Journal of Spirituality in Mental Health, 15:107–122, 2013 Copyright © Taylor & Francis Group, LLC ISSN: 1934-9637 print/1934-9645 online DOI: 10.1080/19349637.2013.776442
Assessing Spirituality: The Relationship Between Spirituality and Mental Health
DAVID R. BROWN Department of Behavioral Sciences, Cincinnati Christian University,
Cincinnati, Ohio, USA
JAMIE S. CARNEY Department of Special Education, Rehabilitation, Counseling/School Psychology,
Auburn University, Auburn, Alabama, USA
MARK S. PARRISH Department of Counseling and Educational Psychology, University of West Georgia,
Carrollton, Georgia, USA
JOHN L. KLEM Department of Rehabilitation and Counseling, University of Wisconsin-Stout,
Menomonie, Wisconsin, USA
This research study investigated the possible relationship between two spirituality variables (religious coping styles and spiritual well-being) and two psychological variables (anxiety and depression). Also studied were differences between those who self-disclosed a spiritual/religious identify and those who did not. Although a relationship was not noted between religious coping styles and the psychological variables, significance was reported in the relationship between spiritual well-being and both psy- chological variables. Overall, this study finds that individuals reporting higher levels of religiosity and spiritual well-being may also experience a reduction in mental and emotional illness.
KEYWORDS spirituality, religiosity, anxiety, depression
Over the past few decades, spirituality has become an increasingly impor- tant consideration in the mental health profession (Richards & Bergin, 2005; Young, Wiggins-Frame, & Cashwell, 2007). This has included an emphasis
Address correspondence to David R. Brown, Department of Behavioral Sciences, Cincinnati Christian University, Cincinnati, OH 45204. E-mail: email@example.com
108 D. R. Brown et al.
and awareness of the importance of integrating spirituality and religion within the counseling process (Parker, 2011; Seybold & Hill, 2001). The inte- gration of spirituality into the counseling process is reflective of recognition that spirituality can be beneficial to client welfare (Koenig, 2010; Miller & Thoresen, 2003; Seybold & Hill, 2001). Moreover, there is significant evi- dence that spirituality may be related to or positively impact overall mental health or well-being (Hodges, 2002; Townsend, Kladder, Ayele, & Mulligan, 2002). This point is reflected in the growing body of research examining the physical, emotional, and psychological effects of spirituality and religios- ity (Hayman et al., 2007). For many individuals, spirituality and religion are central and foundational aspects of their lives and their well-being, and thus critical elements of the counseling process and interpersonal dynamic (Miller & Thoresen, 2003; Parker, 2011; Seybold & Hill, 2001).
As previously noted, evidence suggests that spirituality may be linked to physical health and well-being (Hodges, 2002; Townsend et al., 2002), and that spirituality is a significant component of holistic wellness (Myers & Williard, 2003). From a mental health perspective, religion and spiritual- ity may be involved in how individuals and groups make decisions, solve problems, and cope with life experiences; all of these activities incorporate spiritual themes and subsequently can correspond to overall improved men- tal health (Thurston, 1999; Pargament et al., 1988). In addition, spirituality may be an asset or a coping strategy for dealing with negative life events, as well as with psychological concerns (Koenig, 2010; Pargament et al., 1988); for example, Hayman et al. (2007) reported that spirituality helped buffer the negative effect of stress on self-esteem. There are also indicators that spiritu- ality may relate to how an individual deals with or is affected by depression (Srinivasan, Cohen, & Parikh, 2003; Westgate, 1996) and anxiety (Graham, Furr, Flowers, & Burke, 2001).
LIMITATIONS OF CURRENT RESEARCH IN SPIRITUALITY AND RELIGIOSITY
Although research has supported that spirituality is linked to both positive physical health (Miller & Thoresen, 2003; Townsend et al., 2002) and pos- itive mental health (Koenig, 2010; Hayman et al., 2007), the same research also notes a number of complicating factors in studying spirituality and reli- giosity. A review of literature reveals that similar limitations are noted in many research studies, the most common problem being the definition of spirituality. When questioning if spirituality can be measured, Oakes and Raphel (2008) noted that a common concern is defining the constructs: “these definitional problems make it difficult to know what a measure of spiritu- ality actually assesses” (p. 243). This problem in defining spirituality echoes Speck’s (2005) concerns regarding the inherent difficulty in determining a
Assessment of Spirituality and Mental Health 109
consensual definition of spirituality because of its epistemological nature. Further, Seybold and Hill (2003) suggested that the inconsistencies in defin- ing spirituality and religiosity have resulted in an ambiguous nature of research findings, which then leads to conflicted reporting when linking research outcomes to the mental and physical health issues being treated.
