and fighting, relative to those in the groups that were less religious. These findings suggest that, like during adolescence, religion may 'work' as a social control for emerging adults, albeit, to a lesser degree, being reflective of the decreased level of religious engagement in emerging adults. In our next section, we will examine the findings of studies that have used individual elements of religion, such as attendance at religious services, as well as those focused on religious profiles.

How Religion Impacts Criminal Behavior for Emerging Adults
So far, we have explained what emerging adulthood is, the criminal offending of emerging adults, theoretical explanations on how religion may act as a controlling or bonding force to prevent crime, and the religiosity of emerging adults. How the factors come together and how religion impacts criminal behavior in emerging adults can be determined by combining the existing research. Since religion is a social bond that is eroding, or at least attenuating, among emerging adults, it should be expected that this reduction in religiosity will lead to a rise in criminal behavior of the prolonged adolescent variety. The sort of criminal behavior that emerging adults typically engage in-namely, low level offending such as drug use and vandalism-might be particularly susceptible to religious teachings that link individuals to their communities and denounce substance abuse. In order to investigate the influence of religion on the criminality of emerging adults, we will look at research examining religious engagement and criminal behavior, followed by a look at how religious engagement has influenced the delinquency of youth and criminality of emerging adults and finally, studies examining the influence of religious profiles on offending during emerging adulthood.
While the study of the influence of religion on emerging adults' criminality may be a relatively new area of inquiry, the study of religion's role on criminal behavior is not. Prior research has found ample evidence to support the belief that religious engagement acts as a protective factor or deterrent against criminal behaviors (Salas-Wright et al. 2014). For example, in their meta-analysis of 60 articles exploring the relationship between religiosity and crime, Baier and Wright (2001) found that public religious engagement and personal religious beliefs have a stable inverse (r = −0.13) deterrent effect on criminal behavior. Interestingly, the deterrent effect was even stronger (r = −0.24) when examining non-violent crimes, like drug crimes and other 'victimless' crimes (Baier and Wright 2001). More recently, another meta-analysis of 270 articles, published from 1944 through to 2010, studying the relationship between religiosity and crime, conducted by Johnson andJang in 2010 (Johnson andJang 2010), revealed that over 90% of the journal articles included had reported religiosity to be a protective factor against antisocial behaviors, like crime. Studies looking at multidimensional measures of religiosity have reported a deterrent effect of religion on crime in general (Johnson and Jang 2011;Pickering and Vazsonyi 2010. Other studies, such as Laub and Sampson (2003), have reported that religious engagement may be part of the institutional relationships that are key in the process of desistance from crime, though some research suggests that factors like race and ethnicity may influence the role of religiosity on influencing desistance (Stansfield 2017).
Research has yielded similar findings when focused on the two primary subcomponents of religiosity: religious service attendance and private religiosity (Salas-Wright et al. 2014). Attendance at religious services has been found to reduce the incidence of minor crimes, such as being loud and rowdy, and vandalism (Petts 2009;, drug use , felony theft and selling drugs (Johnson et al. 2000), domestic violence (Ellison and Anderson 2001) and involvement in the criminal justice system (Johnson 2008;Johnson et al. 2000;Ryan et al. 2008). In regard to private beliefs, typically defined as how an individual views the importance of their religious beliefs in their lives, a similar deterrent effect as attendance at religious services has been discovered (Salas-Wright et al. 2014). For example, higher levels of private religiosity have been found to decrease the likelihood of engaging in minor crimes, like status and property offenses (Benda and Corwyn 2001), and more serious crimes, like armed robbery . In sum, both attendance at religious services and private religiosity have been found to be effective as protection from minor and more serious forms of crime (Salas-Wright et al. 2014).
Turning our attention to the influence that religion may have on the delinquency and criminality of youth, we find that religiosity can reduce the antisocial behaviors of young people, including adolescents and emerging adults (Salas-Wright et al. 2014). Studies, such as , have utilized attendance at religious services as a predictor to study social bonds and turning points that influence offending during emerging adulthood. The results of their study found that those who attend religious services more often have an inverse relationship with criminal offending during emerging adulthood. Other studies have found that religiosity can be a protective factor against substance use and abuse for adolescents, including alcohol, tobacco, and marijuana (Johnson et al. 2000Wallace et al. 2003). Baier and Wright (2001); and Salas-Wright et al. (2014), have found that youth who are involved in religious organizations and have a high level of religious engagement have lower rates of violent behaviors, such as hitting, carrying weapons, and violent attacks on others. Studies have also found an association between religiosity and decreased levels of behaviors like theft and selling drugs (Johnson et al. 2000). While the evidence strongly suggests that religion is a viable protective factor against delinquency and crime for adolescents and young adults, it is important to consider demographic factors such as race and gender are important when examining the relationship between religion and antisocial behaviors in youth populations.
