Printer Friendly

Making sense of Quest's multidimensionality: the search for a higher order structure.

The multidimensionality of the Quest construct has been suggested by a number of authors, although there is a paucity of multidimensional Quest scales in the extant literature. Further, the vast majority of researchers continue to utilize unidimensional measures of Quest. In a sample of 436 university students, the Multidimensional Quest Orientation Scale (MQOS; Beck & Jessup, 2004) was subjected to a principal axis factor analysis. The results suggested a nine-factor structure, although the Exploration scale was removed, due to its suspected measurement of apologetics. The remaining eight factors were subjected to a secondary factor analysis, suggesting a higher-order factor structure consistent with Edwards et al.'s (2002) concepts of "hard" and "soft" Quest. Correlational analyses between the MQOS subscales and other measures of religiosity (e.g., extrinsic and intrinsic motivation, religious commitment, and defensive theology) support the usage of multidimensional measures of Quest, due to the unique information provided by each of the dimensions and their respective higher order factors. However, the higher order structure helps to elucidate the broader themes suggested within the multiple dimensions of Quest.

A recent report from the Pew Research Center (American Values Survey, 2012) indicated that Millennial (i.e., those born after 1980) are less religious than previous generations. Further, according to the report, data from 2007 indicated that 15% of Millennials reported uncertainty with God's existence. By 2012, this number had doubled (to 31%). This increase in doubt is exceptional, not only in the history of the Millennial generation, but also across generations (e.g., among Baby Boomers and those from Generation X) who have remained generally stable in previous assessments of their uncertainty of Gods existence.

While today's young adults are reporting more theistic uncertainty than any generation since the Pew Research Center began collecting data on this question in 1987, it is unclear whether this increase in doubt is part of a larger, "less religious" movement in Millennial, or if a different dimension of religiosity is emerging among Millennial in an unprecedented fashion. Some prominent authors in the study of religion have even indicated a need to embrace uncertainty. For example, Armstrong (2009) stated that "a modern theology must look unflinchingly into the heart of a great darkness and be prepared, perhaps, to enter into the cloud of unknowing" (p. 278). Regardless of one's perspective, the psychology of religious doubt seems more relevant than ever.

Quest Motivation: History and Conceptualization

Nearly 40 years ago, Batson and colleagues (e.g., Batson, 1976; Batson, Naifeh, & Pate, 1978; Batson & Ventis, 1982) suggested that Allport's (Allport & Ross, 1967) two dimensions of religiosity, intrinsic (religion as end) and extrinsic (religion as means), did not comprehensively measure religiosity. Therefore, they suggested a third dimension: religion as quest, which "concerned the degree to which the individual sought to face religious issues in all their complexity, while resisting clear-cut, pat answers" (Batson, Flink, Schoenrade, Fultz, & Pych, 1986; p. 175). These three dimensions, according to Batson and Schoenrade (1991a) should be treated as independent and orthogonal.

As conceptualized and written by Batson and Schoenrade (1991a; 1991b), the Interactional Scale (often referred to as the Quest Scale) was designed to assess three aspects: 1) readiness to face existential questions without reducing their complexity, 2) self-criticism and perception of religious doubts as positive, and 3) openness to change. Despite the conceptualization of Quest as being comprised of these three aspects, Batson and Schoenrade (1991b) recommended that the Interactional Scale be scored unidimensionally. The authors reported that several of the items loaded on more than one factor. Further, they indicated that more items (for each proposed dimension) would be necessary to more adequately assess the multidimensional construct validity of the Interactional Scale.

The Multidimensionality of Quest

As was aptly noted by Edwards, Hall, Slater, and Hill (2011), the vast majority of Quest research is based on the utilization of a single score for Quest measurement. Indeed, most of the research on Quest seems to include the usage of Batson's Interactional Scale (1991b) and, as a result, much of the correlational data in the extant literature is based on relations with a unidimensional (i.e., total score) conceptualization of Quest. However, in recent years, some researchers have undertaken efforts to explore the multidimensionality of the Quest construct.

Utilizing student samples from a small private (Christian-affiliated) University, Beck and colleagues (i.e., Beck, Baker, Robbins, & Dow, 2001; Beck & Jessup, 2004) have developed a line of research devoted to assessing the multidimensionality of Quest. Initially, Beck et al. (2001) found support for the facets of openness to change and tentativeness, and later (i.e., Beck & Jessup, 2004) added items for seven additional Quest dimensions. Regarding their Multidimensional Quest Orientation Scale (MQOS) Beck and Jessup (2004) indicated that "five of the MQOS subscales follow directly from Batson's various descriptions of the Quest construct" (p. 284), and these included tentativeness, openness to change, exploration, complexity, and existential motives. Additionally, Beck and Jessup added items to assess openness to other Christian faiths (i.e., the ecumenism items) and items to assess openness to all other world religions (i.e., the universality items). Further, in order to assess one's endorsement of moralistic interpretations of the Bible (i.e., rather than a literal interpretation), Beck and Jessup (2004) included items to assess a posited moralistic interpretation facet of Quest. Lastly, Beck and Jessup (2004) created items to assess religious angst, or the existential emotions that individuals may experience in their religious journey.

