4. Survey Measures
The questionnaire, a broad health risk and behaviors survey, was designed with numerous sections to concentrate on areas such as substance misuse, sexual activity, use of pornography, relationships, personal religious practices, and social support. The survey consisted of 124 questions and was administered using a paper and pencil close-ended question format.
4.1. Dependent Variables
The dependent variable is alcohol use and is defined as the frequency of alcohol use within the year prior to the administration of the questionnaire: “Within the last year about how often have you used alcohol”. The question included seven response categories (never; tried it once; once/month; twice/month; once/week; 3 times/week; daily). Prior year use was chosen since it would relate most closely with the relational independent variables.
4.2. Independent Variables
Independent variables included religious involvement and social support. Christian Religious Internalization Scale (CRIS) (Department of Psychology, New York, NY, USA) is a validated scale developed by Ryan et al. (1993), that was used to measure religious involvement. Religious involvement was defined as participation in both public and private displays of religiosity. The survey question asked “How often do you participate in the following activities listed below”, with public displays measured through: “Attend church services”; “Attend Sabbath School” (similar to Sunday School); and, “Attend school-sponsored religious programs”. Private displays were measured through: “Personal prayer”; “Read the Bible (outside of class assignments); “Read Seventh-Day Adventist literature outside of class assignments”; and “Family worship”. Responses were marked on a nine-point scale (never; less than once a year; about once or twice a year; several times a year; about once a month; two to three times a month; nearly every week; every week; and, several times a week).
4.3. The Multidimensional Scale of Perceived Social Support (PSSS)
The Multidimensional Scale of Perceived Social Support (PSSS) (Department of Pediatrics, Indianapolis School of Medicine, IN, USA) is a validated 12-item instrument designed to assess perceptions about support from family, friends and significant others. The scale was developed by Zimet et al. (1988). The items are divided into factor groups relating to the source of support, with choices ranging from ‘a lot, some, a little and none’. Sample items included “How many close friends do you have (people you feel at ease with, can talk to about private matters, and can call on for help” and “How many faculty/staff you can discuss sensitive issues like drugs/sex with?)”. Cronbach’s Alpha on the items in the PSSS = 0.80.
6. Results
The present study is based on an analyses of data collected from 334 college students enrolled at one university. As shown in
Table 1, more than half the students (60.2%) were female and single (82.7%). Nearly one-third (30.3%) were sophomores and juniors and almost half (49.2%) of the students had a B average grade. Seventh-Day Adventist was the dominant religion (94.4%), followed by those without a religious preference. White/non- Hispanics comprised the largest ethnic group (45.1%), followed by African Americans (19.4%), Latino (16.1%), Asian/Pacific Islander (13.5%), and West Indian (5.9%). Ages ranged from 18 to 63 years. Most participants were young adults (M = 21.79, SD = 5.55) and the majority (61.9%) were aged 18–22 years. The socio-demographic characteristics of the sample approximated those of the wider student population. For instance, the race/ethnic composition of the university population included Whites (37.0%), African Americans (18.3%), Asian/Asian American (13.2%), Latino/a (14.3%), and other groups (11.2%). At the time of the study, the student population was 59.1% female, and 90.2% of students identified as Seventh-Day Adventist. Of interest, the majority of students (51.5%) lived on campus, and those who wished to consume alcohol would normally car pool to a setting where alcohol was consumed. The nearest liquor store was 20 miles from the university.
Please see Frequency of Alcohol Users and Nonusers by Demographics in
Table 2.
What is the relationship between frequency of alcohol use and gender, marital status, student class, GPA, religion, ethnicity and age?
A chi-square test of independence was used to see if there was a relationship between various categorical variables. This test examined the relationship between frequency of alcohol use and gender, marital status, student class, GPA, religion, ethnicity and age. The relationship between frequency of alcohol use and gender was not statistically significant, X2 (3, N = 334) = 6.05, p = 0.109. The relationship between frequency of alcohol use and marital status was not statistically significant, X2 (3, N = 334) = 6.78, p = 0.079.
Participants’ reported use of alcohol did not differ by class standing; freshman, sophomore/junior, senior and grad/professional, X2 (9, N = 334) = 6.27, p = 0.712. The relationship between the frequency of alcohol use and GPA was statistically significant, X2 (5, N = 334) = 14.4, p = 0.026. In the religious category, the relationship between the frequency of alcohol use and religion was statistically significant, X2 (6, N = 334) = 23.8, p = 0.001, however, some cells are extremely small and as a consequence, the statistical test might not be reliable. For the ethnicity category, there was not a statistically significant relationship between ethnic groups and frequency of alcohol use, X2 (12, N = 334) =8.99, p = 0.703. The probability level might have been influenced by the way the variable was measured and the distribution of responses across cells, for example, just having 3 categories for ethnicity (white, black and other)
In the age category, there was a significant relationship between frequency of alcohol use and age X2 (3, N = 334) = 14.2, p = 0.003).
Table 3 presents the frequencies and percentages of responses to each item on the two scales which measured the independent variables. When students were asked “How many close friends do you have (people you feel at ease with, can talk to about private matters, and can call on for help?)” almost one-fourth (23.8%) indicated that they had at least three to five friends compared with those who said none (1.4%). When asked “How many faculty/staff can you discuss sensitive issues like alcohol/drugs/sex with?” almost one third (31.8%) of the students indicated none compared to 12.9% who said one or two. When asked, “How many relatives do you have that you feel close to?” 17.4% shared that they had three to five relatives that they feel close to. When asked “How many of these friends or relatives do you see at least once per month?” 17.1% indicated that they see one or two relatives at least once monthly compared to those who said none (4.2%). When asked “Do you belong to any social, recreational, work, church or other community groups? (For example, social clubs, exercise groups, campus ministries or community service)?” almost one third (31.9%) indicated that they belonged to one to five groups. Finally, when students were asked “How many close friends do you have (people you feel at ease with, can talk to about private matters, and can call on for help)?” about (23.8%) indicated that they had at least three to five friends compared with those who said none (1.4%).
