• Directories | A to Z | Home

Religiosity and Perceived Social Support

Laya Hess, Melanie Hornbeck, Brita Johnson, Lyndel Owens, Violet Thor, and sixth author

 

Abstract. While many studies have examined social support among college undergraduates, we focus specifically on the relationship between social support and religiosity, examining students’ involvement in organized religious groups, measured by hours per week spent in religious group activity. Using an online survey of randomly selected undergraduates at a small, private Lutheran institution in the Midwest, we tested the following hypotheses: 1) Religious students have higher perceived social support; 2) Lutheran students have higher perceived social support than non-Lutheran students; and 3) For religious students, as group religious activity increases, social support increases. We found, however, that there was no significant relationship between religiosity and perceived social support.

 

In their recent book God is Back: How the Global Revival of Faith is Changing the World, John Micklethwait and Adrian Wooldridge (2009) argue that religion is returning to the public sphere around the world.  In the United States, a nation where the role of religion in society is both glorified and contested, a recent San Diego Union Tribune article quoted a Gallup poll: the average church attendance rate ranges from 58% to 24% (2006). As religion’s role in society waxes and wanes, both religious identity and religious participation follow. The relationship, however, between religiosity and social support remains of particular interest to social scientists.

Literature Review
Social support is a popular topic for researchers, especially in the college setting. There are many possible directions to take when looking at social support, including different forms of social support or ways that social support can affect college student behavior. Many researchers have characterized social support as having "instrumental" and "socio-emotional" elements (Ellison and George 1994; Menagi 2008). Instrumental social support is reactive in that support is sought to address a problem directly; for example, seeking tangible assistance, advice, and information. Conversely, socio-emotional support includes sympathy and emotional and moral support.
Studies focusing on social support agree that support is an interpersonal resource that one utilizes, often unknowingly, to maintain or improve psychological well-being. Though support and comfort can be gleaned from many sources, social support generally refers to an interpersonal relationship between individuals within a readily available network characterized by close personal relationships, shared interdependence and identification with common values (Bohus, Woods, and Chan 2005). Acquiring social support can relieve an individual of anxiety or isolation and situate them within a community.
Several researchers have examined different types of social support and how they interact with one another. Rutger and Engels (2002) considered the connection between parent-child relationships and peer-peer relationships in adolescents in the Netherlands, while Dennis, Phinney, and Chuateco (2005) examined the role of familial support in the academic success and social integration of first-generation college students.
Other studies have investigated the distinction between social support received and social support given, and the interactions between these two categories. Several studies have shown a correlation between the amount and quality of support received and support given (Baus, Dysart-Gale, and Haven 2005; Gant, Calsyn, and Winter 1999), though others have shown this correlation to be contingent on perceived responsiveness and/or the appearance of “visible” and “invisible” support (Maisel and Gable 2009).
Research has also examined social support networks and college student behaviors including stimulant use and alcohol consumption. Although these behaviors are usually viewed as deviant (especially for underage college students), they have also been shown to have strong social implications. In a study of collegiate stimulant abuse, researchers found that stimulant use not only helped students cope with life stressors, but also made them feel as if they “belonged” (Hall, Irwin, Bowman, Frankenberger, and Jewett 2005).
To further explore how social support can influence health-related behaviors, researchers have paid close attention to the social aspects of disordered eating. A study by Tiller, Sloane, Schmidt, Troop, Power, and Treasure (1997) showed that clinically diagnosed anorexia nervosa and bulimia nervosa patients received significantly less social support than did a control group of healthy students. This research may point to correlations between levels of social support and risks for disordered eating.
Recent studies have explored the relationship between social support and the internet. Networking websites, e-mail, and blogs have changed the type of communication available to most, and have even been shown to be related to well-being, self-esteem, and levels of extraversion (Subrahmanyam and Lin 2007). For example, one study’s findings support “the rich get richer” model, suggesting that extraverts are more likely to use the internet and have higher perceived levels of social support than introverts, who are more likely to be lonely (Kraut, Kiesler, Boneva, Cummings, Helgeson, and Crawford 2002).