Their results supported the posited nine factors, with rather small intercorrelations (r = .00 to .37, mean r = 0.15, SD = 0.11) between the subscales, suggesting distinct dimensions. Further, with the exception of the complexity subscale ([alpha] = .68), each of the subscales yielded an alpha above .70. However, due to the relatively small sample size (n = 183), Beck and Jessup (2004) were unable to undertake an item-level factor analysis. Therefore, an exploratory factor analysis of the MQOS (at the item-level) has yet to be conducted.

Most recently, Edwards, Hall, Slater, and Hill (2011), undertook a large-scale study of the Quest construct with over 1,000 students from nine Christian colleges in the United States. Utilizing three measures of Quest (i.e., Altmeyer & Hunsberger, 1992; Batson & Schoenrade, 1991b; Dudley & Cruise, 1990), Edwards et al. (2011) found exploratory and confirmatory factor analytic support for five Quest dimensions: QF1 (Beliefs Tentative Open to Change), QF2 (Reason versus Faith), QF3 (Questions Good), QF4 (Doubts Valued), and QF5 (Religion as Quest). Additional correlational analyses with other religious measures (including measures of spiritual well-being, God awareness, and fundamentalism) were utilized to support the construct validity of the factors identified by Edwards et al. (2011).

Quest: The Hard and Soft Dimensions

While the work of Beck and colleagues (i.e., Beck, Baker, & Robbins, 2001; Beck & Jessup, 2004) and Edwards and colleagues (i.e., Edwards, Hall, Slater, & Hill, 2011) seems to support the multidimensionality of the Quest construct, the necessary number of dimensions (i.e., for adequate measurement of Quest's multidimensionality) remains in question. While Beck and Jessup (2004) proposed nine dimensions, Edwards et al. (2011) found evidence for five dimensions. However, a more parsimonious (i.e., higher-order) dimensionality has also been proposed: Edwards, Slater, Hall, Oda, and Eck (2001) originally proposed "hard" and "soft" versions of the Quest construct.

Hard Quest. Hard Quest can be defined as an ongoing, largely existential Quest, potentially leading an individual toward or away from their own traditional religious beliefs (Beck & Jessup, 2004; Edwards, Hall, & Slater, 2002; Edwards et al., 2011). Beck and Jessup (2004) posit that those displaying higher levels of hard Quest may frequently change their religious views and do not tend to value traditional religious practice as central to their spiritual journey. Further, according to their findings, these persons may have lower levels of spiritual well-being and may struggle with feelings of existential isolation and anxiety.

Soft Quest. Those who are committed to a religious worldview (e.g., Christianity) yet manifest Quest attributes may be said to exhibit Soft Quest (Beck & Jessup, 2004). Therefore, while Soft Quest is comprised of openness and doubt, those higher in Soft Quest may have reached some religious conclusions and display religious commitment. As Edwards et al. (2011) indicated, Soft Quest allows for satisfying ones beliefs, "while remaining open to change and embracing doubts as an inevitable component of religious belief" (p. 107).

Beck and Jessup (2004) found support for both Hard and Soft Quest in their initial study of the MQOS. When examining inter-scale correlations, they found that the subscales posited as Hard Quest (i.e., Change, Universality, Religious Angst, Complexity, and Existential Motives) were associated with a number of other religious variables, including decreased intrinsic religiosity, increased extrinsic religiosity, decreased spiritual well-being, and less orthodox beliefs. They further suggested that the MQOS subscales Tentativeness, Ecumenism, Exploration, and Moralistic Interpretation suggest a tentative and curious person, who is "not beholden to Biblical literalism" (p. 290). These scales, therefore, were posited to measure Soft Quest. Beck and Jessup found that Exploration was positively related to intrinsic religiosity, while none of the other subscales (i.e., Tentativeness, Exploration, and Moralistic Interpretation) were related to the remaining study variables (i.e., extrinsic religiosity, spiritual well-being, and orthodoxy). Therefore, the relationships among these posited Soft Quest subscales of the MQOS and other religious variables remain unclear.

Edwards et al. (2011), also utilizing correlations among their subscales and other religious measures, found support for Hard and Soft Quest. Specifically, their factors QF1 and QF4 (labeled Beliefs Tentative Open to Change and Doubts Valued, respectively) provide an apt model for measuring Soft Quest. These two subscales correlated negatively with religious fundamentalism, seemingly indicating an "open cognitive style" (p. 107). Further, Edwards et al. found that QF2 and QF5 (Religion vs. Faith and Religion a Quest) corresponded to their conceptualization of Hard Quest, with each of these subscales being negatively related to spiritual well-being and God-awareness.

How do Hard and Soft Quest Relate to Batson' Original Conceptualization of Quest?

Edwards et al. (2011) reported that their factors QF1 (Beliefs Tentative Openness to Change) and QF4 (Doubts Valued) could potentially be "viewed as core latent dimensions of the quest construct as formulated by Batson" (p. 107), and that these dimensions are representative of Soft Quest. They further stated that (their conceptualization of) Soft Quest seems to yield a more "mature" (p. 107) and open cognitive style of religiousness that is unrelated to religious commitment and spiritual well-being. Edwards et al. indicated that their factors of QF2 (Reason versus Faith) and QF5 (Religion a Quest) seemed consistent with their conceptualizations of Hard Quest (and as a more immature religiousness). Both QF2 and QF5 were inversely related to God-awareness and spiritual well-being.

