Moderators of the relationship between religiosity and alcohol use in college students.

Bodford, Jessica E. ; Hussong, Andrea M.

Alcohol use and abuse among college students is of serious concern for a variety of social and health-related reasons. High levels of consumption have been attributed to impaired academic performance, early onset of substance dependence, risk of vehicle-related accidents, and elevated threat of violence, sexual assault, and other illegal behaviors (Ellison, Bradshaw, Rote, Storch, & Trevino, 2008). In 2007 the U.S. Surgeon General declared underage alcohol use a chief health concern for the nation, emphasizing the importance of research into the predictors of and protectors against college-age drinking (Brown, Salsman, Brechting, 8c Carlson, 2007). Recent investigations into these factors have established a number of prevalent variables related to alcohol use including both demographic (e.g., gender, ethnicity, socioeconomic status) and biological factors (prenatal exposure, generic predisposition, cognitive functioning) that remain relatively constant throughout the lifespan (Chassin, Hussong, 8c Beltran, 2009). However, environmentally determined correlates are more likely to differ as a function of time as aspirations, stress, and peer groups undergo change and maturation, likewise producing varying effects on substance use.

Consequently, although religiosity--that is, the quality of being religious or of devout faith--has been regarded as one such environmental protective factor in adulthood, the mechanisms behind this relationship remain unclear among emerging adults (Brechting et al., 2010; Convyn, 2002). Whereas research asserts that adults frequently use alcohol as a means of coping with stress, anxiety, or depression (e.g., Grunbaum, Tortolero, Weller, & Gingiss, 2000; Strawser, Storch, Geffken, Killiany, & Baumeister, 2004; Wills, Yaeger, & Sandy, 2003), drinking behaviors within the college population are primarily driven by social motives (Bahr, Maughan, Marcos, & Li, 1998; Perkins, 1987; Regnerus, 2006; Westmaas, Moeller, & Woicik, 2007). For this very reason, studies have suggested that public religiosity's impact on alcohol use is largely mediated by peer and parent lifestyles, values, and socialization (Burkett, 1993; Ellison et al., 2008; Perkins, 1987; Regnerus, 2006; Wood, Read, Mitchell, & Brand, 2004). Even so, measurements of religiosity as a construct rarely separate these social factors from more internal, private forms of devotion and, when assessed with respect to substance use and other risky behaviors, religiosity is not often analyzed beyond participants' religious denomination and frequency of service attendance. This study therefore aims to expand this limited area of research within the emerging adult population, at which time college students have the choice to pursue or refrain from opportunities to establish new peer groups, drink while underage, maintain contact with family, and become religiously involved, all of which may alter perceptions of and behaviors toward alcohol.

Religiosity and Alcohol Use

The sparse literature concerning the association between religiosity and substance use has generally not distinguished between religion and spirituality, which have often been combined into a single construct (e.g., Bert, 2011; Good, Willoughby, & Busseri, 2010). Hodge, Andereck, and Montoya (2007) were among the first to distinguish spirituality and religiosity based on the ways in which people exhibit their beliefs. Specifically, they proposed that spirituality be defined in individual, existential, and relational terms of an almost transcendental nature (e.g., I often experience a feeling of unity with the earth and all livingthings) while religion should be displayed as a combination of communal, organizational, and structured expressions (i.e., frequency of church attendance and religious organization participation).

Building off this distinction, a study conducted by Borders, Curran, Mattox, and Booth (2010) assessed religiosity through a set of clear dimensions that not only addressed examples of faith and internal thoughts but also separated private religious practices from spiritual beliefs. This distinction between private religiosity and spirituality is key for future research, because although the two are often considered analogous, it is possible to hold spiritual beliefs without worshipping the deity or set of deities of a particular religion (Worthington, Kurusu, McCollough, & Sandage, 1996).

Public Religiosity and Socialization

Defined broadly, private religiosity is any action that is carried out in accordance with one's religion but is not necessarily seen by the wider community. Examples include internal thoughts and beliefs, prayer and reflection, religious self-ranking, attachment to and importance of religion, and personal religious awareness (Chitwood, Weiss, & Leukefeld, 2008). Public religiosity, on the other hand, constitutes participation in religious activities that entail social contact and communication with others, such as frequency of service attendance, involvement in religiously affiliated retreats and mission work, participation in religiously based organizations, and roles within each of these spheres.

It therefore follows that interactions with similarly involved peers, which may take place through public religious means, may account for a more personal outward religious experience (Borders et al., 2010). Consequently, a small number of recent studies (Brownfield & Sorenson, 1991; Edlund et al., 2010; Menagi, Harrell, & June, 2008) were designed to examine the possibility that social support facilitates the inverse relationship between religiosity and emerging adult alcohol use, focusing specifically on feelings of solidarity and quality of communication with parents, friends, and religious communities. It is possible that these studies failed to find significant support for their hypotheses at least in part because, as is the case in a majority of research into this relationship, religiosity was defined through a composite of private and public manifestations without studying their unique impacts on reported drinking behaviors. Indeed, when Ozorak (1986) measured religiosity only through involvement and active participation, publicly religious students reported that their relationships with many sources of social support were created from or maintained through religious involvement.

It is therefore possible that while family, peer, and religious cohesion could be related to religiosity, their social impacts (c.g., on alcohol use) are most evident when interactions with these groups occur through direct communication in religious meetings, services, and other gatherings. This concept closely follows the theory of primary socialization, which stems from parental influences on norms and values, and secondary socialization, pertaining to involvement in academic and religious institutions (Lubbers, Jaspers, & Ultee, 2009). Both forms of socialization suggest that the morals and values established within these three distinctive social groups (i.e., parents, friends, and religious communities) can directly or indirectly influence college-age drinking because these groups serve as a foundation for attitude and belief formation. If public religiosity thus impacts social networks through associations with individuals or larger groups who are religiously involved in similar ways, the socialization theory would suggest that publicly religious students are more likely to adopt the belief systems, behaviors, and values instituted by the norms of their social groups (Roberts, Koch, 8c Johnson, 2001). Consequently, peers and parents may protect students from alcohol use if they are less inclined toward underage drinking behaviors, regardless of whether those students are religious.

Hypotheses

Collectively, because of the socially determined nature of college-age alcohol use, this study focuses on the differing impacts of public and private facets of religiosity on drinking behaviors in college settings. More specifically, public manifestations entail some level of social interaction through which attitudes in favor of or against alcohol use can be created and emphasized, although such manifestations may not accurately reflect internal belief structures (e.g., students may be outwardly religious without holding private religious beliefs). Because both college-age drinking and public religiosity are motivated by social influences, the relationship between these two constructs may therefore be compromised under certain socially or religiously relevant circumstances while the relationship between alcohol use and private religiosity continues to hold.

Taking into account the fundamental differences between public and private religiosity and their relationships with socialization effects and moral values, it is hypothesized that the negative association between public or private religiosity and alcohol use is moderated under the conditions that:

1. Private religious tendencies are less common than public religious practices.

2. Peer and/or parent alcohol use and approval of underage drinking is high.

3. Peers and/or parents are not religiously involved and previous exposure to religiously affiliated schools is limited, as per the socialization theory.