As social support emerges as an important factor in the college experience, researchers have studied college students’ group membership and its subsequent effect on their satisfaction with social support. Involvement in clubs, sororities, teams, and other groups enhance well-being and can help with social adjustment in college (Garcia 2005). 
Group participation, a central element of students' social support, often takes the form of involvement in organized religious groups and activities (Bonhert, Aikins, and Edidin 2007). Though most social support places the individual within the community, religious social support can involve the individual in a community with similar beliefs and values. Bryant, Choi, and Yasuno (2003) identify group participation as part of both religion and spirituality, indicating that these sources of social support affect students in a multitude of ways, including self-reported degrees of spirituality, attendance of religious services, discussion of religion, and prayer and meditation.
Researchers focusing on the relationship of religion and social support found that students with stronger religious commitment, based on frequency and type of religious activities, had a stronger social network of relationships (Mankowski and Thomas 2000). Bohus, et al. (2005) found that students who scored higher on scales of religious commitment also had higher scores on psychological sense of community.
The results from Knox, Langehough, Walters, and Rowely (1998) and Hill's (2009) studies echoed the conclusion of Bohus et al. (2005) that there is a positive association between perceived social support and religious commitment. Although religious involvement and social support often share a positive correlation, they are often treated as two separate factors, rather than as religious social support. Hill (2009) acknowledged that institutional influences may impact results, as religious institutions may attract students who are already religious and find that their social support is built into the institution.
Increased religious involvement expands the scope of social support (Ellison and George 1994), easing the transition from high school to college and strengthening family relationships (Mulenga, Needham, Schludermann, and Schludermann 2001). Bradley (1995) demonstrated that these findings can be extrapolated to a larger population, by replicating the Ellison and George study (1994) but using nationwide sample of adults instead of a sample from a single community. Experiencing major life changes also impacted a student's ability to seek religious social support (Dennis and Muller 2007). However, other studies found no correlation between religious participation and transition and adjustment (Hunsberger, Pratt, and Pancer 2001).
Students on college campuses participating in religious social support networks have tended to report better than average emotional and physical well-being. Frankel and Hewitt (1994) demonstrated that students involved in religious groups reported better physical health, greater perceived self-esteem, and lower perceived stress. Students with strong emotional religious support networks also abused alcohol less than others (Menagi, Harrell, and June 2008).
The most relevant information for our research comes from Ellison and George’s (1994) study examining the relationship between religious participation and perceived social support among adults in North Carolina. This study found a positive correlation between religious participation and the quality and quantity of perceived social support. Ellison and George (1994) measured quality as the extent to which people felt cared for and valued within their relationships, while they measured quantity of social support by the amount of interactions between two people of a given religious network. Furthermore, they measured quality of social support by asking questions on a scale of one to three about how supported participants felt within their relationships. In order to do this, researchers asked respondents questions on topics such as how often they felt useful, listened to, or could count on others for help. Measuring religious involvement by having people estimate how often they attend “church or other religious meetings” provided an easy and straightforward way of assessing frequency of attendance (Ellison and George 1994). This method provided a concrete way to determine how supported participants felt by their networks.
The limitations of this study included an overrepresentation of the elderly in the sample and a sample that only included people from one part of the country. Thus, a possible confounding variable was the amount to which people in this region relied on religious networks. Although Bradley (1995) showed that Ellison and George’s (1994) results can be generalized to the rest of the United States, there is a considerable age discrepancy between the elderly and college students, most importantly in their sources of social support. College students living on a residential campus like St. Olaf may have social groups that provide more support than religious institutions.
Our research focuses on the effect of religiosity, shown through participation in organized religious activities, on perceived social support experienced by college students. As Bradley (1995) found that religious individuals perceived a greater quality and quantity of social support, we want to examine the relationship between religiosity and perceived social support among college students.