Other noted limitations in studying spirituality and religiosity are that quantitative measures may not fully assess the subjective nature of spirituality (Moberg, 2002) and that a majority of spiritually related assessment instru- ments are developed from a Judeo-Christian perspective (Stanard, Sandhu, & Painter, 2000). Moberg (2002) suggested that qualitative assessments may provide more useful and specific information about an individual’s spiritual- ity than a quantitative measure, as individual responses may better express spiritual needs and experiences. As noted by Stanard et al. (2000), many measures of spirituality and religiosity also lack normative information, thus limiting their usefulness in clinical settings. Because a lack of normative information inhibits the ability to generalize assessed results, Moberg’s (2002) suggestion for a qualitative assessment of spirituality reflects an understand- ing of the individualistic nature of spiritual experience and expressions, as well as articulating the difficulty in even developing a normative under- standing of spirituality. Miller and Thoresen (2003) stated that spirituality and religiosity are best described as latent constructs, which are complex and multidimensional variables. Therefore, such complexity in a construct implies that no single assessment instrument can adequately capture its meaning.
Research in spirituality and religiosity has attempted to address this com- plexity through the development of a multitude of assessment instruments. Hill and Hood (1999) published a review of 125 spirituality/religiosity assess- ment instruments, which were placed into 17 categories; each designed to assess a different construct of spirituality and religiosity. These 17 categories were defined as (a) religious beliefs and practices, (b) religious attitudes, (c) religious orientation, (d) religious development, (e) religious commitment and involvement, (f) religious experience, (g) religious/moral values or per- sonal characteristics, (h) multidimensional religiousness, (i) religious coping and problem solving, (j) spirituality and mysticism, (k) God concept, (l) reli- gious fundamentalism, (m) death/afterlife, (n) divine intervention/religious attribution, (o) forgiveness, (p) institutional religion, and (q) related con- structs. Unfortunately, as noted by Stanard et al. (2000), most of the assessment instruments reviewed by Hill and Hood (1999) suffered from a lack of normalizing data, questionable design, and most were devel- oped from a Judeo-Christian perspective. Some instruments, although initially developed from a Judeo-Christian view, have shown promise through the development of normalized information, validation through repeated use within numerous research studies, and refinement of nonspecific religious vocabulary. Assessment instruments, such as the Spiritual Well-Being Scale (Ellison, 1983; Paloutzian & Ellison, 1982) and the Religious Problem-Solving
110 D. R. Brown et al.
Scale (Pargament et al., 1988), have demonstrated high levels of validity and reliability, thus suggesting greater utility in research and practice.
The increase of interest in the study of spirituality and religiosity has resulted in the development and improvement of assessment instru- ments designed to evaluate various constructs of spirituality and religiosity. Research studies, such as Hayman et al. (2007) and Davis, Kerr, and Robinson-Kurpius (2003), used assessment instruments designed to measure faith maturity and levels of spiritual well-being and religious orientation, respectively, related to various mental health concerns. Davis et al.’s (2003) study reported that “greater spiritual well-being predicted lower trait anxi- ety among at-risk adolescents” (p. 361), although they did caution against over-generalization of these results. Similarly, Hayman et al.’s (2007) study reported that higher levels faith maturity correlated with higher levels of self-esteem and lower levels of stress and body-image concerns. Although further study in the areas of spirituality and religiosity is recommended, suffi- cient evidence has already been collected to demonstrate correlations among spirituality and religiosity with both mental and physical health. It appears that further research should focus in refining an understanding of various spiritual constructs with physical and mental health, as well as determining how to resolve the deficiencies in the qualitative assessment of spirituality and religiosity, as noted previously.
THE CURRENT STUDY
The current study was designed to evaluate the relationship of spiritual well-being and religious problem-solving with anxiety and depression. The selected spiritual constructs were partially determined through the selec- tion of well-validated instruments designed to evaluate spirituality. Because of noted limitations with spirituality assessment instruments, the authors resolved to carefully select assessment instruments that have demonstrated fewer such limitations; a discussion of these instruments is provided below. Furthermore, because anxiety and depression are currently understood as two of the most common psychological concerns throughout the world (Seligman & Reichenberg, 2007), they presented as common, personal characteristics that could be present in a nonclinical sample population.