By examining the effects of these socially motivated factors on alcohol use, we can further our understanding of the conditions under which public and private religiosity separately protect against college-age drinking. This study is the first to empirically overcome the limitations of existing research both by focusing on distinct social influences on drinking behaviors and by classifying religiosity as a topic of primary interest. Not only has prior research on alcohol use failed to place more than just minimal emphasis on religiosity's protective impacts (Chitwood, Weiss, & Leu kefeld, 2008), but scientists with a key interest in the study of religiosity have rarely published in the field of substance abuse.

Because this intermittent overlap has hindered thorough investigation, it is possible that by promoting and expanding this limited area of research we can facilitate our understanding of religiosity's protective influences, generalize these specific components into a broader comprehension of social drinking, and enhance existing prevention and intervention strategies to more effectively reduce college-age alcohol use and its serious repercussions on students, universities, and the larger community.

Method

Procedure

Students enrolled in an introductory psychology course were given the option to complete a self-report inventory as part of a class research participation requirement. The mean completion time of this online questionnaire was 31.09 minutes (SD = 14.01 min). Of those who started the survey, 4.49% left before completion and 1.63% were of legal age (21 or older) and therefore dropped from further analysis. Questions were separated into construct-themed blocks that were displayed in the following order: demographics and background information; parent, peer, and personal alcohol use and views toward alcohol; and parent, peer, and personal religiosity. All aspects of this study were reviewed and approved by the Institutional Review Board.

Participants

The sample was composed of 230 students (60.0% female) at a large southeastern public university, all of whom were between the ages of 17 and 20 years old (M = 18.77, SD = 0.76). Of the respondents, 70.4% were non-Hispanic White, 9.7% were Black or African American, 7.8% were Hispanic, 7.4% were Asian, 3.9% were of two or more ethnicities, and 0.8% were of Native American or Pacific Islander descent. In terms of gender and racial characteristics, the sample was representative of the wider university undergraduate population (Office of the University Registrar, 2010).

70% of students were raised in middle- to upper-class households with a yearly income of at least $70,000 in 2010. At the time of assessment, 79.1% of respondents lived in an on-campus residence hall whereas 7.0% lived in a fraternity or sorority house and the remaining 13.9% lived in an off-campus house or apartment. No participants reported living at home with parents or guardians.

With regard to religious affiliation, 75.2% of students were affiliated with a Christian denomination, 75.7% of whom belonged to Protestant denominations (e.g., Baptist, Methodist, Presbyterian, Non-denominational). Of the 20.9% who were unaffiliated, 7.8% were Agnostic (i.e., not sure if there is a God) and 6.5% each were Atheist (do not believe in God) or nothing in particular. Participants reported that 84.9% of mothers and 81.7% of fathers belonged to Christian denominations, with 7.8 and 11.3% unaffiliated, respectively. Those comprising Jewish, Muslim, Buddhist, Hindu, and other world faiths were uncommon, ranging between 3.9% of participants and 7% each of mothers and fathers.

Measures

Alcohol use. Participants were asked to indicate how often they drank alcohol during a typical month in the school year on a 9-point Likert-type scale ranging in frequency from Not at all to Every day. This item was recoded to reflect increasing intensity in frequency, so that instead of attributing equal weight between 2 (Once a month), 6 (Twice a week), 8 (Four or five days a week), and so on, numbers were distributed closer to the actual frequency of monthly drinking episodes. Students who reported drinking at least once a month were then asked how much they usually drank for each 24-hour period they drank during a typical month, ranging from: Less than 1 drink to 9 or more drinks. During both questions, participants were reminded that a "drink" refers to a can or bottle of beer, a wine cooler, a glass of wine, a shot of liquor straight, or a mixed drink.

A single alcohol use item was created by multiplying the frequency of drinking episodes per month by the number of drinks per episode. The average number of alcoholic drinks consumed per month within this sample was 22.4 (SD = 26.1).

Greek involvement. Participants indicated their degree of affiliation with on-campus Greek life on a 4-point Likert-type scale, namely Not at all affiliated, Non-member who occasionally associates with members, Non-member who frequently associates with members (e.g, regular attendance at fraternity parties), or Active member (Sher, Bartholow, & Nanda, 2001). As is summarized in Table 1, 31.3% of respondents were active members at the time of data collection, 11.3% were non-members who frequently associated with on-campus Greek events, 29.6% were non-members who occasionally associated with these organizations, and the remaining 27.8% were not at all affiliated.

TABLE 1 Preliminary analyses of demographic covariates

Variable M (SD) / % Alcohol use Religiosity
 M(SD) M(SD)

Gender F(1) = 18.37 *** F (1) = 2.58

Female 60.0% 16.56(20.30) 0.32(0.24)

Male 40.0% 31.06(31.04) 0.27(0.21)

Age 18.77 F (3) = 2.29 F (3) = 5.15 **
 (0.76) [dragger]

18 41.7% 18.89(24.25) 0.34(0.24)

19 38.7% 22.30(24.35) 0.29(0.21)

20 19.1% 30.55(31.66) 0.23(0.22)

Income F (5) = 4.10 ** F (5) = 1.45

< $20,000 6.1% 9.32(14.08) 0.36(0.24)

520,000-539.999 7.0% 8.53(10.51) 0.32(0.27)

$40,000-570.000 17.0% 14.46(22.31) 0.34(0.26)

$70,000-5100,000 22.2% 26.59(30.45) 0.24(0.24)

5100.000-5150.000 20.0% 21.79(25.58) 0.34(0.18)

> $150,000 27.8% 30.52(26.35) 0.28(0.21)

Residence F (3) = 3.68 * F (3) = 2.50

On-campus dorm 79.1% 19.85(24.51) 0.32(0.24)

Greek house 7.0% 39.53(34.38) 0.27(0.20)

Off-campus residence 13.9% 28.56(27.45) 0.21(0.16)

Greek affiliation F (3) = 16.53 F (3) = 0.55
 ***

Active member 31.3% 8.40(20.35) 0.33(0.30)

Non-member, frequently 11.3% 18.26(23.14) 0.29(0.22)
associates

Non-member, 29.6% 30.97(25.88) 0.28(0.13)
occasionally
associates

Not at all affiliated 27.8% 35.53(26.27) 0.30(0.19)

*** p < 0.001, ** p <0.01, * p < 0.05, [dragger] p < 0.10

Religious affiliation. Religious affiliations and specific denominations were extracted from the U.S. Religious Landscape Survey (Pew Research Center, 2008), which has exhibited high test-retest reliability (a = 0.93). Participants were asked to identify the religious affiliation to which their mother and father separately adhered during most or all of participants' childhood and adolescence before reporting their own current religious affiliation.

