The Intersection of Race and Spirituality: Underrepresented Students' Spiritual Development at Predominantly White Evangelical Colleges

Abstract

Although a great deal of research assesses the value of both diversity and spirituality in the university setting, little research addresses the experience of underrepresented students and their spiritual development in an environment that is defined by a mission committed to spiritual growth. Utilizing data from the College Students' Beliefs and Values survey, a longitudinal dataset from the UCLA Spirituality in Higher Education project, this study explored the racial differences in students' spirituality at faith-based institutions. Amongst all students, seniors demonstrated significant growth on the spiritual identification and ethic of caring scales, and their religious commitment decreased during college. White students scored significantly higher on the religious commitment scale than non-white students, and non-white students scored significantly higher on the ethic of caring scale.

Evangelical schools place spiritual growth as their highest educational objective. Many researchers agree that, in general, the current state of education precludes a focus on individual student development (Astin, 2004; Chickering, Dalton, & Stamm, 2006; Gallagher, 2007; Subbiondo, 2005). Astin (2004) articulates a bifurcated relationship between academe and internal development—the dialogue of meaning, truth, spirituality, and personal calling within higher education has been replaced with the more practical discussions of business calculus, mass communication law, and cognitive psychology. While these discussions may be essential for a student's future career, “our heavy concentration on objectivity and empirical rationality and on professional and vocational preparation indeed can work against encouraging authenticity and identity, integrity, and spiritual growth” (Chickering, Dalton, & Stamm, 2006, p. 26). However, the idea is not missing at evangelical colleges that college should be a place for students to expand their mind with practical career-oriented courses, while also developing the interior aspects of their lives (Astin, 2004). Hundreds of private, evangelical schools note a more narrow definition of spiritual growth as a primary goal. Conversely, while many of these schools excel in their commitment to spiritual development, they are quite homogenous in makeup (Paredes-Collins, 2009).

Although significant research has found that a diverse learning community yields positive outcomes for students of all backgrounds (Allen, 2006; Bok, 2006; Chang, 1999; Hurtado, 2007, 2006; Hurtado et al., 1998; Ibarra, 2001), many evangelical colleges do not identify the development of a diverse learning community as an institutional priority (Paredes-Collins, 2009). As such, evangelical institutions intentionally engage students in spiritual growth and development, but lack emphasis on promoting a diverse learning community. In terms of structural diversity, an evaluation of 101 evangelical universities (all of which were members of the Council for Christian Colleges & Universities) revealed a conglomerate picture of the student body that is 84% white (Paredes-Collins, 2009).

While a great deal of research assesses the value of diversity and spirituality in the university setting, the gap in the literature concerns the intersection between race and spirituality at predominantly white evangelical schools. Specifically, there is a dearth of research about the experience of underrepresented minority students and their spiritual development in an environment that is defined by a mission committed to spiritual growth. As many of these institutions have so few students of color, it is important to understand how the spiritual growth for this group compares to those who are not in the minority of campus culture. The following research questions guided this study: 1) Are there any significant racial differences in seniors' spirituality at faith-based schools in the CCCU, after controlling for their level of spirituality as a freshman? 2) Are there any significant racial differences in students' spirituality at CCCU schools? Given the homogenous nature of CCCU schools, the expectation was to find significant variance among races in some aspect of spirituality. In consideration of the research on white conservatism and pro-social interests of nonwhites, (Brown, 2009; Kelley, 1995), the expectation was also to find that students of color were more prone to the social-justice oriented aspects of faith and spirituality.

Review of the Literature

Spirituality in Higher Education

To clarify the distinction between spirituality and religiosity, religion involves an association with and commitment to an established set of customs and rituals, whereas spirituality is characterized by an individual pursuit for meaning, purpose, truth, and value, however elusive the quest may be (Chickering, Dalton, & Stamm, 2006; Zinnbauer, Pargament, & Scott, 1999). Someone could be distinctly spiritual and not religious, religious and not spiritual, neither, or both (Bryant, Choi, & Yasuno, 2003). Spirituality integrates an amalgam of internal awareness and outward focus (Astin, 2004; Astin & Keen, 2006; Chickering, Dalton, & Stamm, 2006; Gallagher, 2007; Harris, Thoreson, & Lopez, 2007). This paradoxical awareness of intrinsic value and meaning that is juxtaposed with an outward focus on others can dramatically impact quality of life and positively influence personal development, leading to a greater understanding of individual purpose, vocation, and meaning (Harris, Thoreson, & Lopez, 2007; Astin, 2004). For the purpose of this discussion, both spirituality and religiosity will be referenced.

In a national study of students' beliefs and values, Astin (2004) found that two-thirds of college students identify that it is “essential” or “very important” for their college to enhance self-understanding. Students also identify a desire for their college to play a role in developing personal values. Approximately 50% of all students say that it is “essential” or “very important” that colleges encourage a personal expression of spirituality, and 58% of students place great value on “integrating spirituality into [their] life,” (Astin, 2004, p. 38). Although these numbers clearly suggest that students have a great interest in spirituality, many universities do not offer students the opportunity to explore spirituality in the context of their student experience.