Because literature has suggested that the relationship between men- tal health and spirituality is complex (Miller & Thoresen, 2003; Seybold & Hill, 2001), it is important to note that understanding this relationship is made even more complex when one considers the overlapping and dif- fering constructs and definitions of religiosity and spirituality, as noted by Moberg (2002) and Richards, Bartz, and O’Grady (2009). For the purposes of this study, spirituality will be defined as a sense of connectedness to a higher power and openness to the infinite beyond human existence and
Assessment of Spirituality and Mental Health 111
experience (Burke et al., 1999). Religion (or religiosity) will be defined as “an institutionalized set of beliefs and practices by which groups and individuals relate to the ultimate” (Burke et al., 1999, p. 252). Inherent in these definitions is an understanding that both religion and spirituality are complex constructs, to address this issue the current study focused on two specific aspects of spiritual and religious identity: religious problem-solving (specifically religious coping styles as they correspond to one’s relation- ship with God in a passive, collaborative, or self-initiating approach, as developed by Pargament et al., 1988) and spiritual well-being. The latter (spiritual well-being) consists of a global concept relating to one’s own perception of spirituality and well-being, including one’s sense of quality of life (Ellison, 1983; Paloutzian & Ellison, 1982). Furthermore, it was the intent of this research study to compare across the spirituality measures to determine which demonstrated a more significant relationship with the mea- sures of mental health used in this study (Beck Anxiety Inventory and Beck Depression Inventory-II). In addition, this study will address any differences noted within the sample population, as some participants were recruited from a private, religiously affiliated university, and other participants were recruited from a public university. The findings of this study could provide critical information for counseling professionals about this relationship, as well as how it relates to psychological and mental health concerns.
Participants and Sampling
Both undergraduate and graduate students from two universities were recruited as a convenience sample to participate in this study: one university is a large, public institution in the Southeast (Southeastern), and the second university is a small, private, religiously affiliated university located in the Midwest (Midwestern). All participants were recruited from both undergradu- ate and graduate courses and were offered extra credit to complete an assess- ment packet. A total of 150 surveys were distributed at the end of a class ses- sion and collected the following class session; in all, 121 survey packets were returned (response rate of 81.3%; there was no follow-up). All responses were anonymous. There were 30 male and 91 female participants. Participant ages ranged from 19 to 56 years (M = 24.50 years). Ethnic diversity among participants was slight as 82% self-reported as Caucasian, 13.1% self-reported as African-American, and 4% self-reported with other distinct ethnicities. Furthermore, religious diversity was low; 96.2% (n = 25) of the Midwestern participants reported to be Christian (3.8% [n = 1] reported to be Messianic Jew) or of a Christian denomination, and 90.6% (n = 87) Southeastern par- ticipants reported to be Christian or of a Christian denomination. Other Southeastern participants reported as Agnostic (4.2%, n = 4), Gnostic (1.0%, n = 1), Jewish (2.1%, n = 2), and Seventh-Day Adventist (1.0%, n = 1).
112 D. R. Brown et al.
Participants completed a packet consisting of a demographic ques- tionnaire and four assessment instruments. The demographic questionnaire requested grouping data such as age, ethnicity, gender, religious/spiritual affiliation, and the use of the word “God” in their spiritual/religious affil- iation. To counterbalance the presentation of measures and not affect the participant responses, the order of documents placed in half of the survey envelope packets contained documents in a different order.
The Religious Problem-Solving Scale (RPSS) was used to measure religious coping and problem-solving styles. This scale was designed by Pargament et al. (1988). Consisting of three subscales (Self-Directing, Collaborative, and Deferring), the RPSS contains 36 items on a 5-point Likert scale (1 = never, 2 = occasionally, 3 = fairly often, 4 = very often, and 5 = always) in which item responses indicate how often the individual engages in an activity. According to Thurston (1999), reliability and validity are reportedly strong: Collaborative (r = .94, α = .93), .94 Self-Directing (r = .94, α = .91), and Deferring (r = .91, α = .89). Test-retest reliability returned promising relia- bility estimates: α = .93 (Collaborative), α = .94 (Self-Directing), and α = .87 (Deferring). According to Pargament et al. (1988), in respect to measures of religiousness, the Self-Directing subscale correlated to a significantly negative relationship with a Higher Power, whereas the Collaborative and Deferring exhibited a positive relationship.