Public religiosity. At the time of questionnaire compilation, no scale separated religiosity into its public and private manifestations. Eight items were therefore fashioned after examples of public religious involvement cited in previous studies, namely: frequency of church or religious service attendance in the previous month; on- or off-campus religious organization attendance in the previous month and any active roles (e.g., specific jobs, duties, or titles) held in such events; mission work with a religious purpose or affiliation; and attendance at religious retreats or conventions, studies or meetings, and religious choir. Responses were averaged into a composite scale ranging from 0 to 1 with lower scores indicating lower degrees of public religiosity. Inter-item correlations and internal consistency reliability were modest (r = 0.28, a = 0.77).

Private religiosity. After indicating how often they prayed privately, read religious scripture, listened to religious radio or music, and watched religious television, participants indicated the importance they prescribed to relying on religious teachings in times of trouble, believing in God or a Higher Power or Creator, relying on religious beliefs as a guide for day-to-day living, and turning to prayer when facing personal problems. To expound upon aspects of religiosity extending beyond concrete actions and perceived importance, Rohrbaugh and Jessor (1975) provide a set of highly reliable religiosity subscales embodying the non-social element of private religiosity, three of which were used for the present study: ritual religiosity (i.e., prayer or religious meditation), consequential religiosity (When you have a serious personal problem, how of ten do you take religious advice or teaching into consideration?), and experiential religiosity (Do you agree with the following statement? "Religion gives me a great amount of comfort and security in life"). Each of the 11 items was measured on a 5-point Likert-type scale, and all responses were compiled into a single composite score ranging from 0 to 1 with lower scores indicative of lower levels of private religiosity. Similar to public religiosity scale analyses, this construct demonstrated a modest inter-item correlation (r = 0.49) and level of reliability ([alpha] = 0.71).

Parent alcohol use. The Children of Alcoholics Screening Test (CAST; Jones, 1982) was used to identify parents or guardians who are problematic drinkers by asking participants to report their perceptions of and experiences with each parent's drinking separately. The scale comprises 30 items scored dichotomously and scored by the number of positive answers, with six or more affirmative responses indicating parental alcoholism. The CAST scale yielded a strong inter-item correlation (r = 0.48) and reliability ([alpha] = 0.96) within this sample. Descriptive statistics of all variables of interest are summarized in Table 2.

TABLE 2 Construct and scale descriptive analyses
Construct M(SD) Range Items [alpha] Inter-item r

Drinking

Parent 2.3 (5.5) 0-30 30 0.96 0.48
alcohol use

Peer alcohol 0.5 (0.3) -0.3-0.9 7 0.79 0.79
use

Peer pressure 0.3 (0.2) 0-1 4 0.86 0.62
to drink

Peer approval 2.5 (1.3) 1-7 7 0.84 0.44
or alcohol
use

Parent 1.4 (0.7) 1-7 7 0.85 0.47
approval of
alcohol use

Religiosity

Public 0.1 (0.1) 0-0.6 8 0.77 0.28
religiosity

Private 0.2 (0.2) 0-0.7 11 0.71 0.49
religiosity

Peer 10.0 (8.0) 0-35 8 0.90 0.54
religiosity

Parent 31.1 (11.2) 10-50 10 0.93 0.59
religious
emphasis

Peer alcohol use. Peer alcohol use was assessed using the seven-item scale presented in Bartholow, Sher, and Krull (2003). Of these items, two pertain to peer perceptions of drinking and getting drunk; three pertain to the number of close friends who drink, drink to get drunk, and regularly get drunk; one relates to the amount these friends typically drink per episode; and one relates to how often these friends get drunk. Each question was scored on a Likert-type scale, with a composite score indicating overall peer alcohol use with higher scores indicative of heavier drinking behaviors. Within the current sample, this scale showed a ately high inter-item correlation (r = 0.79) and acceptable internal consistency reliability ([alpha] = 0.79).

Peer pressure to drink. Two items each from the Survey of Personal and Social Development (SPSD2; Jessor, Costa, & Turbin, 2003) and Jamison and Myers (2008) were used to measure the extent to which participants feel pressured by peers to drink. All four items were scaled from 0 (Never) to 3 (All the time) and pertained to how often respondents were encouraged to drink; were encouraged to get drunk; were encouraged to drink the same number of drinks as friends; and had the same number of drinks as friends because their friends thought they should. A composite score was calculated from answers to each of the four items weighted equally and ranging from 0 to 1, with lower scores indicating a lower degree of peer pressure to drink. Responses indicate that these items were relatively closely correlated (r = 0.62) with moderately high inter-item reliability ([alpha] = 0.86) and the combined scale was highly correlated with peer approval of alcohol use (r = 0.48, p <0.001) and peer alcohol use (r = 0.38, p < 0.001).

Peer and parent disapproval of drinking. Peer and parent disapproval of underage alcohol use was assessed using the Injunctive Norms Questionnaire (Baer, 1994). Additional items were used from the Rhode Island Substance Abuse Survey (RISAS; Rhode Island Department of Health, 2000), which was modified from the Monitoring the Future study (Johnston, O'Malley, & Bachman, 1996). These measures ask the extent to which parents or friends approve of various alcohol-related scenarios on a 6-point Likert-type scale ranging from Strongly disapprove to Strongly approve. All responses to these seven items were averaged along a scale From 1 to 7, with lower scores indicating a stronger degree of disapproval. For both peers and parents, these scales revealed a relatively high inter-item correlation (r = 0.44; 0.47) and reliability ([alpha] = 0.84; 0.85).

School affiliation and parent religious emphasis. To assess the degree to which students were exposed to an emphasis on religion in family and academic settings during adolescence, participants were asked to report whether their middle and/or high school was religiously affiliated. Analyses revealed that a majority (75.7%) of respondents did not attend a religiously affiliated middle or high school. A smaller proportion attended either a religiously affiliated middle school (5.2%) or high school (5.7%) while the remaining 13.5 % of students spent the approximately seven years prior to college in a religious academic environment. Participants were then asked to report the extent to which their parents or guardians emphasized family religion as they were growing up. This construct was evaluated through the 10-item Religious Emphasis Scale (RES; Altemeyer, 1988). Respondents selected the amount to which each statement described their adolescence on a 6-point Likert-type scale from None at all to A great deal. A total parent religious emphasis score was determined by adding all 10 items. This scale displayed a high inter-item correlation (r = 0.59) and degree of reliability ([alpha] = 0.93) within the present sample.

Peer religiosity. The same items comprising the Public Religiosity scale were asked concerning friends' religious involvement. Each question prompted participants to respond with the number of their five closest friends who met that particular criterion, with possible scores ranging from 0 to 5. All eight responses were added to produce a single scale of peer public religiosity, which demonstrated high inter-item correlations (r = 0.54) and internal consistency ([alpha] = 0.90).

Results

Preliminary Analyses

To assess the possible influence of demographic co-variates, participants' alcohol use and total religiosity (that is, the sum of public and private religiosity scores) were regressed separately on gender, age, income, residence, and Greek affiliation. Results from these analyses of variance are reflected in Table 1, indicating that although gender, income, residence, and Greek affiliation were significantly related to alcohol use, only age bore a close negative relationship with total religiosity (F[3] = 5.15,p = 0.002).