Table 4 presents the responses to the questions relating to religious involvement: namely, how often do you participate in activities such as; attending church services, personal prayer, read the Bible (outside of class assignments), family worship, attend Sabbath School, read Seventh-Day Adventist literature outside of class assignments and attend school-sponsored religious programs. Almost three quarters of the respondents (71.0%) indicated that they attend church nearly every week, every week or several times per week as compared to 1.8% who said “never.” Almost four-fifths (79.3%) indicated that they prayer nearly every week, weekly or several times per week as compared to 1.5% who said “never”. Almost one third (60%) indicated that they read their Bible nearly every week, weekly or several times per week as compared to 2.8% who said “never”. Almost one third (30.3%) indicated that they have family worship nearly every week, weekly or several times per week as compared to (21.7%) who said “never.” Almost one third (32.8%) indicated that they attend Sabbath School several times a year, about once a month or 2 or 3 times a month as compared to 21% who said “never.” 44.7% indicated that they attend school-sponsored religious programs nearly every week, weekly or several times per week as compared to 10.1% who said “never.” Just over a third (34.4%) indicated that they read SDA literature several times a year, about once a month or at least two to three times a month compared with 20.7% who said “never.
Are social support and religious involvement predictors of frequency of alcohol use?
Table 5 shows that Linear Regression was calculated to predict alcohol use based on two independent variables; social support and religious involvement. The model summary gave a measure of how well the overall model fits, and how well the predictor variables (social support and religious involvement) predicted alcohol use.
R is a measure of how well our predictions predict the outcome, but the square of
R was taken to get a more accurate measure.
R2 gives the amount of variance in frequency of alcohol use explained by social support and religious involvement. A significant equation was found (F = 2333) = 33.464,
p < 0.000 with an
R2 of 0.167. Therefore, we can say that 16.7% of the variance in the dependent variable can be explained by the two independent variables. The analysis shows that social support did not significantly predict alcohol use, (ß = 0.027, t (335) = 0.536,
p < 0.05), however, religious involvement did significantly predict alcohol use, (ß = −0.413, t (335) = −8.115,
p > 0.05). These findings do not imply causality, because this was a cross-sectional study and not a longitudinal study.
In
Table 6, we conducted multivariate analyses with four variables (Gender, Age, Ethnicity and GPA). We controlled for these four variables while looking at religious involvement and social support to find out how each affects alcohol use individually. The following was found: gender was a strong predictor of frequency of alcohol use and this was statistically significant (
p-value = 0.013, 0.001 respectively). Examination of students between the ages of 18–22 and students 23 and above, showed that both were statistically significant (
p = 0.000, 0.000 respectively), and were strong predictors of frequency of alcohol use. When comparing ethnic groups Whites, Hispanics and other ethnic groups (
p < 0.001) used alcohol less because of their religious involvement. African American (
p > 0.05), almost reached significance, but were the only group that was not a predictor of frequency of alcohol use. Another predictor of less frequency of alcohol use was GPA (
p value = 0.000).
7. Discussion
The purpose of this study was to examine whether religious involvement and social support played a role in reducing the frequency of alcohol use. The results of this study did support past research findings that consistently suggested that religious involvement is a protective factor against alcohol use in adults, and is related to lower frequency of alcohol use [
34,
35]. In one study, both religiousness dimensions and emotional social support were related to less frequent alcohol use, but mediation was not supported. This present study revealed that only religious involvement was a predictor for reduced frequency of alcohol use, and social support was not a predictor of lower frequency of alcohol use.
Based on the findings of this study, a major protective element in keeping college students from drinking alcohol is the strength of their religious involvement. At least in this study, social support did not provide a protective effect on avoiding alcohol consumption. These findings indicate that it is relevant to assess the degree to which religious practices and beliefs saturate various life purviews.
We must interpret these findings with caution because the sample was limited to students at one religiously conservative school (Seventh-Day Adventist). Given that, there are very few religious traditions that reject all alcohol use for their members it is difficult to extrapolate the role that protective role that religion plays in other teetotalling traditions. However, the present study does provide meaningful firsthand evidence that religious involvement (distinct from religiosity) may actually decrease the frequency of alcohol use within some populations.
The current study has other important limitations as well. First, the sample consisted of 45.1% Caucasian students as compared to 19.4% African Americans, and still lower percentages for other ethnic groups. Perhaps the role of religious involvement and its connection to social support may be different for ethnic minority groups, particularly those that adhere to non-Western religion. Future research on the relationship between alcohol use and religious involvement should consider the role of ethnicity and non-Western religions.
Alcohol normally plays a role in building social support networks in college settings. Because drinking behaviors are normative to college life [
36] social support from friends and colleagues in some cases may take the form of drinking in a group. Thus, social support may enhance drinking behaviors during this critical transition into adulthood. However, in this study, social support was not associated with either increased or decreased drinking behaviors Instead, religious involvement had an independent and primary effect, perhaps relating to the strength of the internalized beliefs around abstaining from alcohol that are deeply engrained into Seventh-Day Adventist church members.
One of the reasons for conducting this study was to identify risk and protective factors for alcohol use among one conservative, non-alcohol-using university. Future studies should include other teetotalling colleges and should further investigate different motivating factors for drinking as related to religious involvement and drinking behaviors.
Finally, although the hypothesis that religious involvement and social support would be related to alcohol use was not supported, this hypothesis might be more applicable in other less conservative populations or participants with a history of heavy alcohol use patterns.