Beck and Jessup (2004), however, indicated that their subscales Change, Universality, Religious Angst, Complexity, and Existential Motives seemed representative of the Hard Quest dimension, and that this dimension (i.e., Hard Quest) is more consistent with Batson's original conceptualization of Quest. Beck and Jessup stated that those high in (their conceptualization of) Hard Quest have tendencies to be "more agnostic about Christian dogma" (p. 290).

The Current Study

The current study was focused primarily on two areas. First, the multidimensionality of the Quest construct was re-examined through factor analysis of the MQOS. While Beck and Jessup (2004) did find support for their posited nine-factor structure, they were unable to conduct an item-level factor analysis of their data due to a relatively small sample size (n = 183). Additionally, their sample consisted of students from a small private University, and the representation of religious affiliation was quite homogeneous (i.e., 61% indicated a religious affiliation with the Church of Christ). Beck and Jessup (2004), therefore, indicated the need for replication and continued efforts for validation.

Additionally, Beck and Jessup (2004) noted that the MQOS subscales Change, Universality, Angst, Complexity, and Existential Motives each yielded similar correlations with the other measures of religiosity utilized in their study. Despite small intercorrelations between these subscales, they indicated the possibility that these aforementioned subscales are not actually measuring distinct constructs. Therefore, additional attention was devoted to the nature of the relationships of each of the MOQS scales with other measures of religiosity: intrinsic and extrinsic religiosity (Gorsuch & McPherson, 1989), religious commitment (Worthington et al., 2003), and defensive theology (Beck, 2006).

As Beck (2006) found that his Defensive Theology Scale (DTS) was negatively related to the Interactional Scale (Batson & Schoenrade, 1991b), it is expected that the DTS total score will also be negatively related to each of the subscales of the MQOS. Further, Beck and Jessup (2004) found that several of their proposed MQOS subscales (i.e., Change, Universality, Religious Angst, Complexity, and Existential Motives) were positively related to extrinsic religiosity, and these same relationships are expected in the current study. Similarly, as reported by Beck and Jessup (2004), negative relationships between MQOS subscales (i.e., Change, Universality, Religious Angst, Complexity, and Existential Motives) and intrinsic religiosity are expected. Finally, while there are no data published (to date) on the relationship between the Quest construct and religious commitment (as measured by the Religious Commitment Inventory; Worthington et al., 2003), it is plausible that Quest should be negatively related to religious commitment. Particular aspects within Quest (e.g., as defined by Beck & Jessup, 2004) include tentativeness of religious beliefs and openness to changing religious views, and seem especially likely to be negatively related to religious commitment.

Second, the MQOS was analyzed for a higher-order structure. While multidimensionality may yield important insights into the nature of a construct, parsimony in measurement is also rather desirable. Therefore, following an item-level factor analysis of the MQOS to determine factor structure, the subscales were subjected to a higher-order factor analysis.

Although there seems to be a relative paucity of published data on higher-order structures for multidimensional scales of Quest, some authors (e.g., Edwards, Hall, Slater, & Hill, 2011) have found some support for the dimensions of "hard" and "soft" Quest. Of particular interest in the current study was the comparison of the higher-order structure of the MQOS to these concepts. Furthermore, the nature of the relationships between the higher-order structures and the aforementioned religious measures was assessed. An examination of these relationships with other religious constructs (i.e., comparing the data from the sub-scale and higher order levels) stands to answer important theoretical and statistical questions. Perhaps most importantly, while increased parsimony may be meaningfully achieved through a higher-order structure, it is also plausible that subscale-level relationships may yield important information that may be otherwise disguised within an aggregate structure.

Method

Participants

The sample was composed of 436 college students (71% female, 29% male) from a mid-sized public university in southeast Texas, United States. The mean age was 22.00 years, with a standard deviation of 4.88 years. The ethnic composition was Caucasian (n = 243), Hispanic/Latino (n = 77), African American (n = 69), Multi-Racial (n = 15), Bi-Racial (n = 12), American Indian or Alaska Native (n = 11), Asian or Asian-American (n = 7), and Hawaiian or other Pacific Islander (n = 2). Three hundred forty-nine participants (80%) indicated a Christian religious affiliation. The largest of these groups (i.e., Non-Denominational Christian, Roman Catholic, and Baptist) constituted 64% of the sample (n = 277). Eleven percent reported having no preference for any affiliation, while 5% reported being Agnostic.

A portion of the current sample was utilized for a study of preferences for religious help-seeking (Crosby & Varela, In Press). Quest was not examined in that study.

Procedure

Through a university-based research system, students were invited to participate in a web-based study of religious attitudes and beliefs. After the presentation of informed consent, participants consented and responded to eleven scales (presented in counterbalanced order). Upon completion of the study, participants were provided with extra- or research-credit by their course instructors. A demographic questionnaire was utilized to assess the demographic characteristics of the sample (i.e., age, gender, and race) as well as their religious affiliation. These data were collected as part of a larger study on the psychology of religion and of the eleven scales administered to participants only four were used in the current study. These scales are described below.