Methods
Our research was conducted in the fall of 2009 using a survey questionnaire. Survey research is a way by which data can be collected from a large sample and generalized to a larger population (Neuman 2007). Thus, it is the most efficient method of data collection regarding the application of our results to the St. Olaf student body. For our research, survey questionnaires were distributed using the online program “Form Creator.” After choosing a sample, we sent each selected student an email requesting their participation, with a link to the survey. 
We had three hypotheses: 1) Religious students have higher perceived social support than non-religious students; 2) Lutheran students, as members of the largest religious denomination on campus, have higher perceived social support than non-Lutheran students; and 3) For religious students, there is a positive correlation between hours spent in religious group activity and perceived social support.
Variables
For each of our three hypotheses an independent and dependent variable was identified and then tested. Religiosity was the independent variable for our first hypothesis, while social support was the dependent variable. In our second hypothesis, religious denomination was the independent variable and social support was the dependent variable. In our third hypothesis, total time spent in religious activity was the independent variable and social support was the dependent variable.
To gauge religiosity, we asked respondents if they had participated in any religious group activities during the current semester. They were then asked to estimate how much time (in hours) they spent in their activities on an average week. Survey categories included: sacred text study, organized prayer and group meditation, worship service, musical religious groups or organizations (not including any choirs in the St. Olaf music department), non-musical religious groups or organizations, faith-based volunteering, and other religious group activities, in which they were asked to further specify. We focused on religious group participation since we wanted to investigate the connection between religious involvement and perceived social support.
To assess social support we used a four-point Likert scale (Neuman 2007). Participants were asked to respond to ten statements using “Strongly agree,” “Agree,” Disagree,” and “Strongly Disagree.” Each statement was an indicator for a facet of our definition of perceived social support, including tangible support, intimacy, sense of belonging, self-disclosure, and availability (Hale, Hannum, and Espelage 2005). The statements were: “I know someone who would lend me $20 to shop at Target”; “My friends do not seek me out to spend time with them”; “I feel like I am part of the St. Olaf College community”; “I have someone in whom I can confide personal information”; “The people I’m closest with truly care about the current state of events in my life”; and “I know someone who is available to talk with me at any time.” Since we tested components of our definition of perceived social support independently, we were able to more precisely identify which areas of social support correlated with religiosity.
We measured religious denomination, an independent variable, with a simple question asking participants to self-identify from a list of denominations and to write in their denomination if it was not listed. The listed denominations, chosen based on St. Olaf student demographics and on-campus religious groups, were agnostic/atheist, Baha’i, Buddhist, Catholic, Hindu, Jewish, Muslim, Protestant (Lutherans), and Protestant (non-Lutherans), and Other. In order to increase reliability for our study, we followed a four-step process outlined by Neuman (2007). This process includes conceptualizing concepts clearly, using a precise level of measurement, having multiple indicators for each variable, and conducting pilot tests. Since we clearly conceptualized religiosity and social support, answers were contained to our construct. By measuring religiosity in hours, we increased the precision of our measurement using concrete numbers.
The Hale, et al. (2005) model for social support uses multiple dimensions of social support, including tangible support, sense of belonging, intimacy, and self-disclosure. We also included availability of support as an indicator of one’s social support. Our ten-part index expands on these indicators which allows for a broad construct and reliable results. 
              We also wanted our survey to meet Neuman’s (2007) criteria for measurement validity in its various forms. Because of this, we wanted the survey to measure our conceptual definitions of religiosity and social support as comprehensively as possible. Neuman (2007) defines three types of validity that must be achieved: face validity, where the indicator clearly appears to measure the construct; content validity, where all parts of the conceptual definition are measured; and criterion validity, where our results should agree with an external source.
To increase face validity, we conducted pilot tests where participants (other researchers) could comment on our measures. Since face validity is based mostly on consensus of the scientific community, this method was the best way to enhance it. For content validity, we made sure to include two items for each part of our conceptual definition of social support. We also included exhaustive categories for our index of religious activity, so that respondents could indicate hours per week in a variety of activities.
Sample and Sampling Procedure
The data were collected in the fall of 2009 at St. Olaf College in the upper Midwestern United States.  The target population was college students in classes 2010 to 2013. We excluded students in the following categories: those who are under age 18, enrolled in our Sociology/Anthropology research methods course, currently studying abroad, participants in the focus group, and those who are not full-time students. These students were excluded to prevent bias, and many of the students studying abroad have limited computer access which hinders their timely response. Susan Canon, Director of the Office of Institutional Research at St. Olaf, generated an email alias using a simple random sampling technique. Simple random sampling is the most efficient way of providing a sample that accurately represents the population parameters (Neuman 2007).
            Our attempted sample size was 25% of the population at St. Olaf after our exclusions (about 2,813 students). According to Neuman (2007), with a smaller population (under 1,000 people) one should use a sample size of about 30%, but since the population under study was a little larger than that, we do not require quite as large a sample size (2007). The larger the population, the smaller the sample size, and because our population was relatively homogeneous, we could have a goal sample size of 25% of the people.
            The number of respondents for the survey was 333, meaning the response rate was 47.4%. The characteristics of the sample were 32.2% (104/322) male, 67.1% (216/322) female, and 0.6% (2/322) other. In terms of race, 90.9% (292/321) identified as White Non-Hispanic, 0.3% (1/321) as Black or African-American, 2.2% (7/321) as Hispanic, 3.7% (12/321) as Asian, and 3.3% (9/321) as Other. No respondents identified as American Indian or as Native Hawaiian or Pacific Islander. The survey was comprised of 27.0% (90/321) first years, 22.2% (74/321) sophomores, 21.3% (71/321) juniors and 25.2% (84/321) seniors, and 0.6% (2/321) who identified as other. 
Ethics
            The main ethical issues faced were privacy, protection from harm, voluntary participation, ensuring the parties were informed and protecting potentially vulnerable groups. There was a potentiality for psychological harm and offense as a result of answering personal questions regarding religion, spirituality, and social support. By asking students to answer personal questions regarding religiosity and social support, we risked potentially offending participants. These risks were neutralized by the voluntary nature of the survey and the incentives offered. Because we generated a random sample of students who met certain criteria (over 18, not participants of focus groups, not fellow researchers of SOAN 371), we avoided exploitation, exclusion or overprotection. As the survey sample was generated electronically, respondents’ answers were anonymous and their identities hidden, further ensuring privacy. The cover letter explained the survey’s goals and risks, and reinforced the voluntary nature of participation, helping to ensure informed consent. Conducting this study in an ethical manner aided researchers in obtaining fair and accurate data.
            As this study involved gathering and analyzing information about people, it was subject to approval by St. Olaf’s Institutional Review Board (IRB) and had to comply with the St. Olaf College Code of Ethics for Projects with Human Subjects. For such studies, St. Olaf IRB requires researchers to maintain an appropriate risk/benefit ratio, appropriate procedures for subject selection, appropriate protection of privacy and confidentiality, and appropriate provisions for obtaining informed consent (St. Olaf College Code of Ethics, 2004).