Looking at demographic factors, evidence suggests that religion may act as a protective factor, but that how it does so may vary. For example, using a national sample of adolescents, Caputo (2008) found that gender moderates the relationship between religiosity and risk factors, with the protective influence of religion being stronger for females relative to males. Other studies have shown similar findings, such as Johnson and Morris (2008), who found that religiosity was also only significant in females, and not in males. More recently, Salvatore and Markowitz (2014) used data from Add Health to cross-sectionally examine the influence of social bonds on turning points in delinquency and crime from childhood through emerging adulthood. Using a life course perspective, they sought to identify and compare the turning points and social bonds previously identified in the life course literature that predicted offending across multiple waves of the Add Health data. At each wave, Salvatore and Markowitz found several gender-specific relationships between social bonds/turning points and delinquency/offending. Many of their findings identified that previously-identified variables 'worked' as expected for both males and females. However, their findings in regard to religion as a social bond (measured by attendance at religious services) identified that religion only predicted a decrease in offending in males at wave 3 (when the sample was in emerging adulthood), and at wave 4, higher levels of religious participation were predictive of decreased levels of offending for both males and females. Other studies, like Salas-Wright et al. (2013), found little difference in the relationship between religiosity and violent offending based on gender.
Scholars have also theorized that race may be a key predictor in influencing the role of religion as a social control (Johnson et al. 2000). This theory is supported by studies such as Markowitz and Salvatore (2012) who used multiple waves of data from the National Longitudinal Study of Adolescent Health to examine patterns of offending by race. Results of the study showed that during wave 3 (when the sample was emerging adulthood), there was a predictive effect of religion (measured by attendance at religious services) for more serious offenses by race, with African Americans who had lower levels of attendance at religious services, being more likely to report serious offending. This suggests that race may play a unique role in religion or mediate the role of religion as a social control during emerging adulthood.
As discussed above, religiosity profiles are key components in understanding the role that religion plays in emerging adult populations. However, to date, research exploring their influence has been limited. One such study that has addressed this area is Salas-Wright et al. (2014), which used data from the National Survey on Drug Use and Health and the National Epidemiologic Survey on Alcohol on Related Conditions to conduct latent profile analyses on the influence of latent religiosity classes and antisocial behaviors, substance use, and substance use disorders. The results of their study revealed a four class solution for both samples. In particular, the emerging adults classified as publicly and privately devoted to religion had a much lower likelihood of being involved in a variety of risk behaviors or meeting criteria for substance abuse disorders. Further, the study also found that the greater the level of religiosity an individual has, the lower the likelihood of risky behaviors.
In sum, the literature to date regarding the influence of religiosity on the offending of youth populations largely supports the idea that higher levels of religious engagement act as a protective factor against antisocial behaviors, including substance use and abuse, and non-violent and violent crime. For emerging adults, in particular, research has evidenced that this relationship holds true despite some evidence suggesting they have a reduced participation in religious services. Demographic factors, like gender and race/ethnicity, may also shape the role of religion as protective factors against antisocial behaviors, like crime and substance use, in youth populations.

Conclusions
This paper examined the influence that religion has on offending behaviors during emerging adulthood. While religion has been a component of prior studies examining emerging adulthood, there has yet to be a comprehensive summary examining the role of religion as a key social bond during emerging adulthood that further explores how that bond may influence offending. As much of the prior literature (e.g., Arnett 1998; Salvatore and Taniguchi 2012) has found religion to be a relevant variable, there is a need for further research examining the role of religion as a factor influencing offending during emerging adulthood and whether other social bonds can suitably take the place of religion in mitigating criminal offending. This is especially important given the diminished religiosity of emerging adults and the open question about whether these young people will turn to religion later in life. Studies are needed to explore the long-term effects of religion as a social bond on not only offending during emerging adulthood, but on the prevalence of other risky and dangerous practices during this life stage (e.g., unsafe sex, substance use/experimentation). Longitudinal studies need to be conducted as emerging adults age, to provide a retrospective look at how religion did or did not impact these behaviors for those who experienced emerging adulthood. As a relatively new stage of the life course, this is a much needed area of study and one that will continue to be of value as young people continue to push off the responsibilities traditionally seen as part of becoming a full-fledged adult. Further, studies need to examine the role of religion as a social bond for emerging adult samples in non-Western nations. A large portion of the existing research has largely focused on Western nations, and as such, there is a lack of research examining how religion 'works' as a social bond for emerging adults in other cultures, such as in the youth-heavy Middle East.