More specifically, the results showed that males drank significantly more than females (F[1] = 18.37, p < 0.001) and alcohol use increased only marginally with age (F[3] = 2.29, p = 0.08, ns). There were significant differences in alcohol use as a function of income distribution (F[5] = 4.10, p = 0.001) in that students tended to drink significantly more if they came from a combined family income of $150,000 or higher (M = 30.42, SD = 26.35, p = 0.02) compared to those whose incomes ranged from $20,000 to $70,000 (M= 11.50, SD = 16.41, p = 0.02). Tukey HSD post-hoc analyses also indicated that residence played a significant role in alcohol usc (F[3] = 3.68, p = 0.01) with higher levels of drinking reported by students who lived in a fraternity or sorority house at the time of assessment (M= 39.53, SD = 34.38) rather than in an on-campus dormitory (M = 19.85, SD = 24.51, p = 0.02).

As is suggested by these findings in particular, a Tukey HSD post-hoc analysis indicated that levels of drinking increased with degree of Greek affiliation (F[3] = 16,531p < 0.001), with respondents who were active members reporting significantly higher use than those who were not at all affiliated (At = 8.40, SD = 20.35, p < 0.001) and non-members who occasionally associated with on-campus fraternities and sororities (M= 18.26, SD = 23.14, p < 0.001), but not compared with non-members who frequently associated with these organizations (M = 35.53, SD = 26.08, p = 0.84, ns).

Correlations between alcohol use and religiosity constructs revealed that, as predicted, alcohol use was significantly negatively related to public religiosity (r =-0.18, p <0.01) and private religiosity (r =-0.16, p = 0.02). Public religiosity was significantly positively correlated with private (r = 0.55, p < 0.001).

Public Religiosity as a Predictor Variable

Prior to conducting ordinary least squares regression analyses to test moderating hypotheses, public religiosity and private religiosity were centered to increase the interpretability of interaction effects and to avoid potential problems with multicollinearity. The OLS regression model comprised the following elements:

ALC = [[beta].sub.0] + [[beta].sub.1](PUBREL) + [[beta].sub.2](MOD) + [[beta].sub.3](PUBREL x MOD),

where standardized beta coefficients ([beta]) represented the number of standard deviations the dependent variable ([gamma], alcohol use) would change for each increase in standard deviation of the predictor variable ([X.sub.1], public religiosity [PUBREL]; [X.sub.2], the moderator variable [MOD]; and [X.sub.3], the interaction of the two [PUBREL x MOD]). Significant interaction effects were further probed using the computational procedures described by Hayes and Matthews (2009). Because no demographic variables were significantly related to both alcohol use and religiosity, no covariates were included in the regression analyses.

Nine models estimated the moderating effects of the variables that corresponded to the following three hypotheses of the current study, each model separately taking into account the impacts of public and private religiosity on alcohol use. These hypotheses include: (1) private religiosity, (2) peer and parent alcohol use, disapproval of use, and peer pressure, and (3) peer, parent, and high school religious involvement. Results are depicted in Table 3.

TABLE 3

Results of moderation analyses of alcobol use and public religiosity

 OLS Interaction
 Model

Prediction R-sq F (df) P b se t

1. Private 0.52 4.09 (3,226) 0.01 -225.5 122.8 -1.84
religiosity

2. Peer 0.45 62.67 (3,226) 0.00 -87.8 35.3 -2.49
alcohol use

Parent 0.03 2.67 (3,226) 0.05 0.9 1.9 0.49
alcohol use

Peer pressure 0.26 26.87 (3,226) 0.00 -5.7 57.3 -0.10
to drink

Peer 0.20 19.36 (3,226) 0.00 3.3 11.8 0.28
disapproval
or alcohol
use

Parent 0.06 4.40 (3,226) 0.01 20.2 20.0 1.01
disapproval
of alcohol
use

3. Peer 0.05 4.31 (3,226) 0.01 -1.8 1.5 -1.24
religiosity

Parent 0.05 3.60 (3,226) 0.01 -2.4 1.4 -1.68
religious
emphasis

Prediction P

1. Private 0.07
religiosity

2. Peer 0.01
alcohol use

Parent 0.63
alcohol use

Peer pressure 0.92
to drink

Peer 0.78
disapproval
or alcohol
use

Parent 0.31
disapproval
of alcohol
use

3. Peer 0.22
religiosity

Parent 0.09
religious
emphasis

Private religiosity. The first model tested the hypothesis that low levels of private religiosity would diminish the negative association between public religiosity and alcohol use. Although this overall regression model was significant ([R.sup.2] = 0.52, F[3,226] = 4.09, p = 0.01), the interaction was only marginally so (b =-225.5, p = 0.07, ns). Probing of this interaction showed that public religiosity was a significant predictor of alcohol use at high levels of private religiosity (b =-34.8, p = 0.05) but not at mean (b = -2.0, p = 0.93, ns) or low levels (b = 30.8, p = 0.42, ns), indicating that greater private religiosity only predicted lower levels of alcohol use for those students who had high levels of public religiosity (see Figure 1).

[FIGURE 1 OMITTED]

Peer and parent alcohol use and disapproval. Five models tested the hypothesis that high levels of (1) peer or (2) parent alcohol use, low levels of (3) peer or (4) parent disapproval of underage drinking, and high exposure to (5) peer pressure would diminish the negative association between public religiosity and alcohol use.

Although the moderating effect of parent alcohol use (b = 0.9, p = 0.63, ns) on the relation between private religiosity and alcohol use was non-significant, the moderating effect of peer alcohol use on this association was significant (b =-87.8, p = 0.01). Further probing of this moderating effect indicated that public religiosity most strongly predicted alcohol use when peer use was below average (b = 23.1, p = 0.06, ns), at which time alcohol use was positively related to public religiosity (see Figure 2). The effect of public religiosity on alcohol use was non-significant at mean (b = 0.1, p = 0.99, ns) and high levels of peer use (b =-23.0.p = 0.26, ns), peer disapproval (b = 3.3, p = 0.78, ns), or parent disapproval (b = 20.2, p = 0.31, ns) on the relation between religiosity and alcohol use.

[FIGURE 2 OMITTED]

Peer, school, and parent religiosity. This hypothesis stated that if peers or parents were religiously involved, the relationship between alcohol use and public religiosity would be significantly more negative than that observed in participants whose peers and parents showed higher levels of religious involvement. Similarly, secondary socialization theory would suggest that exposure to religiosity through religiously affiliated middle and/or high schools could also play a role in the development of drinking behaviors as a function of religiosity. Although the regression of public religiosity on alcohol use was significant across levels of peer religiosity ([R.sup.2]= 0.05, F[3,226] = 4.31, p = 0.01), an interaction was not present (b =-1.8, p = 0.22, ns).