Measures

Multidimensional Quest Orientation Scale. (1) The Multidimensional Quest Orientation Scale (MQOS; Beck & Jessup, 2004) is a 62 item measure of Quest orientation that was designed to include nine subdimensions (i.e., tentativeness, change, ecumenism, universality, exploration, moralistic interpretation, religious angst, complexity, and existential motives). Participants are asked to respond to the items on a 7-point Likert-type scale (1 = strongly disagree to 7 = strongly agree). Although the original authors (Beck and Jessup, 2004) were unable to complete an item-level factor analysis (due to sample size), their results supported the nine-factor structure. Alpha values for each of the nine subscales ranged from .68 to .90, with only one scale (Complexity, [alpha] = .68) yielding an alpha coefficient below .70. Generally low correlations among the subscales also suggested that they were each assessing relatively unique aspects of Quest. With the exception of Moralistic Interpretation, all of the subscales correlated with Batsons Interactional Scale (Batson & Schoenrade, 1991 b). The items, as presented to participants, may be viewed in Appendix A.

Revised Religious Orientation Scale. The Revised Religious Orientation Scale (Revised ROS; Gorsuch & McPherson, 1989) is a 14-item scale measuring intrinsic and extrinsic (I/E) religiosity. Respondents rate item endorsement on a 5-point Likert-type scale (strongly disagree to strongly agree), with 8 items forming the intrinsic subscale and 6 items forming the extrinsic subscale. Gorsuch and McPherson (1989) found support for two sub-factors (3 items each) within the Extrinsic scale, which were labeled Extrinsic-Personal ([E.sub.p]; e.g., "What religion offers me most is comfort in times of trouble and sorrow") and Extrinsic-Social ([E.sub.s]; e.g., "I go to church because it helps me to make friends"). Gorsuch and McPherson (1989) indicated that their purpose was to establish the best items for measuring I/E religiosity, and their factor analytic results supported an overall reduction in the number of items included in the original ROS (All-port & Ross, 1967). In the current study, the mean for the 8 item intrinsic scale was 26.18 (SD = 7.54) with an alpha of .86. Further, the mean score for the Extrinsic-Personal ([E.sub.p]) scale was 9.71 (SD = 3.08) with an alpha of .79, and the Extrinsic-Social ([E.sub.s]) scale had a mean of 5.90 (SD = 2.70) with an alpha of .80.

Religious Commitment Inventory. The Religious Commitment Inventory (RCI; Worthington et al., 2003) is a ten-item measure of intrapersonal and interpersonal religious commitment. The authors reported that a unidimensional (i.e., total) score is most parsimonious, although factor analytic results supported the usage of the intra- (6 items) and interpersonal (4 items) commitment scales. Intrapersonal items were designed to measure beliefs and activities focused on the self (e.g., "It is important to me to spend periods of time in private religious thought and reflection") whereas the interpersonal items were focused on the communal aspects of religious involvement (e.g., "I enjoy spending time with others of my religious affiliation"). Participants are asked to respond through the use of a Likert-type scale (1 = not at all true of me to 5 = totally true of me). Each of these scales was found to be positively related to frequency of religious activities and participation in organized religion. The alpha coefficients were acceptable (i.e., intrapersonal [alpha] = .92, interpersonal [alpha] = .87) and the test-retest reliability coefficients (3 weeks) for the intra-and interpersonal scales were .86 and .83, respectively. In four college student samples, RCI-10 total scale means ranged from 19.10 (SD = 3.20) to 22.00 (SD = 6.1). For the current study, the total mean was 26.81 (SD = 10.64), while the sub-scale means were 16.52 (SD = 6.74) for the RCI Intrapersonal scale and 10.29 (SD = 4.35) for the Interpersonal scale. The alpha estimates in the current study (Intrapersonal [alpha] = .92, Interpersonal [alpha] = .86) were consistent with those reported by Worthington et al. (2003), while the current sample displayed generally higher levels of religious commitment than did the samples in Worthington et al.

Defensive Theology Scale. The Defensive Theology Scale (DTS; Beck, 2006) is a unidimensional scale designed to measure the extent to which individuals use their faith to repress existential anxiety. The DTS is 22 items long and respondents are asked to indicate their level of agreement with items using a 7-point Likert-type scale (1 = Strongly Disagree to 7 = Strongly Agree). Beck (2006) found that the DTS was negatively correlated to the Interactional Scale (Batson & Schoenrade, 1991b). While descriptive statistics (i.e., mean and SD) were not reported for the DTS, Beck found acceptable internal consistency for the DTS ([alpha] = .86). In the current study, the mean for the DTS was 90.95 (SD = 30.00), while the alpha coefficient was .95.