While spirituality has positive ramifications for the general population, its ability to promote development among college students is even more dramatic. An expanding collection of research has shown that spirituality is associated with college engagement (Kuh & Gonyea, 2005), service involvement (Astin, 2002; Kuh & Gonyea, 2005; Chickering, Dalton, & Stamm, 2006), leadership (Yasuno, 2004), and identifying purpose in life (Greenway, 2005). According to Astin (2004), both public and private institutions of higher education would benefit from engaging their students in a discussion of spirituality, internal development, and personal vocation. Encouraging students to grasp an understanding of their intrinsic value, their connection to others, and their ability to impact humankind through their actions could have a dramatic impact on the world they inhabit (Chickering, Dalton, & Stamm, 2006). Through such engagement, students can understand the power of forgiveness and their ability to affect their own situation through peaceful equanimity (Astin & Keen, 2006). Further, extending beyond student spirituality, Lindholm and Astin (2008) found that faculty's spirituality emerged as a significant predictor for engaging in student-centered pedagogical practices, after controlling for faculty demographics, institutional, and personal characteristics.

Although research has shown that spirituality has implications for both physical and mental health (Harris, Thoreson, & Lopez, 2007; Hill & Pargament, 2003), little research has focused on the possible implications for spiritual struggle. Bryant and Astin (2008) note that while most of the research regarding the relationship between health and spirituality focus on the positive implications, the studies have utilized rather unsophisticated definitions of religion and spirituality. Most of the research on spirituality and college students does not assess the outcomes of spiritual disequilibrium, instead focusing on the positive outcomes of spirituality and religion. Bryant and Astin (2008) were among the first to explore the negative effects of spiritual struggle on the well-being of college students. Defined as the “intraphysic concerns about matters of faith, purpose, and meaning of life” (p. 2), Bryant and Astin (2008) found that students indeed experience spiritual struggle. Female students, psychology students, students who had a difficult experience, and students whose worldview has been challenged were more likely to experience spiritual struggle. Further, spiritual struggle was significantly associated with higher levels of psychological distress and lower levels of physical health, well-being, and self-esteem. Contrary to previous research (Chickering & Reisser, 1993; Holcomb & Nonneman, 2004; Pargament et al., 2005), Bryant and Astin (2008) did not find a significant relationship between spiritual struggle and spiritual growth. The only positive outcome of spiritual struggle was a greater propensity for religious tolerance. Bryant and Astin's work notes that students are likely to struggle spiritually in college, and schools would be remiss to ignore the serious implications of such experiences.

Evangelical Institutions

The milieu of higher education in the United States comprises a wide variety of institutional types. Approximately 19 million students attend 4,409 degree-granting, accredited public, private not-for-profit, and private for-profit institutions. Of these schools 900 are religiously affiliated, as reported by the institution. Over the past two decades, student enrollments at these private, religiously affiliated colleges and universities have steadily risen. With an enrollment of 1,285,409 in 1990 and 1,772,614 in 2008, religiously affiliated schools experienced an increase of 73% (U.S. Department of Education, NCES, 2009). Of these 900 schools, 106 are members of the CCCU, and 31 maintain affiliate membership (CCCU, 2010). Each of these faith-based institutions implements myriad different programs to integrate spirituality into curricular and co-curricular activities. Compulsory chapel programs, Bible studies, informal small groups for accountability or discussion, campus ministry programs, spiritual life advisor programs, formal religion or ministry courses, and opportunities to engage in service learning, faculty mentorship, and community prayer are some of the ways these schools integrate spirituality and religion into the curricular and co-curricular.

Between 2002 and 2009, a private philanthropic organization called the Lilly Endowment awarded numerous grants to church-related liberal arts schools to interweave spirituality, faith and vocation, service, and mentorship opportunities into their curriculum and program offerings (Gallagher, 2007; Lilly Endowment, 2009). Of the 400 schools that applied for a Lilly grant, 88 Christian schools were each given approximately $2.5 million over a few years to create a variety of programs that would “encourage theological reflection and religious practices that recover the wisdom of the Christian tradition for our contemporary situation” (Gallagher, 2007, p. 32). The Lilly Grant provides a tangible illustration of how evangelical institutions integrate spirituality into the educational experience. Each institution had the flexibility to create programs and curricula that would have the greatest impact on their specific student population. Examples of programs supported by the Lilly Endowment include the following (Gallagher, 2007):

	Seattle Pacific University established SERVE (Spiritual Educational Resources for Vocational Exploration), an initiative that includes 16 different programs for students and faculty.
	Gordon College in Massachusetts established unique first-year courses called “Christianity, Character, and Culture” that include discussions of calling and vocation within the context of Christian faith.
	Wesleyan Asbury College in Kentucky created an honors senior seminar course entitled “Leading Lives of Significance” with the university president as the instructor.
	Creighton University in Nebraska: established a new living community for sophomore students where prayer, discussion about vocation, and spiritually focused study is encouraged.
	Santa Clara University: offered myriad opportunities for faculty, staff, and alumni to “stand in solidarity with the poor” through mission trips, cultural immersion programs, and service-learning projects around the world.

Lilly Grant recipients consistently report positive feedback regarding their individual programs. According to Gallagher (2007), students have shown an increase in “thoughtful reflection about their vocational choices in light of their faith commitments” (p. 36), a deeper understanding of their purpose in life, and an increased commitment to engage in ministry.