To assess the influence of exposure to religiously affiliated academic settings, a one-way ANOVA evaluating participants' religious educational background suggested that although those who attended a religiously affiliated middle and/or high school did not report significantly different levels of alcohol use (F[3] = 1.97, p = 0.12, ns) or private religiosity (1131 = 0.37, p = 0.77, ns) compared to those who were not as frequently exposed to such environments, mean scores of public religiosity varied significantly between the four groups (F[3] = 2.66, p = 0.05). More specifically, a Tukey HSD post hoc test revealed that those who attended a religiously affiliated middle school displayed higher levels of public religiosity than those who attended an affiliated high school (NI = 0.02, SD = 0.03, p = 0.05) but not those who attended both a middle and high school with a religious affiliation (M = 0.09, SD = 0.12, p = 0.66, ns). Those who did not attend an affiliated middle or high school did not report significantly different levels of public religiosity compared to others (0.19 [less than or equal to] p [less than or equal to] 0.70, ns).

After separately assessing the effect of peers and religious environments, parent religious emphasis, public religiosity and their interaction were regressed on alcohol use, yielding a significant overall model (R2 = 0.05, F[3,226] = 3.60, p = 0.01) and a marginally significant interaction effect (b =-2.4, p = 0.09, ns). Further probing revealed that the effect of public religiosity on alcohol use was most evident at high levels of parent religious emphasis in the home (b =-56.1, p <0.01) but not at mean (b =-29.4, p = 0.12, ns) or low levels (b = = 0.93, ns). A graph depicting these trends can be found in Figure 3, which suggests that greater public religiosity is associated with lower alcohol use only for those whose parents highly emphasize religiosity.

[FIGURE 3 OMITTED]

Private Religiosity as a Predictor Variable.

All previously tested models were repeated in such a way that private religiosity and the interaction term of private religiosity and a moderating variable corresponding to any given hypothesis were regressed on alcohol use as per the equation ALC = ([[beta].sub.0] + [[beta].sub.1](PRIVREL) + [[beta].sub.2](MOD) + [[beta].sub.3](PRIVREL x MOD). All regression models were significant (0.04 [less than or equal to] R2 [less than or equal to] 0.45, 3.31 [less than or equal to] F[3,226] [less than or equal to] 61.51, 0.001 [less than or equal to] p [less than or equal to] 0.02), and results are presented in Table 4.

TABLE 4

Comparison of public and private religiosity as predictor variables

 Private
 religiosity

Prediction R-sq F (df) p b se t p b

2. Peer 0.45 61.51 0.00 -76.7 30.8 -2.49 0.01 -87.8
alcohol use (3.226)

Peer 0.21 19.48 0.00 11.8 8.2 1.45 0.15 3.3
disapproval (3,226)
of use

Parent 0.07 5.62 0.00 53.0 19.2 2.76 0.01 20.2
disapproval (3.226)
of use

 Public
 religiosity

Prediction se t p

2. Peer 35.3 -2.49 0.01
alcohol use

Peer 11.8 0.28 0.78
disapproval
of use

Parent 20.0 1.01 0.31
disapproval
of use

Peer alcohol use was a significant moderating factor (b =-76.7, p = 0.01), indicating that the relationship between alcohol use and private religiosity was only significantly negative for those with peers who drank heavily (b =-25.5, p = 0.06, ns), but not at mean (b = 14.7, p = 0.18, ns) or low levels (b =-5.4, p = 0.56, ns). Although these trends appear to be roughly identical to those characterizing the relation between alcohol use and public religiosity (see Figures 4 and 2), it is worthy of note that public religiosity significantly predicted increased drinking levels at low levels of peer use while private religiosity significantly predicted decreased drinking at high levels of peer use.

[FIGURE 4 OMITTED]

In contrast, parent disapproval of alcohol use yielded a significant interaction effect when predicted through private religiosity (b = 53.0, p = 0.01) but not public religiosity (b = 20.2, p = 0.31, ns). These trends, which are shown in Figure 5, indicate that the effect of private religiosity on alcohol use was significantly dependent on parents' disapproval of underage drinking (b = -60.21, p < 0.001). Parent ambivalence--that is, neither approving nor disapproving of alcohol use--also played a heavy role on the negative association between alcohol use and private religiosity, although not as strongly (b = -23.03, p = 0.05). For participants whose parents approved of underage drinking, alcohol USC was not associated with private religiosity (b = 14.1, p = 0.44, ns).

[FIGURE 5 OMITTED]

Discussion

The purpose of this study was to expand a previously limited area of research investigating the impact of religiosity on alcohol use within the college student population. Our findings underscore a number of conditions under which religiosity does not as effectively protect against alcohol use as a result of social, familial, and religious factors.

Demographic Variables

Analyses of demographic covariates supported a number of trends observed in previous studies with respect to the relationship between alcohol use and gender, income, and Greek affiliation. More specifically, higher levels of drinking have been noted in college males compared to females, possibly as a result of increased risk-taking behavior (Neumann, Leffingwell, Wagner, Mignogna, & Mignogna, 2009) and perceived masculine norms related to intoxication (Iwamoto, Cheng, Lee, Takamatsu, 8c Gordon, 2011; Carey & DeMartini, 2010). For those who hail from affluent socioeconomic backgrounds, if leisurely lifestyles were evident in the home it is possible that so too was the pursuit of pleasure and relaxation in place of responsibilities (e.g., early morning part-time jobs necessitating sleep and sobriety), which may have emphasized recreational drinking from an early age. Research has also indicated that Greek organizations may encourage the perception of alcohol as a facilitator of social interaction with others, often underestimating and misperceiving the risks associated with underage use (Barry, 2007; Park, Sher, Wood, & Krull, 2009; Capone, Wood, Borsari, & Laird, 2007; Park, Sher, & Krull, 2008). These findings collectively support the results of preliminary analyses within this study.

As has been consistently observed in past research, alcohol use was significantly negatively correlated with religiosity. Because this study is original in its separation of religiosity into public and private manifestations when analyzing alcohol use, it is further worthy of note that this strong negative trend was observed in public and private domains of religiosity. Because we would expect these scales to be theoretically similar, their high degree of convergent validity suggests that this is indeed the case.

Hypothesis Evaluation

The purpose of this study was to identify the conditions under which religiosity effectively protected against alcohol use, looking specifically at factors that would either strengthen or weaken the negative relationship between public and private religiosity and alcohol use. Our analyses indicated that not all tested factors were significant moderators of this relationship. These significant and non-significant findings shed light on the potential for social, familial, and religious factors to impact college-age drinking behaviors.

Private religiosity. Public religiosity was a significant predictor of alcohol use only at higher levels of private religiosity, in which case the relationship between public religiosity and alcohol use was significantly negative. Participants who scored low in private religiosity tended to drink slightly more, rather than less, as public religiosity increased. This marginally significant interaction effect therefore supports our hypothesis that religiosity is not necessarily negatively correlated with alcohol use in college students, as has been indicated in past studies. Even when public religiosity is high, religiosity's protective influence diminishes in the absence of private religious actions such as prayer, belief in supernatural beings, and perceived importance of religion.