Results
We measured students' religiosity based on hours per week spent in religious group activity. Of the 320 respondents who answered the question “During this current semester, have you participated in any religious group activities (on campus or off campus)?”, approximately 32.8% (105/320) participated in religious group activities, and approximately 67.2% (215/320) did not participate. Of respondents who participated, the mean time spent in religious group activity was 3.23 hours per week. 
To measure perceived social support, we created an index using indicators of tangible support, self-disclosure, intimacy, sense of belonging, and availability. Higher scores on the index indicated higher perceived social support, while lower scores indicated lower perceived social support. Participants reported on a scale of 0-30. The mean social support score was 25.68, the median was 27, and the mode was 30 (see Figure 1). 

Figure 1: Total Perceived Social Support

  

 

 

 

 

 

Of our respondents, 34.9% (111/320) identified as Protestant-Lutheran, 24.4% (83/320) as agnostic or atheist, 17.5% (65/320) as Protestant non-Lutheran, 11.3% (34/320) as Catholic, 1.3% (4/320) as Buddhist, 0.6% (2/320) as Jewish, and 0.3% (1/320) as Muslim (See Figure 2). There were no respondents who identified as Baha’i or Hindu. The proportion of respondents who identified as “Other” was 6.3% (20/320). In total, 34.9% identified as Lutheran (111/320) and 65.1% (189/320) identified as non-Lutheran.