In contrast to the positive feedback regarding faith commitment reported by Lilly grant recipients, Smith's (1998) “subcultural identity” theory claims that religiously committed youth fare better in terms of religious development when they are in a non-religious environment as they are forced to rely on their own faith identity to cope with life. In a national study, Hill (2009) found otherwise. Hill's (2009) research showed that students attending Catholic and mainline protestant universities were more likely to decrease religious participation than those who attend evangelical schools. Further, evangelical students who attended Catholic institutions were more likely to decrease their religious involvement, while Catholic students attending evangelical institutions were likely to increase their religious involvement. In a finding that could relate to the propensity for spiritual development of underrepresented students at evangelical colleges, black Protestants were more likely to decrease in religious participation when they attend evangelical institutions. Although Hill (2009) acknowledges that the data may be impacted strongly by the school-selection bias of students, the findings are no less interesting.

Diversity in Higher Education

As a focus on spirituality lends numerous benefits to student development, so do rich and meaningful interactions between those of varying races, ethnicities, and backgrounds. When exposed to a diverse learning environment, students of all backgrounds have the potential to experience developmental and educational benefits (Bok, 2006; Chang, 1999; Hurtado et al., 1998). Research shows that students will experience growth in cognitive, social, and critical thinking skills, a greater commitment to civic engagement, an enhanced understanding and appreciation for those with various opinions and experiences, and an amplified desire to provide assistance to those who are in need (Allen, 2006; Bok, 2006; Chang, 1999; Hurtado, 2006, 2007; Hurtado et al., 1998; Ibarra, 2001). Research has also shown that educational, purposeful interaction between diverse groups of students is related to personal growth and development, vocational preparation, intellectual development, and diversity competence (Hu & Kuh, 2003). However, in order for students to experience the benefits associated with a diverse learning environment, institutions of higher education need to engage in the facilitation of a constructive educational community and commit to the advancement of diversity as an institutional priority (Allen et al., 2006; Anderson, 2007; Bok, 2006; Ibarra, 2001).

Institutions must not underestimate the importance of actively facilitating a diverse learning community. Although structural diversity is an essential component to creating such a community, research has shown that the lived experience on campus is very different for underrepresented students compared with white students (Ancis et al., 2000; Chavous, 2005). Johnson et al. (2007) found that white students reported significantly higher levels of sense of belonging than all underrepresented student groups, and black students were the least likely to report positive racial climates. Ancis et al. (2000) found that white students had a more positive perception of the campus racial climate, and black students perceived more racial tension on campus and felt they received unequal treatment from faculty. Utilizing Allport's (1954) intergroup contact theory as a theoretical framework, Chavous (2005) found that white students rated the campus climate more positively than black students in each of the areas Allport (1954) identified as essential to successful intergroup interactions: 1) meaningful association, 2) everyone sharing equal status on campus, 3) interdependence/everyone as having something to offer, 4) university support of diversity, and 5) personal association with diverse peers. Further, white students reported a significantly higher sense of community than black students. While the generalizability of this study is low as it was conducted at a single predominantly white institution (PWI), the results confirm previous research that black students perceive the racial climate less positively than their white counterparts (Ancis et al., 2000; Chavous, 2005; Hurtado et al., 1999; Johnson et al., 2007; Rankin & Reason, 2005).

Diversity at Evangelical Institutions

Despite the evidence that suggests the great value of integrating multiple forms of diversity into the learning environment, evangelical colleges lag behind their public and private non-sectarian counterparts (Haralu, 2005; Paredes-Collins, 2009). Although the CCCU, a consortium of 111 evangelical institutions of higher education, has an average minority enrollment of 16%, actual diversity enrollments at typical CCCU schools are much lower, as a few non-traditional, non-residential outliers skew the overall average (Paredes-Collins, 2009). Further, many schools within the CCCU do not identify campus diversity as an institutional priority, resulting in a lack of allocated resources or the basic policies and practices necessary to promote a multicultural learning experience for students (Paredes-Collins, 2009). Unfortunately, the issue of diversity on evangelical campuses across the country is an understudied subject area (Haralu, 2005).

What has been studied is the difference in student development for various racial and ethnic groups at predominantly white institutions (PWI). For example, African American students are better adjusted at historically black institutions versus predominantly white institutions (Fleming, 1984; Hughes, 1987). In a small, qualitative phenomenological study about the identity development of black students, Lazarus Stewart (2008) focused on the interaction and influences of the various aspects of multiple identities. The following themes emerged from the study: validity of multiple identities (each student acknowledged that they had to navigate multiple identities, depending on the scenario, and there was more to their identity than their race), the critical role that higher education played in their ability to understand and articulate their identity, and finally, the difficulty of navigating identity integration (e.g., students felt pressure to navigate their multiple identities, representing different identities at different times). Furthermore, religion and spiritual development take on particular significance for African Americans (Hughes, 1987). Wijeyesinghe (2001) found that in relation to racial identity, multiracial people find spiritual beliefs as a source of strength and refuge from racism that sustains them through racial identity development and “assist[s] them in deriving greater meaning from their racial ancestry or identity” (p. 143).