A possible explanation for this finding is that students who attend services, organizational meetings, retreats, and other forms of publicly religious involvement do so for reasons that are not related to religion at all (hence the lack of private religiosity) but rather peer involvement, convenience, or customs relayed by family members in the past. When public religiosity is balanced or exceeded by indications of private religiosity, which pertain solely to religious beliefs and internal practices, its long-held protective influence against alcohol use persists. However, for those who view religious involvement as synonymous with a pastime, voluntary commitment, or means of meeting others without engaging in private religiosity, these defenses may prove ineffective.

Peer and parent alcohol use and disapproval. Although the three trends (i.e., low, mean, and high levels of peer use as shown in Figures 2 and 4) depicting alcohol use and peer use appear similar across levels of public and private religiosity, two key differences set these findings apart. First, public religiosity was a significant predictor of alcohol use only for those whose peers drank little alcohol, yielding an overall positive association between use and public religiosity. Second, private religiosity significantly predicted drinking behaviors only at high levels of peer use to produce a negative relationship between use and private religiosity. It would therefore appear that the protective impact of private religiosity buffers the effects of high peer use above and beyond the influence of public religiosity, which may in fact suggest a more ready compliance with social norms for those whose peers do not engage in extreme drinking behaviors.

In suggesting that drinking tendencies are lower in those who are not publicly religious, this positive relationship contrasts with what might be expected as per the socialization theory--namely that young and emerging adults typically mimic peer drinking behaviors (e.g., Bartholow, Sher, & Krull, 2003; Brownfield & Sorenson, 2010; Ellison et al., 2008; Perkins, 1987; Sher, Bartholow, & Nanda, 2001). Because those who are highly religiously involved (i.e., high in public religiosity) are more likely to be exposed to religious communities, this greater degree of contact to such peer groups may lead to a self-other comparison in which students who are highly publicly religious view their peers' drinking tendencies as more conservative than their own if those friendships were established and maintained within religious contexts. Unlike many on-and off-campus parties, alcohol is generally not featured in religious institutions and organizational meetings, thereby leading to the perception that peers established within those networks always behave in the same way that they appear to act during these religiously affiliated events. Students might then report perceived peer use in comparison to their own. It would be beneficial for future research to further analyze this possible explanation by assessing the ways in which participants met their closest peers as well as the contexts in which they most often spend time together.

As predicted, the regression of private religiosity on alcohol use was significantly altered across levels of parent disapproval of alcohol use, although not in the way that we expected. Similar to past findings, students who were highly privately religious tended to drink less if parents disapproved of underage alcohol use (Blobaum & Anderson, 2006; Welte, Barnes, Hoffman, & Dintcheff, 1999; Wood et al., 2004), but also tended to drink more if parents approved of such behaviors. It is possible that students who are high in private religiosity feel more compelled to please their parents by complying with parental attitudes toward underage alcohol use, although this explanation has yet to be studied empirically. That these trends did not appear when assessing the influence of public religiosity on alcohol use suggests that these manifestations of religiosity either are not inherently protective against use, as was the case for those with ambivalent parents (i.e., neither approving nor disapproving of alcohol use), or are not related to compliance with parental attitudes.

Peer and parent religiosity. It is possible that peer religiosity did not explain variability in alcohol use because when compared with private and religiously affiliated universities, student religious norms are less salient in larger, public university settings. Consequently, even though examples of religious involvement can be outwardly observed, participants might not be aware of friends' religious involvement unless they are involved in similar organizations or religious institutions during the school year.

Within families, however, frequent contact within the home increases the likelihood that participants have a close understanding of parent religiosity. This possibility is supported by the marginally significant moderating effect observed with respect to parent religious emphasis in the home. More specifically, higher parent religious emphasis predicted lower levels of drinking only in those who reported high levels of public religiosity, but in those with low public religiosity, parent religious emphasis had the opposite effect. Past studies in the domain of religiosity research have pinpointed a tendency for young and emerging adults to rebel against parents' own religious behaviors by reducing such religiosity while increasing levels of substance use and risk-taking behaviors (e.g., Leonard, Cook, Boyatzis, Kimball, & Flanagan, 2012; Putney & Middleton, 1961; Rice, 1999; Wilson & Sherkat, 1994). Within the current study, it is possible that students who were frequently exposed to religious practices during their high school years (i.e., high parent religious emphasis) but who displayed few forms of public religiosity in college were demonstrating a need to go against parents'

conventions, thereby adopting a higher degree of alcohol use. Further supporting this claim was the observation that for those who did not grow up with parents who emphasized religiosity in the home, alcohol use remained stagnant despite changes in levels of public religiosity. However, it is worthy of note that such acts of rebellion were not found in students who attended private, religiously affiliated schools, with mean levels of alcohol use approximately equal to those of students who did not attended public middle and high schools. To attempt to identify such a rebellion effect longitudinally, it would be beneficial for future studies to measure religiosity over the course of high school and college years.

Implications and Directions for Future

Research Parent religious emphasis may lead to higher use in students who stop attending religious services or who stop showing other forms of religious involvement after leaving home for college. On this point, it would be beneficial for future studies to investigate this trend longitudinally from high school into the first several years of college, sampling both from public and private, religiously affiliated high schools. Furthermore, it is possible that there is an impact of participant religious affiliation, which was not analyzed within this study beyond as a demographic variable. Within certain affiliations it is sometimes the case that illicit substances are seen as a way of enhancing religious experience in either public or private settings. Because customs within affiliations may vary, additional studies may be interested in developing these questions to be specific to each to best capture diverse forms of public and private religiosity as well as the perceived norms of alcohol and substance use within each affiliation.

The impact of peer alcohol use is an important predictor of drinking behaviors regardless of public or private religiosity, stressing the importance of college prevention programs targeting not religious or familial domains but instead use among friends and peer groups. Because of the high degree of external validity of this sample compared to the gender, racial, and ethnic characteristics of the wider university, we feel that we can safely extend these findings to large, public universities comprising similar proportions of student demographics as well as a comparable distribution of parental income. However, generalizability is limited such that all participants were students in an introductory psychology course for which they received research credit upon questionnaire completion. Future studies may wish to expand external validity across universities or through college-oriented survey websites.

While this initial study has indicated the importance of distinguishing between private and public religiosity in the study of underage alcohol use, these findings may serve as a basis for future work in considering how other factors also associated with the college transition (e.g., social groups, educational pursuits, religious beliefs) might together impact drinking behaviors. Furthermore, it may prove valuable to assess the impact of fundamentalism and students' views of God or other deities (e.g., as forgiving or condemning) on illicit substance use (Cook & Hillman, 2006; Cook, Larson, & Boivin, 2003). A similar possible direction for future research would be to capture peer closeness as measured through communication and bonding as well as the degree to which respondents consider their peers to be good role models to determine whether participants' affinities with their peers impact the influence of peer alcohol use on drinking tendencies as a function of religiosity. Similarly, it would be beneficial to evaluate perceived alcohol use norms within the wider university to better understand if student alcohol use stems from a desire to fit in with their close peers or to fit in with the larger student body. In this way, it would not be peer pressure to drink that would dictate alcohol use but instead individuals' beliefs of college student drinking norms regardless of whether or not those students--be they friends, classmates, acquaintances, or strangers--pressure them to engage in such behaviors.