Results

Principal Axis Factor Analysis

An examination of the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (i.e., 0.86), Bartlett's Test of Sphericity ([x.sup.2] [1891] = 13804.60, p < .001), and the communalities indicated that a structural analysis of the Multidimensional Quest Orientation Scale (MQOS; Beck & Jessup, 2004) would be appropriate. Therefore, a principal axis factor analysis with orthogonal (varimax) rotation was performed on the 62 items of the MQOS. Orthogonal rotation was chosen due to the history of relatively low correlations among the subscales of the MQOS (as discussed by Beck and Jessup, 2004). Further, as will be discussed below, the data from the current sample supported the use of orthogonal rotation due to a replication of the aforementioned low intercorrelations between subscales. Initially, thirteen factors had eigenvalues greater than 1.0, accounting for 55.35% of the total variance. However, the scree plot indicated either a seven-factor or nine-factor solution, respectively accounting for 40.10% and 48.18% of the total variance. Each of these solutions (i.e., 7, 9, and 13 factors) was examined following orthogonal rotation, yet the theoretical meaningfulness of the nine-factor solution led to a decision to use the nine-factor solution. Of the 62 MQOS items originally published by Beck and Jessup (2004), 58 had structure coefficients of .40 or greater (see Table 1). The skewness statistics for the MQOS subscales (as well as each of the other religious measures) were acceptable and ranged from -.66 to .61.
TABLE 1

Factor Loadings, Means, Standard Deviations, and Communalities for
the Nine Factor Solution

 Factor
 Loadings
 (a)