In contrast, Helms (1995) developed a model of white racial identity where the central task for whites is “the recognition and abandonment of internalized White privilege and the creation of a nonracist, self-defining White identity” (p. 189). Interestingly, Tranby and Hartmann (2008) found that white conservative Protestants were significantly more likely to consider “whiteness” as important to their identity than other whites and racial groups. Brown (2009) found that white evangelical Protestants are significantly more conservative in their attitude about race-based policies as compared to white mainline Protestants, secular whites, and whites of other faiths. For example, when compared to white evangelicals, white mainline Protestants are almost two and a half times as likely, and secular whites and whites of other faiths are more than twice as likely to think that the historical marginalization of blacks warrants special consideration today.

Emerson and Smith's (2000) book, Divided by Faith, qualitatively chronicled the race-based attitudes and ideals of evangelical Christians. Tranby and Hartmann (2008) have extended the study initially conducted by Emerson and Smith, to reveal more pervasive anti-black sentiments and stereotypes, along with the subtle commitment to the majority white culture as being “normal” among white evangelicals. Tranby and Hartmann (2008) purport that “evangelical individualist culture is more deeply racialized … than Emerson and Smith imagine” (p. 342). For example, white conservative evangelicals were significantly less likely to think that prejudice and discrimination or laws and institutions favor whites, than all other Americans combined (Tranby & Hartmann, 2008).

The research about the white evangelical attitudes toward people of color calls into question the experience of underrepresented students at evangelical colleges. In a national study of religious participation in college, Hill (2009) found that compared to other students, black Protestants significantly decreased their participation in religious services when they attended evangelical institutions. “Although white evangelicals and black Protestants share many conservative theological beliefs, barriers between groups exist in how best to incorporate an understanding of racial identity in their religious understandings. It is possible that these barriers may leave black Protestants as outsiders on evangelical campuses” (Hill, p. 532). The gap in the literature concerns the intersection between race and spirituality at predominantly white evangelical schools, particularly as it relates to the experience of underrepresented minority students and their spiritual development in an environment that is defined by a mission committed to spiritual growth.

Methodology

The Instrument

This study utilized data from the College Students' Beliefs and Values (CSBV) survey, a longitudinal dataset from the Higher Education Research Institute of California (HERI) Spirituality in Higher Education project, a Templeton-funded initiative. The instrument was developed in spring 2003, with extensive pilot testing on a group of 3,700 juniors from 46 institutions who had completed the 2000 Cooperative Institutional Research Program (CIRP) survey as a freshman. In fall 2004, the instrument was included as an addendum to the annual CIRP Freshman Survey, which includes extensive questions about student demographics, attitudes, behaviors, and expectations about college. The CSBV incorporated an additional 160 questions about student attitudes, perspectives, and practices in terms of spirituality and religiosity.

The following factor scales, as previously defined by the HERI Spirituality in Higher Education project, were considered as dependent variables of student spiritual growth: spiritual identification (alpha = .89), equanimity (alpha = .72), religious commitment (alpha = .97), ethic of caring (alpha = .82), and compassionate self concept (alpha = .78). These factors were developed through the process of principle components analysis with varimax rotation. Table 1 presents information about the individual items on each scale.

TABLE 1 Description of Spirituality Factors

Factors	Description
Spiritual Identification	Composite score on the following items: Personal goal measured on a 4-point scale, “Not important” to “Essential:” integrating spirituality into my life, seeking out opportunities to help me grow spiritually. Self-description measured on a 3-point scale, “Not at all” to “To a great extent:” having an interest in spirituality, believing in the sacredness of life. Self-rating measured on a 5-point scale, “Lowest 10%” to “Highest 10%:” spirituality, on a spiritual quest. Belief measured on a 4-point scale, “Disagree strongly” to “Agree strongly:” people can reach a higher spiritual plane of consciousness through meditation or prayer. Spiritual experience measured on a 3-point scale, “Not at all” to “Frequently:” viewing a great work of art, participating in a musical or artistic performance, engaging in athletics, witnessing the beauty and harmony of nature, meditating, listening to beautiful music.
Equanimity	Composite score on the following items: Self-description measured on a 3-point scale, “Not at all” to “To a great extent:” feeling good about the direction in which my life is headed, being thankful for all that has happened to me, seeing each day, good or bad, as a gift.
Religious Commitment	Composite score on the following items: Personal goal measured on a 4-point scale, “Not important” to “Essential:” seeking to follow religious teachings in my everyday life. Self-rating measured on a 5-point scale, “Lowest 10%” to “Highest 10%:” religiousness. Belief measured on a 4-point scale, “Disagree strongly” to “Agree strongly:” I find religion to be personally helpful, I gain spiritual strength by trusting in a Higher Power. Self-description measured on a 3-point scale, “Not at all” to “To a great extent:” feeling a sense of connection with God/Higher Power that transcends my personal self, felt loved by God. My spiritual/religious beliefs measured on a 4-point scale, “Disagree strongly” to “Agree strongly:” are one of the most important things in my life, provide me with strength, support, and guidance; give meaning/purpose to my life; lie behind my whole approach to life; have helped me develop my identity; help define the goals I set for myself.
Ethic of Caring	Composite score on the following items: Engagement measured on a 3-point scale, “Not at all” to “To a great extent:” trying to change things that are unfair in the world. Personal goal measured on a 4-point scale, “Not important” to “Essential:” helping others who are in difficulty; reducing pain and suffering in the world; helping to promote racial understanding; becoming involved in programs to clean up the environment; becoming a community leader; influencing social values; influencing the political structure.
Compassionate Self-Concept	Composite score on the following items: Self-rating measured on a 5-point scale, “Lowest 10%” to “Highest 10%:” kindness; compassion; forgiveness; generosity.