Figure 2: Religious Denomination Percentages (excluding Other)

 

Hypothesis 1: Religious students have higher perceived social support than non-religious students.
To assess our first hypothesis, we used a Mann-Whitney U test for nonparametric variables to examine the difference in perceived social support between religious and non-religious students. Because our social support results were heavily skewed negatively, and thus not normally distributed, we only used tests for nonparametric variables.  We found no significant difference in the results of perceived social support between the two groups (U = 9370, p > .05). Religious students had a mean score of 26.22, and non-religious students had a mean score of 25.46. The significance value falls below the 95% confidence level and therefore is not statistically significant and cannot be generalized to the larger St. Olaf population (see Table 1 and Figure 3). However, the results are significant at the 90% confidence level, suggesting a trend, even if no significant relationship was found.
Table 1: Mann-Whitney U Test Results for Religious v. Non-Religious Students

 

 Figure 3: Total Perceived Social Support for Religious v. Non-Religious Students

 

 

 

Hypothesis 2: Lutheran students have higher perceived social support than non-Lutheran students.
Of the sample, 34.9% (111/320) identified as Lutheran and 65.1% (189/320) identified as non-Lutheran. We used the Mann-Whitney U test to examine the difference in perceived social support for Lutherans and non-Lutherans. No significant difference was found in the results (U = 9839, p > .05) (See Table 2). Lutherans had a score of 25.68, and non-Lutherans had a score of 25.64. The significance value showed no relationship between Lutherans and non-Lutherans and perceived social support (see Figure 4). 
Table 2: Mann-Whitney U Test Results for Lutherans v. Non-Lutherans

 

 

 Figure 4: Total Perceived Social Support for Lutherans v. Non-Lutherans

 

 

 

 

 

 

Hypothesis 3: Among religious students, as hours spent in religious activity increases, perceived social support increases.
To test our final hypothesis we used a Spearman rho correlation coefficient for the relationship between religious activity and perceived social support. We found a weak correlation that was not significant (r = -.086, p > .05). Total religious activity was not related to perceived social support (see Table 3).  

Table 3: Spearman's rho Correlation Coefficient for Religious Activity and Social Support

Correlations
			Total Perceived Social Support	TotRelInd2
Spearman's rho	Total Perceived Social Support	Correlation Coefficient	1.000	-.086
		Sig. (2-tailed)	.	.393
		N	320	101
	TotRelInd2	Correlation Coefficient	-.086	1.000
		Sig. (2-tailed)	.393	.
		N	101	104

 

Discussion
The results of the survey do not support our first hypothesis that religious students have higher levels of perceived social support than non-religious students. Instead, students who said they were religious reported similar amounts of support to students who said they were not religious. These results suggest that at St. Olaf, the main sources of social support for students come from outlets other than religious groups. The high levels of social support reported by St. Olaf students in general may account for these results. Survey respondents reported generally high levels of support, and thus the data do not indicate a difference between the social support of religious and non-religious students.
The survey results also do not support our second hypothesis, that Lutheran students have more social support than non-Lutheran students, and rather suggest that the two groups feel similar levels of social support. This runs contrary to results in the Knox et al. study (1998), where students who identified with the religious denomination of the college felt more supported than those who identified with other denominations or as non-religious. Differences in the proportions of religious denominations from the sample to the larger college population may account for these results.
Our last hypothesis, that among religious students, as hours spent in religious group activity increases, their perceived sense of social support increases, was also not supported. There was not a significant relationship between hours spent in religious group activity and the student’s level of social support. These results also support the idea that religious groups are not the sole provider of social support for students.
Possible alternative explanations of our results include an inaccurately designed index for social support. Since most students reported high levels of social support, we may need to further differentiate among the higher levels in order to gain significant results. Another possible explanation is imprecise representation of social support due to social desirability bias, as respondents may have been influenced by the perceived expectations of the survey, or the institutional emphasis on community at St. Olaf College.
            While our research found no significant relationship between religious participation and social support, previous research that measured religiosity (religious activity in groups) separately from spirituality (individual religious activity) did find significant relationships (Knox et al. 1998). If we had included individual activity in our survey, we might have found different results. Further, some studies show that the amount of religious activity changed when students entered college (Bryant et al. 2003; Hunsberger et al. 2001). If students had a strong religious social support network at home, maybe they did not feel a need or desire to participate as much in on-campus religious groups.