Item 1 2 3 4 5 6 7 8 9
#

53 .766 .157 .048 .017 .084 .128 .098 .098 .045

11 .753 .058 .086 .075 .086 .073 .132 .071 .058

61 .716 .034 .032 .083 .067 .104 .003 .076 .014

44 .707 .023 -.008 .065 .063 .046 -.046 .098 .033

38 .643 -.086 -.064 .047 .065 -.052 .011 .121 .078

6 .609 -.106 .027 .187 -.075 -.025 .108 .048 .017

59 .561 -.052 .057 .314 .021 -.015 .144 .112 .085

28 .551 -.028 -.035 .320 .006 -.015 .135 .071 .079

54 .548 .100 .189 .055 .078 .059 .001 .083 .123

24 .060 .742 .006 -.008 .198 .139 .089 .070 .005

43 .063 .731 -.006 .013 .146 .061 .082 .054 -.053

26 -.130 .672 .038 .036 .167 .096 .023 .005 .151

52 -.055 .672 .017 .020 .119 .279 .113 -.034 .020

36 .142 .663 -.065 -.049 .134 -.029 .271 .061 .032

13 .033 .652 -.070 .028 .147 .169 .090 .104 -.029

4 -.044 .574 .108 .124 .091 .174 .018 -.118 .014

21 .025 .516 .095 .025 .236 .164 .008 -.005 .174

22 .080 .030 .827 .044 -.010 .002 -.097 .016 .032

23 -.078 .041 .772 .019 -.022 .180 -.177 .007 .016

50 .001 .064 .735 .107 -.067 .070 -.006 .139 .073

17 .000 -.100 .714 .105 -.054 .063 -.144 -.031 .012

18 .119 .000 .694 .044 -.037 .045 .027 -.026 .008

45 .145 .095 .658 .156 .032 .018 -.022 .215 .165

46 .098 .051 .070 .837 -.005 .056 .027 .047 .063

9 .170 .033 .107 .750 .104 .055 -.001 .082 .109

10 .196 -.032 .064 .740 -.026 -.028 .047 .018 .024

62 .161 .077 .175 .711 -.010 .050 -.067 .070 .112

8 .152 .006 -.092 .694 .024 .045 .014 .123 .027

2 .061 .087 .162 .646 -.029 .053 .007 .088 .124

41 -.051 .135 -.058 -.075 .745 .022 .057 -.043 -.019

56 .042 .197 -.095 .019 .619 .118 .092 .086 .061

19 -.072 .143 -.155 -.113 .605 .082 .034 .041 -.086

57 .037 .157 -.062 .122 .574 .110 .257 .110 .202

34 -.035 .142 .004 -.253 .551 .008 .057 .069 -.036

5 .133 .037 -.083 .080 .535 .035 .069 .084 .037

14 .226 .176 .186 .058 .453 .119 -.058 .149 .152

48 .103 .205 .066 .060 .420 .131 -.024 .064 .216

49 .153 .121 .079 .158 .413 .040 -.052 -.012 .133

29 -.068 .214 .089 .004 .071 .729 -.024 -.090 .025

55 .027 .211 .068 .048 .135 .722 .022 .032 .003

12 .045 .170 .017 .077 .066 .701 .053 .045 .102

30 .083 .064 .063 .017 .043 .603 .033 -.029 .085

1 .070 .111 .089 .000 .055 .582 -.009 -.020 .136

58 .098 .113 .042 .049 .160 .566 -.073 .102 .082

37 .091 .189 -.087 .036 .083 .129 .848 .041 .108

15 .107 .149 -.147 .036 .096 .060 .796 .136 .156

40 .063 .202 -.093 .005 .087 .012 .777 .033 .083

60R .211 .044 -.241 .017 .061 -.175 .594 .202 -.008

39 .217 .087 .000 .062 .091 .033 .117 .655 .124

20 .170 .005 .187 .085 .133 .017 .021 .574 .183

25 .036 .134 .063 .103 .116 .084 -.042 .751 .179

33 .088 -.057 -.044 .163 .029 -.112 .180 .711 .033

47 .296 -.064 .080 .051 .072 .016 .155 .593 .128

3 .122 .018 .245 -.012 -.022 .079 -.007 .129 .681

42 .181 -.028 .181 .049 -.024 .086 .028 .049 .657

31 -.060 .158 -.120 .083 .181 .117 .166 .056 .579

32 .013 .074 .049 .084 .081 .178 .047 .181 .553

7 .101 -.027 -.011 .124 .131 .100 .018 .107 .449

16 * .103 .204 .220 .107 .374 .153 .032 .107 .272

27R .064 .057 .085 -.133 .050 -.148 .196 .141 .043
*

35 * .161 .000 .209 .039 .094 -.050 .258 .259 .350

51R .020 .041 -.190 .114 .041 -.081 .120 .024 .298
*

Item Mean SD [h.sup.2]
#

53 3.61 1.67 .658

11 3.49 1.61 .622

61 3.55 1.64 .543

44 3.38 1.64 .524

38 2.91 1.62 .455

6 2.54 1.55 .439

59 3.27 1.72 .460

28 2.81 1.64 .438

54 3.76 1.78 .381

24 5.25 1.72 .626

43 5.40 1.81 .576

26 5.45 1.67 .532

52 5.47 1.71 .562

36 5.19 1.81 .563

13 5.35 1.73 .501

4 4.83 1.94 .411

21 4.81 1.66 .389

22 3.60 1.74 .705

23 3.82 1.81 .669

50 3.97 1.77 .590

17 3.32 1.80 .560

18 3.09 1.74 .503

45 4.10 1.81 .562

46 3.45 1.84 .728

9 3.64 1.82 .637

10 2.90 1.80 .596

62 3.66 1.76 .592

8 2.97 1.75 .532

2 3.74 1.78 .482

41 4.32 1.67 .591

56 4.67 1.69 .466

19 4.61 1.70 .445

57 4.70 1.71 .505

34 4.52 1.64 .399

5 4.39 1.64 .333

14 4.59 1.55 .388

48 4.66 1.54 .305

49 4.36 1.64 .262

29 5.06 1.55 .604

55 4.88 1.53 .594

12 4.56 1.58 .549

30 4.33 1.39 .389

1 4.56 1.59 .385

58 4.46 1.52 .395

37 4.60 2.06 .809

15 4.33 2.07 .746

40 4.63 2.11 .673

60R 4.51 2.21 .533

39 4.09 1.70 .526

20 3.92 1.77 .453

25 4.38 1.72 .651

33 3.87 1.78 .592

47 3.81 1.72 .499

3 4.02 1.73 .563

42 4.07 1.82 .511

31 4.82 1.62 .463

32 4.51 1.63 .394

7 4.49 1.62 .267

16 * 4.63 1.49 .362

27R 4.44 1.64 .117
*

35 * 3.98 1.63 .339

51R 4.68 1.61 .164
*

Note: (a) No cross-loaders were observed in the matrix of
factor loadings.

1 = Change, 2 = Ecumenism, 3 = Exploration
(Apologetics) 4 = Religious Angst, 5 =
Tentativeness, 6 = Moralistic Interpretation,
7 = Universality, 8 = Existential Motives,
and 9 = Complexity.

* Indicates item that did not sufficiently load
on any factor (i.e., factor loading < .40)


With the exception of the four items not loading on any of the nine factors, the structure of the scale was remarkably similar to that of Beck and Jessup (2004). However, after an examination of the bivariate correlations among each of the subscales, as well as their respective correlations with other measures of religiosity, the construct validity of the Exploration subscale was a concern. Specifically, Exploration was negatively correlated to Universality (r = -.21, p < .001). Further, it was positively related to religious defensiveness (r = .28, p < .001) measured with the Defensive Theology Scale; Beck, 2006) and religious commitment (r = .54, p < .001 [Intrapersonal]; r = .45, p < .001 [Interpersonal] measured with the Religious Commitment Inventory, Worthington et al., 2003). An examination of these relationships, as well as the content of the items on the Exploration subscale seemed to indicate that this subscale is, at least to some extent, measuring apologetics rather that Quest-motivated exploration. For example, some items state that the respondent has spent "more time compared with most people I know investigating the foundations of my religious faith" and "time studying the teachings of religions around the world" (p. 293). Therefore, it was determined that this subscale would be removed from the final solution in the current study.