Procedures

During fall 2004, the CIRP/CSBV survey was completed by 112,232 entering freshmen from 236 diverse institutions across the country. While survey distribution at each respective institution was at their discretion, the paper-and-pen surveys were typically administered during the first few weeks of the semester or during new student orientation. In spring 2007, 14,527 students from 136 of the original 236 institutions completed a follow-up version of the CSBV. The longitudinal nature of the dataset enables the measurement of growth during college. While the follow-up sample represented various types of institutions (public, private non-sectarian, Roman Catholic, mainline Protestant, and evangelical colleges), the participants from evangelical schools are the primary focus of this study.

Participants

A total of 3,238 students attending an evangelical institution completed both the original CIRP/CSBV in 2004 and the CSBV follow-up survey in 2007. Of the respondents, 988 were men and 2,250 were women. Approximately 12% of the students identified with one of the following underrepresented racial categories: black/African American, American Indian/Alaskan Native, Asian/Asian American, Mexican American/Chicano/a, and other Latino/a, Native Hawaiian/Pacific Islander, or Puerto Rican. The underrepresented students in this study were considered both in the aggregate (non-white), and individually. As the instrument allowed students to identify with multiple races, each race variable was coded dichotomously (e.g., white/Caucasian or not, black/African American or not). Table 2 presents more information on the sample. Due to small numbers, this study did not disaggregate the results for students of Native Hawaiian/Pacific Islander or Puerto Rican descent. With 12% of this group comprised of underrepresented students, the demographic breakdown of the dataset is similar to that of evangelical institutions nationwide.

TABLE 2 Participants

Group	N	Percentage within group
Total	3,238
White/Caucasian	2,957	91.3
(Non-White/Caucasian)	281	8.7
Black/African American	70	2.2
American Indian/Alaskan Native	57	1.8
Asian American/Asian	103	3.2
Mexican American/Chicano	74	2.3
Other Latino	51	1.6
Native Hawaiian/Pacific Islander	18	0.6
Puerto Rican	15	0.5
Other	74	2.3

Results

Statistical Analysis

The following research questions guided this study: 1) Are there any significant racial differences in seniors' spirituality at faith-based schools in the CCCU, after controlling for their level of spirituality as a freshman? 2) Are there any significant racial differences in students' spirituality at CCCU schools? The following factors were utilized as the measures of spirituality: spiritual identification, equanimity, religious commitment, ethic of caring, and compassionate self-concept. To ensure homogeneity of the regression slopes, a custom multivariate analysis of covariance (MANCOVA) was conducted with the 2007 spirituality factor scores as the dependent variable and 2004 spirituality factor scores as the covariate. Race (white/Caucasian or not) was the independent variable. The custom MANCOVA revealed a significant interaction between the covariates and the dependent variables, making MANCOVA an unsuitable test for further analysis. As such, this study not able to address the first question.

To move forward with the second question, several one-way MANOVAs were conducted to determine if there were indeed significant racial differences in the spirituality factors at evangelical colleges. In each analysis, the spirituality factors were the dependent variables, and race was the independent variable. As race was coded as multiple dichotomous variables, a MANOVA was conducted for each individual race, compared with all other races combined (e.g., white/Caucasian students were compared to all who did not identify as white/Caucasian, black/African American were compared to all who did not identify as black/African American). Although each of the dependent variables was significantly and positively correlated, most of the correlations were low to moderate. The analyses were conducted on both the 2004 and the 2007 scores, to allow for comparison. Table 3 presents the means and standard deviations for both sets of scores.

TABLE 3 Means and Standard Deviations for Spirituality Scores in 2004 and 2007

	Spiritual identification (α = .89)*				Equanimity (α = .72)*				Religious Commitment (α = .97)*
	2004		2007		2004		2007		2004		2007
Group	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD
Overall	32.05	4.39	32.48	4.52	12.77	1.80	12.81	1.79	41.91	5.63	41.11	6.02
Race/Ethnicity**
White	32.09	4.35	32.52	4.50	12.78	1.80	12.80	1.78	41.98	5.56	41.18	5.98
African American	31.38	5.29	32.42	5.33	12.61	2.09	12.92	1.98	40.60	6.19	39.51	7.07
Asian/Asian American	31.58	5.10	32.77	4.92	12.58	1.75	12.93	1.71	40.94	7.34	40.23	6.65
Mexican/Chicano	32.48	4.89	32.14	4.96	12.63	1.94	12.89	1.72	41.56	6.09	40.28	6.25
Other Latino	32.02	3.64	32.75	4.54	13.31	1.55	12.87	1.66	43.69	2.99	42.29	3.96
	Ethic of caring (α = .82)*				Compassion (α = .77)*
	2004		2007		2004		2007
Group	M	SD	M	SD	M	SD	M	SD
Overall	17.70	3.75	19.85	4.05	15.43	2.28	14.98	2.22
Race/Ethnicity**
White	17.56	3.69	19.75	4.02	15.40	2.28	14.97	2.21
African American	18.82	4.50	21.81	4.72	15.40	2.57	15.00	2.27
Asian/Asian American	19.63	3.61	21.41	3.62	15.92	2.27	15.05	2.34
Mexican/Chicano	18.80	4.29	19.94	4.77	15.72	2.52	15.46	2.16
Other Latino	18.34	4.34	19.81	4.30	15.80	2.16	14.85	2.60