The Spiritual Well-Being Scale (SWBS; Ellison, 1983; Paloutzian & Ellison, 1982) was designed to provide a global measure of a respondent’s quality of life and one’s perception of spiritual well-being (Boivin, Kirby, Underwood, & Silva, 1999). The instrument is constructed of two subscales: religious well-being (Religious) and existential well-being (Existential), as well as an overall score of spiritual well-being (SWB). The SWBS is a 20-item assessment answered on a 6-point Likert-type scale (1 = strongly agree to 6 = strongly disagree), where reliability and validity appear to be high (Stanard et al., 2000): Religious (r = .96, α = .96), Existential (r = .86, α = .78), and SWB (r = .93, α = .89), with a slight correlation between the two subscales (r = .32), a high correlation between SWB and the Religious subscale (r = .90), and a moderate correlation between SWB and the Existential subscale (r = .59). However, the test-retest reliability coefficients with four samples on a 1-, 4-, 6-, and 10-week interval resulted in high reliability; the SWB global scale ranged from .82 to .99, the Religious subscale ranged from .88 to .99, and the Existential subscale ranged from .73 to .98 (Paloutzian & Ellison, 1991). Internal consistency reliability coefficients for the two subscales and the global scale reported results for SWB ranging from .89 to .94, results for the Religious subscale ranging from .82 to .94, and results for the Existential subscale ranging from .78 to .86.
Assessment of Spirituality and Mental Health 113
The Beck Anxiety Inventory (BAI; Beck, Epstein, Brown, & Steer, 1988) was used in this study to measure symptom levels of anxiety. The BAI was created from three existing anxiety assessments to more accurately discriminate anxiety-related diagnoses from non-anxiety-related diagnoses. Consisting of 21 items, the BAI is answered on a 4-point Likert scale (0 = not at all to 3 = severely; I could barely stand it) to indicate severity of anxi- ety symptoms. Beck et al. (1988) reported that the BAI displayed high levels of internal consistency (α = .92). Furthermore, Dowd and Waller (1998) reported that internal consistency reliability coefficients ranged between .85 and .94 and test-retest reliability over one week resulted in a coefficient of .75.
The Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1996) was utilized in this study to evaluate levels of depression in participants. The BDI-II consists of 21 items, each scored on a 4-point Likert scale (0 = not present to 3 = severe). Beck et al. (1996) reported that the BDI-II exhibits high Cronbach’s alphas: outpatients (α = .92) and college students (α = .93). Internal consistency reliability was measured using corrected item-total cor- relations for both the clinical (range = .39 to .70) and convenience (range = .27 to .74) samples. The test-retest reliability resulted in a reliability coef- ficient of α = .93. The BDI-II also correlated well with the BAI with a small subject sample (n = 297; r = .60).
Because the spirituality assessments were developed from a Judeo-Christian perspective, participants were asked to respond to the following on the demographics questionnaire: “Does your spiritual/religious identity use the word ‘God’?” One hundred percent of participants indicated that the word “God” is used in their spiritual/religious identity. Familiarity with the word “God” clearly did not invalidate the results. Analysis of assessment results reported that Midwestern participants reported higher scores on the Collaborative (M = 43.88; SD = 7.039; α = .88) and Deferring (M = 30.15; SD = 7.358; α = .83) subscale than Southeastern participants (M = 37.31; SD = 12.531; α = .96 and M = 29.25; SD = 10.595; α = .93, respectively). Conversely, the Southeastern participants reported higher scores on the Self- Directing subscale (M = 25.32; SD = 12.015; α = .96) than the Midwestern participant (M = 21.96; SD = 6.109; α = .87). Overall, the mean responses on the Collaborative (M = 38.72, SD = 11.861, α = .96), Deferring (M = 29.45, SD = 9.967, α = .92), and Self-Directing (M = 24.60, SD = 11.080, α = .96) subscales suggested that participants were more likely to use a collabo- rative relationship with God to address problems and cope with negative life experiences, as well as demonstrating strong reliability factors. Overall mean results for the Religious (M = 52.59, SD = 10.061, α = .95) and Existential
114 D. R. Brown et al.