The correlation coefficients among each of the nine subscales (including Exploration) are displayed in Table 2. Similar to the findings of Beck and Jessup (2004), the correlations between MQOS subscales were generally small, ranging from .05 to .42 (mean r = 0.18, SD = 0.12; compared to Beck's and Jessup's range of .00 to .37, mean r = 0.15, and SD = 0.11). Therefore, while these subscales do seem to share some variance, the generally low relationships among them suggest that they are each assessing distinct aspects of Quest.
TABLE 2

Correlations among the Nine MQOS Subscales

 1 2 3 4 5 6 7

1. Change (.88)

2. Ecumenism .08 (.88)

3. Exploration .13 .05 (.88)
 **

4. Religious .35 .10 * .21 (.89)
Angst ** **

5. .17 .42 -.02 .07 (.82)
Tentativeness ** **

6. Moralistic .12 .37 .16 .12 * .26 (.84)
Interpretation ** ** ** **

7. Universality .23 .28 -.21 .06 .23 .05 (.88)
 ** ** ** **

8. .35 .10 * .14 .25 .23 .07 .25
Existential Motives ** ** ** ** **

9. Complexity .23 .16 .19 .22 .23 .26 .17
 ** ** ** ** ** ** **

 8 9

1. Change

2. Ecumenism

3. Exploration

4. Religious
Angst

5.
Tentativeness

6. Moralistic
Interpretation

7. Universality

8. (-83)
Existential Motives

9. Complexity .32 (.69)
 **

* = p < -05, ** = p < .01

Alpha coefficients for each of the MQOS subscales are presented
parenthetically on the diagonal.


Higher-Order Factor Analysis

While each of the previously discussed subscales seem to have both statistical and theoretical meaningfulness, an additional principal axis factor analysis was conducted on the eight MQOS subscales (i.e., excluding Exploration) to assess the data for a higher-order factor structure. An examination of the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (i.e., 0.70), Bartlett's Test of Sphericity ([x.sup.2] [28] = 479.59, p < .001), and the communalities indicated that a higher-order analysis of the MQOS would be appropriate.

A principal axis factor analysis with orthogonal (varimax) rotation was performed on the eight sub-scales (the Exploration subscale was not included). Again, orthogonal rotation was chosen due to the relatively low correlations among the subscales. Three factors had eigenvalues greater than 1.0, accounting for 59.91% of the total variance. Further, the scree plot indicated a three-factor solution. In the rotated solution, each of the eight subscales yielded loadings of .40 or greater (see Table 3).
TABLE 3

Factor Loadings, Means, Standard Deviations, and Communalities for
the Higher-Order Solution

 Factor
 Loadings

MQOS Factor 1 2 3 Weighted Mean SD
 Mean (a)

Change (H) .59 .03 .16 3.26 29.31 10.68

Existential .55 -.02 .29 4.01 20.07 6.70
Motives (H)

Religious Angst .50 .11 -.02 3.39 20.36 8.63
(H)

Complexity (H) .42 .24 .15 4.38 21.91 6.00

Moralistic .15 .67 .03 4.64 27.85 6.82
Interpretation
(SC)

Ecumenism (SC) -.02 .53 .50 5.22 41.75 10.29

Universality .20 .01 .52 4.52 18.07 7.22
(SU)

Tentativeness .14 .35 .45 4.54 40.82 9.43
(SU)

MQOS Factor [h.sup.2]

Change (H) .38

Existential .39
Motives (H)

Religious Angst .26
(H)

Complexity (H) .25

Moralistic .47
Interpretation
(SC)

Ecumenism (SC) .54

Universality .31
(SU)

Tentativeness .35
(SU)

Note: (a) For comparison of the participants'
endorsement across subscales, the means of the
subscale total scores were each divided by the
respective number of items on the subscale.

(H) indicates Hard Quest, (SC) indicates Christian
Soft Quest, (SU) indicates Universal Soft Quest


Cross-loading was observed on the second and third factors, with Ecumenism having sufficient factor loadings (i.e., greater than .40) on both factors. Given the relatively larger factor loading and theoretical meaningfulness, Ecumenism was retained on the second factor only. It should be noted that Beck and Jessup (2004) created the Ecumenism and Moralistic Interpretation scales for Christian samples, as they are comprised of items specifically related to Christianity.

Whereas the first factor seemed to measure Hard Quest (as previously described), and the second and third factors were consistent with previous descriptions of Soft Quest, the Christian-specific items were clustered on factor two and the more religiously-universal items were clustered on factor three. Therefore, the first factor (termed Hard Quest) included Change, Existential Motives, Religious Angst, and Complexity; the second factor (termed Christian Soft Quest) included Moralistic Interpretation and Ecumenism; and the third factor (termed Universal Soft Quest) included Universality and Tentativeness. The Hard Quest factor was positively related to Christian Soft Quest (r = .22, p < .001) and Universal Soft Quest (r = .33, p < .001), while Christian Soft Quest and Universal Soft Quest were moderately positively correlated (r = .39, p < .001).