MANOVA results revealed significant racial differences among the combined spirituality variables in 2004 for white/Caucasian, African American, Asian/Asian American, and other Latino/a students. In 2007, MANOVA results revealed significant differences for white/Caucasian, African American, and Asian/Asian American students. There were no significant differences among the combined spirituality variables for other Latino students in 2007. Mexican/Chicano students did not differ significantly during 2004 or 2007. Table 4 presents the multivariate F ratios for 2004 and 2007.

TABLE 4 Multivariate and Univariate Analysis of Variance F Ratios for Spirituality Factors in 2004 and 2007

	MANOVA		ANOVA
	F	h	Spirituality	Equanimity	Religious Commitment	Caring	Compassion
White/Caucasian
2004 ^a	14.3***	.024	2.89	0.56	5.33*	49.19***	4.51*
2007 ^b	7.18**	.011	1.54	0.62	4.40*	21.15***	0.18
African American
2004 ^a	2.56*	.064	1.54	0.53	3.62	6.01*	0.01
2007 ^b	5.4***	.009	0.02	0.28	4.79*	15.81***	0.01
Asian/Asian American
2004 ^a	7.49***	.013	1.06	1.07	2.78	25.00***	4.42*
2007 ^b	4.57***	.007	0.43	0.49	2.26	15.61***	0.11
Mexican/Chicano ^c
2004 ^a	1.75	.121	0.62	0.4	0.26	5.68*	1.09
2007 ^b	1.34	.243	0.42	0.15	1.41	0.04	3.44
Other Latino
2004 ^a	2.64*	.005	0.002	4.12*	4.56*	1.36	1.24
2007 ^b	.545	.001	0.17	0.06	1.84	0.01	0.16

Analysis of variance (ANOVA) was conducted on each dependent variable as a follow-up test to MANOVA. In 2004, white/Caucasian students had significantly higher scores on the religious commitment scale than non-white students (F(1, 2899) = 5.33, p < .05, partial η² = .002), and non-white students had significantly higher scores on both the ethic of caring (F(1, 2899) = 49.19, p < .001, partial η² = .017) and the compassionate self-concept scale (F(1, 2899) = 4.51, p < .05, partial η² = .002). African American students had significantly higher scores on the ethic of caring scale than non-African American students (F(1, 2899) = 6.01, p < .01, partial η² = .002), and Asian/Asian American students had significantly higher scores on the ethic of caring scale (F(1, 2899) = 25, p < .001, partial η² = .009) and the compassionate self-concept scale than non-Asian/Asian American students (F(1, 2899) = 4.42, p < .05, partial η² = .002). Mexican/Chicano students had significantly higher scores on the ethic of caring scale than non-Mexican/Chicano students, (F(1, 2899) = 5.68, p < .05, partial η² = .002). Finally, other Latino/a students had significantly higher scores on the equanimity scale (F(1, 2899) = 4.12, p < .05, partial η² = .001) and the religious commitment scale (F(1, 2899) = 4.56, p < .05, partial η² = .002) than non-Latino students.

In 2007, univariate analysis revealed that white/Caucasian students maintained significantly higher scores on the religious commitment scale than non-white students (partial η² = .001), and non-white students maintained significantly higher scores on the ethic of caring (partial η² = .007) scale. The differences on the compassionate self-concept scale diminished in significance. African American students maintained significantly higher scores on the ethic of caring scale (partial η² = .005) and significantly decreased in their religious commitment (partial η² = .002) as compared to non-African American students. Asian/Asian American students maintained significantly higher scores on the ethic of caring scale (partial η² = .005) as compared to non-Asian/Asian American students, but the differences in compassionate self-concept diminished in significance. Mexican/Chicano student scores did not significantly differ in any way from non-Mexican/Chicano students. Lastly, other Latino student scores on the equanimity scale and the religious commitment scale decreased, resulting in non-significant differences between Latino/a and non-Latino/a student scores. Table 4 presents both the multivariate and univariate analyses for each racial category on the spirituality factors.

Lastly, paired samples t tests were conducted to assess the direction of the changes in student spirituality. Paired samples t tests indicated that students at evangelical colleges had on average significantly higher scores on the spiritual identification (t (2793) = 5.89, p < .001, d = .11) and ethic of caring (t (2793) = 29.9, p < .001, d = .02) scales as seniors than when they were freshmen. Overall, students had significantly lower scores on the religious commitment (t (2793) = −7.89, p < .001, d = .19) and the compassion (t (2793) = −10.39, p < .001, d = .20) scales. The differences, although statistically significant, are small using Cohen's (1988) guidelines. There was not a significant difference in equanimity scores (t (2793) = 1.04, p = .30).