Conclusion
Our study examined the relationship between religiosity and social support among undergraduates at St. Olaf College. Considering the college’s reputation as a supportive community and its affiliation with the ELCA, we hypothesized that religious students have higher perceived social support than nonreligious students, that Lutheran students have higher perceived social support than non-Lutheran students, and that among religious students, as time spent in religious group activities increased, perceived social support also increases. None of our three hypotheses were supported.
            Previous studies generally agreed that people involved in religious group activities often report having stronger social support networks, better physical and mental well-being, and higher psychological sense of community (Ellison and George 1994; Frankel and Hewitt 1994; Bohus, et al. 2005). While our study did demonstrate that religious students had high perceived social support, it also found that non-religious students had high support. In fact, students overall reported extremely high levels of support regardless of their religious participation or religious affiliation. Thus, our findings differed from previous studies in that there was no significant relationship between religiosity and social support among college undergraduates.
            Our results are useful to the college administration regarding integration of all students, regardless of religiosity, into the St. Olaf College community. Even though St. Olaf College is affiliated with the Evangelical Lutheran Church of America, our data suggest that non-religious students tend to feel just as supported as religious students on this campus, and the college can use this information to continue to reach out to students of all religious (or non-religious) backgrounds.
            Although the sample primarily consisted of females and Caucasians, it closely represented the demographic composition of the St. Olaf College student body. The similar demographics are acceptable since results gathered from our sample can only be generalized to the St. Olaf College population.  Over-reporting, a limitation sometimes due to social desirability bias, is potentially problematic but its occurrence is unverifiable. However, surveys are still the most appropriate method with which to measure our data. Also, our definition of religiosity was very narrow, only examining respondents’ temporal involvement in religious group activities, and thus, we did not include respondents’ religious beliefs or values. This definition did allow us to quantify religious involvement, which was very useful for the purposes of this study, but in the future, researchers might look at a broader definition of religiosity. Future research might also explore different indicators of social support than the five indicators used in this study (tangible support, intimacy, self-disclosure, sense of belonging, and availability). It would be worthwhile to conduct this research in a number of different tertiary institutions, including large, urban universities featuring more diverse populations in terms of gender and race.

References

Baus, Ray, Deborah Dysart-Gale, and Phil Haven. 2005. "Caregiving and Social Support: A Twenty-first Century Challenge for College Students." Communication Quarterly 53:125-42.
Bohnert, Amy M., J. W. Aikins, and J. Edidin. 2007. “The Role of Organized Activities in Facilitating Social Adaptation Across the Transition to College.” Journal of Adolescent Research 22:189-208.

Bohus, Steve, K. Caleb Chan, and Robert H. Woods, Jr. 2005. “Psychological Sense of Community among Students on Religious Collegiate Campuses in the Christian Evangelical Tradition.” Christian Higher Education 4:19-40.

Bradley, Don E. 1995. "Religious Involvement and Social Resources: Evidence from the Data      Set ‘Americans’ Changing Lives’” Journal for the Scientific Study of Religion 34:259 267.
Bryant, Alyssa N., Jeung Yun Choi, and MaikoYasuno. 2003. “Understanding the    Religious and Spiritual Dimensions of Students' Lives in the First Year of College.” Journal of College Student Development 44(6):723-745.
Dennis, Dixie L., and Susan M. Muller. 2007. “Life Change and Spirituality Among a College          Student Cohort.” Journal of American College Health 56:1.