Because our findings held significant differences stemming from two distinct measures of religiosity, this study represents the importance of separating public and private religiosity when studying alcohol use and possibly other substance related behaviors in college students, where outwardly visible forms of religious devotion are not necessarily indicative of internal beliefs, religious importance, and dedication to prayer. Future studies would benefit from separating the historically single construct of religiosity into its distinct domains to better assess the protective its nature against other negative factors including depression, anxiety, and violent behavior, among others.

This study was one of the first to empirically evaluate both social and distinct public and private religious influences operating on alcohol use, focusing on religiosity as a topic of key interest. It is our hope that these findings and those of future research will further our understanding of the predictive and protective factors of alcohol use within the emerging adult population, thereby enhancing college prevention and intervention strategies to more efficiently reduce student drinking behavior and its consequences on society.

References

Altemeyer, B. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco, CA US: Jossey-Bass.

Baer, J. S. (1994). Effects of college residence on perceived norms for alcohol consumption: An Examination of the first year in college. Rsychology of Addictive Behaviors, 8, 43-50. doi:10.1037/0893-164X.8.1.43

Bahr, S. J., Maughan, S. L., Marcos, A. C., & Li, B. (1998). Family, religiosity, and the risk of adolescent drug use. Journal of Marriage & the Family, 60, 979-992.

Barry, A. E. (2007). Using theory-based constructs to explore the impact of Greek membership on alcohol-related beliefs and behaviors: A Systematic literature review. Journal Of American College Health, 56(3), 307-315.

Bartholow. B. D., Sher. K. J., & Krull, J. L. (2003). Changes in heavy drinking over the third decade of life as a function of collegiate fraternity and sorority involvement: A Prospective, multilevel analysis. Health Psychology, 22, 616-626. doi:10.1037/0278-6133.22.6.616

Bert, S. C. (2011). The Influence of religiosity and spirituality on adolescent mothers and their teenage children. Journal of Youth and Adolescence, 40, 72-84. doi:10.1007/s10964-010-9506-9

Blobaum, E. M., & Anderson, J. E (2006). The Impact of exposure and perceived disapproval of underage drinking. Criminal Justice Studies: A Critical Journal of Crime, Law & Society, 19, 171-192. doi:10.1080/14786010600764559

Borders, T. F., Curran, G. M., Mattox, R., & Booth, B. M. (2010). Religiousness among at-risk drinkers: Is it prospectively associated with the development or maintenance of an alcohol-use disorder? Journal of Studies on Alcohol & Drugs, 71, 136-142.

Brechting, E. H., Brown, T. L., Salsman, J. M., Sauer, S. E.. Holman, V. T., & Carlson, C. R. (2010). The role of religious beliefs and behaviors in predicting underage alcohol use. Journal of Child & Adolescent Substance Abuse, 19, 324-334. doi:10.1080/1067828X.2010.502494

Brown, T. L., Salsman, J. M., Brechting, E. H., & Carlson. C. R. (2007). Religiousness, spirituality, and social support: How are they related to underage drinking among college students? Journal of Child & Adolescent Substance Abuse, 17, 15-39.

Brownfield, D. & Sorenson, A. M. (1991). Religion and drug use among adolescents: A Social support conceptualization and interpretation. Deviant Behavior, 12, 259-276.

Burkett, S. R. (1993). Perceived parent's religiosity. friend's drinking, and hellfire: A Panel study of adolescent drinking. Review of Religious Research, 35. 134-154.

Capone, C., Wood, M. D., Borsari, B., & Laird, R. D. (2007). Fraternity and sorority involvement, social influences, and alcohol use among college students: A prospective examination. Psychology Of Addictive Behaviors, 21(3), 316-327. doi:10.1037/0893-164X.21.3.316

Carey, K., & DcMartini, K. (2010). The motivational context for mandated alcohol interventions for college students by gender and family history. Addictive Behaviors, 35(3), 218-223. doi:10.1016/j.addbeh.2009.10.011

Chassin. L., Hussong, A., & Beltran, I. (2009). Adolescent substance use. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology. Vol I: Individual bases of adolescent development (3rd ed.) (pp. 723-763). Hoboken, NJ: John Wiley & Sons, Inc.

Chitwood, D. D., Weiss, M. L., & Leukefeld, C. G. (2008). A Systematic review of recent literature on religiosity and substance use. Journal of Drug Issues, 38, 653-688.

Cook, K. V. & Hillman, E. (2006). Christian views of self and God: Context matters. Journal of Psychology and Theology, 34(2), 133-141.

Cook, K. V., Larson, D. C., & Boivin, M. D. (2003). Moral voices of women and men in the Christian liberal arts college: Links between views of self and views of God. Journal of Moral Education, 32(1). 77-89.

Corwyn, R. F. (2002). Church attendance or expressions of religiosity: Predictors of adolescent drug use. Marriage 6- Family Review, 5(1), 79-91.

Edlund. M. J., Harris, K. M., Koenig, H. G., Han, X., Sullivan. G., Mattox, R., & Tang, L. (2010). Religiosity and decreased risk of substance use disorders: Is the effect mediated by social support or mental health status? Social Psychiatry e- Psychiatric Epidemiology. 45. 827-836. doi:10.1007/s00127-009-0124-3

Ellison. C. G., Bradshaw, M., Rote, S., Scorch, J., & Trevino, M. (2008). Religion and alcohol use among college students: Exploring the role of domain-specific religious salience. Journal of Drug Issues, 38, 821-846.

Good, M., Willoughby, T., & Busseri, M. A. (2010). Stability and change in adolescent spirituality/religiosity: A Person-centered approach. Developmental Psychology, 1-13. doi:10.1037/a0021270

Grunbaum, J. A., Tortolero, S., Weller, N., & Gingiss, P. (2000). Cultural, social, and interpersonal factors associated with substance use among alternative high school students. Addictive Behaviors, 25, 145-151.

Hayes, A. and Matthews, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41, 924-936.

Hodge, D. R., Andereck. K., & Montoya, H. (2007). The Protective influence of spiritual-religious lifestyle profiles on tobacco use, alcohol use, and gambling. Social Work Research, 31, 211-219.

Iwamoto, D., Cheng, A., Lee, C. S., Takamatsu, S., & Gordon, D. (2011). "Man-ing" up and getting drunk: The role of masculine norms, alcohol intoxication and alcohol-related problems among college men. Addictive Behaviors, 36(9), 906-911. doi:10.1016/j.addbeh.2011.04.005

Jamison, J. J., & Myers, L. B. (2008). Peer-group and price influence students drinking along with planned behaviour. Alcohol and Alcoholism, 43, 492-497. doi:10.1093/alcalc/agn033

Jessor, R., Costa, F. M., & Turbin, M. S. (2003). SPSD2: Survey of Personal and Social Development; University of Colorado.