Religion, with its emphasis on milestones and turning points, such as the Jewish bar-mitzva or the Catholic confirmation seeks to usher the individual from childhood to adulthood through a time-honored path. As post-modern values take hold, including diminishing religiosity in Western societies and a continued emphasis on individuality (Inglehart 1997), this set life path has lost its luster. Religious institutions are adjusting through, for instance, a renewal of initiatives to set up interfaith spaces on college campuses (see Dupuis 2016) as a means of bridging the values of emerging adults with older generations. Yet, these institutions need to continue to redefine themselves and their missions in a world where social justice goals are increasingly popular, but where traditional religious views on lifestyle choices, like homosexuality or premarital cohabitation, are out of sync with the youth. Any finding that links religious practice to positive outcomes, such as fewer arrests or reduced criminal behavior, should be touted by religious institutions as they seek to attract a new generation that may be suspicious of organized religion. Still, from a criminological perspective, religion is not the sine qua non of a safe, law-abiding society. So, while religion is an important factor to consider in the criminal offending of emerging adults, it should be noted that societies in Western Europe that do not highly value religion also have very low levels of crime (Theodorou 2015;Harrendorf et al. 2010). Many highly religious societies are also prone to high rates of violence, as depicted in the Old Testament (see Pinker 2012) and as seen in Indonesia, Nigeria and Pakistan (Theodorou 2015).
The role of religion as a key social bond and control that may influence the criminal offending of emerging adults has policy and theoretical implications. As criminal offending is common in emerging adults, albeit this being typically lower level (non-violent) offending, diversion programs could be utilized to process emerging adults who commit crime. As religion is a social bond which encourages prosocial behavior, religiously-oriented treatment programs could be utilized as a diversion tactic. In terms of prevention, churches, mosques, temples, and synagogues could develop programs to target emerging adult populations, offering guidance and support during the exploration common in emerging adults. Education and guidance could be offered in the areas of sexual behaviors, decision-making, and other faith-specific areas. These strategies could also be utilized on college campuses. As colleges struggle with on-campus drug use, sexual exploration and victimization, interfaith centers could offer prevention and treatment strategies aimed at bringing the college community together to help support emerging adults. These actions would all have the additional result of building community, an important facet of democracies (Putnam 2000).
In regard to theory, religion acts as a key social bond inhibiting criminal behaviors. Several studies dealing with samples of emerging adults support this idea and suggest religion as a social bond relevant for emerging adults. Theories, such as social control theory and the age-graded theory of social control, may want to add additional focus on religion as a social bond that offers a strong attachment to conventional society and may be useful in preventing offending during emerging adulthood and influencing desistance for those offending during this stage of the life course.
Targeting religiously-oriented primary and secondary prevention strategies towards emerging adults may prove challenging. Colleges and universities may look like ideal locations for such programs, but not all emerging adults attend college after high school, and further, not all institutions may support the use of religious-based programming or have the funds to provide such programming. As previously stated, emerging adults may not see religious institutions as particularly desirable or welcoming, especially if they participate in lifestyles traditionally denounced by religious leaders. Religious community-based interventions for emerging adults could serve as another potential venue. However, traditional youth centers may not provide services for those over the age of 18 and may face funding challenges. For all of these reasons, the role of religion and its influence on emerging adults for theory and policy is a rich area which has only just begun to be explored.
Author Contributions: C.S. contributed to all sections using his expertise in emerging adulthood, as well as prior experience conducting empirical studies to research the relationship between emerging adulthood and crime. He provided the conceptual and narrative framework for the present study. G.R. contributed to all sections drawing on his expertise on religion and crime. He particularly contributed to the conclusions and theoretical framework.
Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.