Dennis, J.M., J.S. Phinney, and L.I. Chuateco. 2005. “The Role of Motivation, Parental Support, and Peer Support in the Academic Success of Ethnic Minority First-Generation College Students.” Journal of College Student Development 46.3:223-236.

Ellison, Christopher G. and Linda K. George. 1994. "Religious Involvement, Social Ties, and         Social Support in a Southeastern Community.” Journal for the Scientific Study of          Religion 33:46-61.

Frankel, B. Gail, and W.E. Hewitt. 1994. “Religion and Well-Being among Canadian University Students: The Role of Faith Groups on Campus.” Journal for the Scientific Study of Religion 33(1):62-73.

Gant, Judy, Robert Calsyn, and Joel Winter. 1999. “Predicting Agreement Between Providers and Recipients of Social Support.” Journal of Social Behavior & Personality 14:4.

Garcia, Gina Ann. 2005. “The Relationship of Perceptions of Campus Climate and Social Support to Adjustment to College for Latina Sorority and Non-sorority Members.” Unpublished manuscript.

Hale, Cara, James W. Hannum, and Dorothy L. Espelage. 2005. “Social Support and Physical Health: The Importance of Belonging.” Journal of American College Health 53(6):276-284.
Hall, K.M., M.M. Irwin, K.A. Bowman, W. Frankenberger, and D. Jewett. 2005. “Illicit use of prescribed stimulant medication among college students.” Journal of American College Health 53(4):167-174.

Hill, J. 2009. “Higher Education as Moral Community: Institutional Influences on Religious Participation During College.” Journal for the Scientific Study of Religion 48(3):515-534.

Hunsberger, B., M. Pratt, and S. Pancer. 2001. “Religious Versus Nonreligious Socialization:
Does Religious Background Have Implications for Adjustment?” International Journal for the Psychology of Religion 11(2):105-128.

Knox, David, Steven O. Langehough, Conner Walters, and Michael Rowley. 1998. “Religiosity and spirituality among college students.” College Student Journal 32(3): 430-441.

Kraut, Robert, Sara Kiesler, Bonka Boneva, Jonathon Cummings, Vicki Helgeson, and Anne Crawford. 2002. “Internet Paradox Revisited.” Journal of Social Issues 58:49-74.

Maisel, Natalya, and Shelly Gable. 2009. "The Paradox of Received Social Support: The Importance of Responsiveness." Psychological Science 20:928-932.

Mankowski, E. S., and E. Thomas. 2000. “The relationship between personal and collective identity: A narrative analysis of a campus ministry community.” Journal of Community Psychology 28(5):517-528.

Menagi, Feyza S., Zaje A. T. Harrell, and Lee N. June. 2008. “Religiousness and College
Student Alcohol Use: Examining the Role of Social Support.” Journal of Religion &
Health 47(2):72-98.

Micklethwait, John, and Adrian Wooldridge. 2009. God is Back: How the Global Revival of Faith is Changing the World. New York: Penguin.

Mulenga, Morgan, Doug Needham, Eduard H. Schludermann, and Shirin M. Schludermann.         2001. “Fear of Rejection versus Religious Commitment as Predictors of Adjustment
Among Reformed and Evangelical College Students in Canada.” Journal of Beliefs and Values    22:2.

Rutger, C., and M.E. Engels. 2002. “Parenting Practices, Social Skills and Peer Relationships in Adolescence.” Social Behavior and Personality 30(1):3-18.

Subrahmanyam, Kaveri, and Gloria Lin. 2007. “Adolescents on the net: Internet use and well-being.” Adolescence 42(168).

Tiller, Jane M., Gaynor Sloane, Ulrike Schmidt, Nicholas Troop, Michael Power, and Janet L. Treasure. 1997. “Social support in patients with Anorexia Nervosa and Bulimia Nervosa.” International Journal of Eating Disorders 21:1.

2006. “Church or synagogue attendance by state.” San Diego Union Tribune, May 2. (Retrieved from http://legacy.signonsandiego.com/uniontrib/20060502/news_lz1n2thelist.html on December 17, 2009.)