Johnston, L. a, O'Malley, P. M., & Bachman, J. G. (1996). National survey results on drug use from the Monitoring the Future Study, 1975-1995: Vol. I. Secondary school students (NIH Publication No. 96-4139). Rockville, MD: National Institute on Drug Abuse.

Jones, J. W. (1982). Preliminary test manual: The Children of Alcoholics Screening Test. Chicago: Family Recovery Press, Distributed through Camelot Unlimited.

Leonard, K. C., Cook, K. V., Boyarzis, C. J., Kimball, C. N., & Flanagan, K. S. (2012). Parent-child dynamics and emerging adult religiosity: Attachment, parental beliefs, and faith support. Psychology of Religion and Spirituality, 1-14. doi:10.1037/a0029404

Lubbers, M., Jaspers. E., & Ultee, W. (2009). Primary and secondary socialization impacts on support for same-sex marriage after legalization in the Netherlands. Journal of Family Issues, 30, 1714-1745.

Menagi, F. S., Harrell, Z. A. T., & June, L. N. (2008). Religiousness and college student alcohol use: Examining the role of social support. Journal of Religion & Health, 47, 217-226. doi:10.1007/s10943-008-9164-3

Neumann. C., Leffingwell, T., Wagner. E., Mignogna, J., & Mignogna, M. (2009). Self-esteem and gender influence the response to risk information among alcohol using college students. Journal Of Substance Use, 14(6), 353-363. doi:10.3109/14659890802654540

Office of the University Registrar. (2010.June). University of North Carolina at Chapel Hill. Historical Enrollment Statistics: Frequently Requested Reports. Available from http://regweb.unc.edu/stats/freq_reports.php.

Ozorak, E. (1986, August 22). The Development of religious beliefs and commitment in adolescence. Paper presented at the 94th Annual Convention of the American Psychological Association, Washington, DC.

Park, A., Sher, K. J., Wood, P. K., & Krull, J. L. (2009). Dual mechanisms underlying accentuation of risky drinking via fraternity/sorority affiliation: The role of personality, peer norms, and alcohol availability. Journal Of Abnormal Psychology, 118(2), 241-255. doi:10.1037/a0015126

Park, A., Sher. K. J., & Krull, J. L. (2008). Risky drinking in college changes as fraternity/sorority affiliation changes: A person-environment perspective. Psychology Of Addictive Behaviors, 22(2), 219-229. doi:10.1037/0893-164X.22.2.219

Perkins, H. W. (1987). Parental religion and alcohol use problems as intergenerational predictors of problem drinking among college youth. Journal fir the Scientific Study of Religion, 26, 340-357.

Pew Research Center. (2008, February). U.S. Religious Landscape Survey. Religious Affiliation: Diverse and Dynamic. Available from http://religions.pewforum.org/pdf/report-religious-landscapc-study-full.pdf.

Putney, S., & Middleton, R. (1961). Rebellion, Conformity, and Parental Religious Ideologies. Sociometry, 24(2), 125-135.

Regnerus, M. D. (2006). Adolescent delinquency. In H.R. Ebaugh (Ed.), Handbook of Religion and Social Institutions (pp. 265-282). New York: Springer Publishing Co.

Rhode Island Department of Health. (2000, June). The 1998 Rhode Island Adolescent Substance Abuse Survey: Report if statewide results. Available from http://www.health.ri.gov/chic/statistics/asasl.htm.

Rice, E. (1999). Religion and the adolescent: A psychodynamic perspective. Psychoanalytic Psychology, 16(1), 58-75. doi:10.1037/0736-9735.16.1.58

Roberts, A. E., Koch, J. R., & Johnson, D. (2001). Religious reference groups and the persistence of normative behavior: An Empirical test. Sociological Spectrum, 21, 81-98. doi:10.1080/02732170150201829

Rohrbaugh, J. & Jessor, R. (1975). Religiosity in youth: A Personal control against deviant behavior. Journal of Personality, 43, 136-155.

Sher, K. J., Bartholow, B. D., & Nanda, S. (2001). Short- and long-term effects of fraternity and sorority membership on heavy drinking: A Social norms perspective. Psychology of Addictive Behaviors, 15,42-51. doi:10.1037/0893-164X.15.1.42

Steele. C. M., Southwick, L. L., & Critchlow, B. (1981). Dissonance and alcohol: Drinking your troubles away. Journal of Personality and Social Psychology, 41(5). 831-846. doi:10.1037/0022-3514.41.5.831

Strawser, M. S., Storch, E. A., Geffken, G. R., Killiany, E. M., & Baumeister, A. L. (2004). Religious Faith and Substance Problems in Undergraduate College Students: A Replication. Pastoral Psychology, 53(2), 183-188. doi:10.1023/BPASP.0000046829.61450.4c

Welte, J. W., Barnes, G. M., Hoffman, J. H., & Dintcheff. B. A. (1999). Trends in adolescent alcohol and other substance use: Relationships to trends in peer, parent and school influences. Substance Use & Misuse, 34, 1427-1-449. doi:10.3109/10826089909029391

Westmaas, J., Moeller, S., & Woicik, P. B. (2007). Validation of a measure of college students' intoxicated behaviors: Associations with alcohol outcome expectancies, drinking motives, and personality. Journal of American college Health, 55, 227-237.

Wills, T. A., Yaeger, A. M., & Sandy, J. M. (2003). Buffering effect of religiosity for adolescent substance use. Psychology of Addictive Behaviors, 17, 24-31.

Wilson, J., & Sherkat, D. E. (1994). Returning to the fold. Journal For The Scientific Study of Religion, 33(2), 148.

Wood. M. D., Read, J. P., Mitchell, R. E., & Brand, N. H. (2004). Do parent still matter? Parent and peer influences on alcohol involvement among recent high school graduates. Psychology of Addictive Behaviors, 18, 19-30. doi:10.1037/0893-164X.18.1.19

Worthington, E. L., Kurusu, T. A., McCollough, M. E., & Sandage, S. J. (1996). Empirical research on religion and psychotherapeutic processes and outcomes: A 10-year review and research prospectus. Psychological Bulletin, 119(3), 448-487.

Jessica E. Bodford

Arizona State University

Andrea M. Hussong

University of North Carolina at Chapel Hill

Author Information

BODFORD, JESSICA E. Address: Arizona State University, P.O. Box 871104, Tempe, AZ 85287-1104. Email: jessica@bodford.com. Currently: Graduate Student (Social Psychology) Arizona State University. Degrees: B.S. (Psychology), B.A. (Spanish) University of North Carolina at Chapel Hill. Specializations: Cyberpsychology, culture, and decision-making.

HUSSONG. ANDREA M. Ph.D. Address: University of North Carolina at Chapel Hill CB#3270. Chapel Hill, NC 27599-3270. Title: Director, Center for Developmental Science, Professor of Psychology. University of North Carolina at Chapel Hill. Specializations: Developmental risk of land resilience to substance use.