The first question posed in this study asked if there are any significant racial differences in seniors' spirituality at CCCU schools, after controlling for their level of spirituality as a freshman. As a custom MANCOVA revealed a significant interaction between covariates and dependent variables, we were unable to answer this question. The second question asked if there were any significant racial differences in students' spirituality at CCCU schools. Indeed, the data showed significant racial differences in student's spirituality.

Discussion and Implications

Consistent with previous research (Welch & Mellberg, 2008), students at evangelical colleges had significantly higher scores on the spiritual identification scales as seniors than when they were freshmen. Students also showed significant growth in ethic of caring. Consistent with research at public and private/non-sectarian colleges, students decreased in their religious commitment during college (Astin, 1993). Several other factors emerged as being particularly noteworthy. Although the effect sizes of the significant differences in the combination of spirituality factors were smaller than typical by Cohen's standards (1988), the differences between white and non-white students at evangelical institutions in both religious commitment and ethic of caring merits further exploration. Although white students had higher levels of religious commitment than non-white students, they also showed significantly lower levels of ethic of caring. For white evangelicals, religious commitment could be analogous to a strict conservatism about following doctrine. This sentiment is reflected in the religious commitment scale, which includes the following items: I seek to follow religious teachings in my everyday life, I find religion to be personally helpful, my spiritual/religious beliefs are one of the most important things in my life.

Items in this scale reflect Kelley's (1995) definition of religious conservatism, where religious teachings are strictly followed and represent paramount importance. According to Kelley (1995), religious conservatives are more likely to identify a set of rules or guidelines (i.e., converting others, providing worship, religious instruction, ministerial services, and the sacraments) as the most central components to their faith. Actions such as helping the poor, supporting marginalized groups, and reaching out to the community, are important—but seen as secondary to the more individualized/inner components (Kelley, 1995). Particularly among white evangelicals, perhaps this self-reliance is connected with the commonly held belief that each person has a set of choices to make that will influence their position in life. This concept was confirmed by Brown (2009), who found that white evangelical Protestants are significantly less likely than all other whites to think that the historical marginalization of African Americans warrants special consideration today.

In terms of race relations, this reproduces the “pull yourself up by your bootstrap” mentality that whites apply to non-whites (Brown, 2009; Tranby & Hartmann, 2008). In this study, religious commitment for whites was significantly higher than non-whites at evangelical colleges, in both 2004 and 2007. Figure 1 represents these changes. This significant difference, in conjunction with the fact that evangelical colleges are predominantly white schools and 88% of the sample is white, may be indicative of some of the aforementioned white evangelical sentiments.

FIGURE 1 Change in “Religious Commitment” for whites and non-whites between 2004 and 2007.

Interestingly, religious commitment for both white and non-white students followed a trend similar to one found at public and private non-sectarian schools—it decreased (Astin, 1993). However, although it decreased for all students at evangelical colleges, there is still significant difference between the groups. Given that evangelical schools are predominantly white, it was hypothesized that religious commitment of non-whites might decrease as a result of being in a white religious environment. However, because both groups decrease, this presumption cannot be confirmed. The reasons for the decrease and potential differences between groups are unknown, and merit further exploration.

The other scale, ethic of caring, represents a contrast to the religious commitment scale. Sample items from this scale include: (it is important to me to) help promote racial understanding, reduce pain and suffering in the world, and try to change things that are unfair in the world. Opposite the religious commitment scale, white students scored significantly lower on ethic of caring than non-white students. This may be further indication of whites having a more individualized faith, and believing that people are responsible for their social location, thus reducing any consequential desire to change things that are unfair or promote racial understanding. These results support our original hypothesis that students of color would be more prone to the social justice-oriented aspects of faith and spirituality. Similar to the changes in religious commitment, the non-white and white students change at a comparable pace. However, as seen in Figure 2, they both increase in ethic of caring. This augmented concern about issues of inequity and care for others could be attributed to the evangelical college environment. white students, however, are still significantly lower than non-white students after three years of college.

FIGURE 2 Change in “Ethic of Caring” for whites and non-whites between 2004 and 2007.

While non-white and white students move in the same direction on both scales, it is important to note that white students score higher on religious commitment and lower on ethic of caring, and the opposite is true for non-white students. Figure 3 shows a comparison of the rate of change for white and non-white students with religions commitment and ethic of caring. Again, this supports our hypothesis that non-white students would be more focused on the social aspects of their faith.

FIGURE 3 Change in “Religious Commitment” and “Ethic of Caring” between 2004 and 2007.

Religion, spirituality, and race intersect at evangelical schools in a predominantly white environment, which enrolls a majority of students with lower concern for a sense of justice and correcting mistreatment. Research that shows the benefits of a diverse learning environment are relevant at evangelical schools, given that non-white, underrepresented students diversify the student body in terms of worldview as it relates to issues of justice (Allen, 2006; Bok, 2006; Chang, 1999; Hurtado, 2006, 2007; Hurtado et al., 1998). This further substantiates the call for predominantly white evangelical schools to work toward diversifying their student body, and to continue to educate students around faith-based mandates toward justice, which appear to be minimal from the traditionally white evangelical perspective.

In terms of the significant differences between individual racial groups, the most notable change is with African American students. In 2007, African American students significantly decreased in their religious commitment, as compared to non-African American students. This finding is consistent with Hill's (2009) research, where black Protestants significantly decreased their participation in religious services when they attended evangelical institutions. A similar decrease in religious commitment also occurred with other Latino students. It is possible that these students did not identify with the predominant religious culture on campus, and thereby decreased their own involvement. This finding bears great implication for evangelical colleges that are committed to providing an engaging spiritual environment on campus. If colleges are not intentional about creating an adaptive spiritual environment where students of different backgrounds, races, and denominations can find some sense of familiarity, marginalization continues.

Limitations

The results of this study must be considered with great caution, due to the limitations of the study. While the sample size was adequate and fairly reflective of the typical predominantly white evangelical campus (Paredes-Collins, 2009), the study would have been strengthened with a larger group of underrepresented students. As evangelical colleges are vastly predominantly white, this limitation stretches beyond this study and into the general limitations of studying evangelical colleges. Further, as the instrument was designed in such a way that students could identify multiple races, the dichotomous coding of the variables (e.g., white/Caucasian or not, African American or not) made it impossible to compare each racial category with the others. This categorization made it difficult to separate or measure the differences between groups. As the longitudinal nature of the dataset allows for the measurement of growth, after controlling for entering characteristics, MANCOVA was the appropriate method for analyses. However, most likely due to the relationship between the covariates and the factors, the assumptions of independence of observations and homogeneity of covariance were checked and not met, rendering MANCOVA an inappropriate method for further inquiry. As such, MANOVA was selected, relinquishing the ability to control from entering characteristics. Lastly, while there were significant differences between racial groups on various measures of spirituality, the majority of the effect sizes were smaller than typical, diminishing the ability to claim practical significance.

Future Research

As the intersection of race and spirituality at evangelical colleges is an understudied and under-theorized area of research, there is room for a great deal of expansion. Specifically, future researchers should consider studying spiritual and identity development in conjunction with sense of belonging among students of color at evangelical schools. This study only scratched the surface in terms of the experience of underrepresented students at faith-based institutions. It would also be advantageous to identify the predictors of success for students of color at evangelical colleges, in order to help schools adapt their campuses to be more integrative to such students. Only then will evangelical schools be able to move forward in terms of creating a diverse learning community for all students.

Conclusion

The primary objective of this study was to explore any racial differences in students' spirituality at faith-based schools. Significant differences were found between white students and non-white students, and the contrasting results between ethic of caring and religious commitment were most compelling. While there is more work to be done in this area, it is useful for evangelical schools to see that they are having the same directional impact on both white and non-white students, in terms of ethic of caring. However, given that the significant differences between groups are maintained during college, this may be a useful indicator that is not only beneficial for predominantly white evangelical schools to dedicate resources to diversify their collegiate community, but also to tailor the educational experience around certain proclivities.

Campuses where the student body comprises approximately 90% white evangelicals may be well served to design learning outcomes with specific indicators for specific groups. If white students continue to graduate with significantly lower scores in ethic of caring, perhaps the environment could benefit from being geared more specifically to building these attributes through service learning or cultural and civic engagement programs. Creating programs and curriculums that are designed around these practical outcomes for spirituality may also enhance the campus climate experience for non-white students. Speculating on this potential outcome provides a dual motive for institutions to make data driven decisions when it comes to curricular and co-curricular programs. These outcomes also serve to highlight that campuses benefit from engaging in a more holistic student experience design that moves beyond the goal of recruiting a more diverse student body (Chang, 2000; Hu & Kuh, 2003; Hurtado et al., 1998). As evangelical schools pursue diversity as a goal, the recruitment process must be accompanied by the incorporation of diversity into the overall student experience (Anderson, 2007). The results presented in this paper serve the purpose of pointing toward the benefit of this holistic design. Many evangelical schools require their faculty, staff, and sometimes students to sign a statement of faith. Race and ethnic relations connect with spirituality on many levels and increasing clarity and intention toward these outcomes should align with the missions of the schools, albeit misaligned with the tradition of white evangelical individualism and racial attitudes. The intersection between race and spirituality provides some evidence of the need and benefit of moving toward working through this tension.

Additional information

Notes on contributors

KRISTIN PAREDES-COLLINS

Kristin Paredes-Collins is currently a doctoral student in Higher Education at Azusa Pacific University.

CHRISTOPHER S. COLLINS

Christopher S. Collins, PhD, is Assistant Provost in the Office of Institutional Effectiveness at Pepperdine University.

Notes

Note. Students were permitted to mark more than one racial category, resulting in total numbers of individual race categories that exceed 100%.

*Represents the α levels for the individual constructs in 2004.

**As the survey allowed students to select more than one race, each race was coded as an individual, dichotomous variable.

Note. Multivariate F ratios were generated from Pillai's statistic.

^a 2004 Multivariate df = 5, 2895, Univariate df = 1, 2899.

^b 2007 Multivariate df = 5, 3104, Univariate df = 1, 3108.

^c Multivariate F ratio generated from Wilk's Λ.

*p < .05, **p < .01, ***p < .001.