Correlational Analyses: Lower and Higher Order Factors and Other Religious Variables

Table 4 displays correlations among subscales and other religious constructs (i.e., intrinsic and extrinsic religiosity, religious commitment, and defensive theology). Raw score totals were used for the correlations between each of the nine MQOS factors and the other variables of interest. However, it should be noted that the various factors of the MQOS contain different numbers of items (see Table 1). In order to give equal weight to each of the factors (e.g., Change, Existential Motives, etc.) within the higher-order dimensions (Hard Quest, Christian Soft Quest and Universal Soft Quest), the total scores for each of the factors were transformed to Z-scores, and then summed to create a total score.
TABLE 4

Correlations among MQOS Factors and Other Religious Variables

 ROS ROS ROS RCI RCI
 Intrinsic Extrinsic-S Extrinsic-P Intra Inter

Change -.32 ** .14 ** -.02 -.25 -.23
 ** **

Existential -.26 ** .01 .03 -.15 -.18
Motives ** **

Religious Angst -.18 ** .18 ** .12 * -.09 -.11
 *

Complexity -.02 -.06 .16 ** .04 -.04

Hard Quest -.29 ** .10 * .11 * -.16 -.21
 ** **

Moralistic .23 ** -.05 .31 ** .25 .19
Interpretation ** **

Ecumenism .03 -.09 .16 ** .05 -.01

Christian Soft .16 ** -.09 .28 ** .18 .11*
Quest **

Universality -.47 ** -.18 ** -.03 -.44 -.42
 ** **

Tentativeness -.10 * -.04 .01 -.05 -.10
 *

Universal Soft -.36 ** -.13 ** -.01 -.31 -.34
Quest ** **

 Defensive

 Theology

Change -.29 **

Existential -.21 **
Motives

Religious Angst -.17 **

Complexity -.03

Hard Quest -.26 **

Moralistic .22 **
Interpretation

Ecumenism -.03

Christian Soft .11 *
Quest

Universality -.36 **

Tentativeness -.14 **

Universal Soft -.32 **
Quest

** = p < .01, * = p < .05


Quest Correlations with Intrinsic and Extrinsic Religiosity. Results of the correlational analyses indicated that the Hard Quest scales (Change, Existential Motives, and Religious Angst) tended to be negatively correlated to intrinsic religiosity and positively correlated to extrinsic religiosity ([E.sub.s]: Change and Religious Angst; [E.sub.p]: Religious Angst and Complexity).

The Christian Soft Quest Scales (Moralistic Interpretation and Ecumenism) were each correlated to [E.sub.p], while Moralistic Interpretation was positively related to intrinsic religiosity. The Christian Soft Quest sub-scales were unrelated to [E.sub.s]. The Universal Soft Quest subscales (Universality and Tentativeness) were negatively related to intrinsic religiosity, and unrelated to [E.sub.p]. However, Universality was negatively related to [E.sub.s].

Quest Correlations with Religious Commitment and Defensive Theology. The results of correlational analyses with the Hard Quest scales revealed a general pattern of small negative correlations with both intrapersonal and interpersonal religious commitment (RCI; Worthington et al., 2003) and defensive theology (DTS; Beck, 2006). However, Complexity was unrelated to both religious commitment and defensive theology.

Of the Christian Soft Quest Scales, only Moralistic Interpretation yielded a significant relationship with either of these constructs. Specifically, Moralistic Interpretation was positively related to intrapersonal religious commitment and defensive theology. In a manner similar to that of the Hard Quest subscales, the subscales within Universal Soft Quest (i.e., Universality and Tentativeness) revealed a pattern of negative correlations to religious commitment and defensive theology. However, Tentativeness was unrelated to intrapersonal religious commitment.

Discussion

Interestingly, the study of the multidimensionality of Quest has now come full circle: from a unidimensional measure, to multidimensional measures, to (currently) a search for the higher order structure of Quest. As was previously discussed, Batson and Schoenrade (1991b) have recommended that the Interactional Scale be scored unidimensionally. Ultimately, the result is simply a sum-score for a construct that others have argued is much more complex. As Batson posited that Allport and Ross (1967) could not comprehensively measure religiosity with the Religious Orientation Scale, other authors have questioned the unidimensional conceptualization of Quest.

If we are to lend any credence to a dimensional conceptualization of Quest, whether it is through Batson's three aspects of Quest, or the multidimensional aspects purported by Beck and Jessup (2004) or Edwards et al. (2011), we must also ask if the individual dimensions provide us with insight into the manifestations of Quest in the lives of people. Correlations between Quest factors may be of basic psychometric interest, but what else do these factors tell us? As will be discussed below, each of the dimensions, while part of a theoretically and statistically cohesive unit (i.e., the Quest construct), provides us with unique information.

The primary purposes of the current study were to 1) examine the multidimensionality of the Quest construct through an exploratory factor analysis of the Multidimensional Quest Orientation Scale (MQOS; Beck & Jessup, 2004) and 2) following the selection of a factor solution of the MQOS, to analyze the data for higher order dimensionality. Additionally, correlations analyses between the factors of the MQOS (as well as the higher order dimensions) and measures of religiosity, religious commitment, and religious defensiveness were completed. The purpose of these correlational analyses was to examine the concurrent validity of the MQOS and to examine the unique information provided by the factors and their respective higher order dimensions. First, however, the factor analytic results warrant discussion.

The Structure of the Quest Construct

The Factor Structure of the Multidimensional Quest Orientation Scale. The results of the current study support the previously hypothesized factor structure of the Multidimensional Quest Orientation Scale (MQOS; Beck & Jessup, 2004). As was previously discussed, the current factor analysis of the MQOS should be considered exploratory (rather than confirmatory) in nature, due to the small sample size of Beck and Jessup (2004) and the resulting inability to conduct an item-level factor analysis. Each of the nine factors proposed by Beck and Jessup (2004) were supported by the results of the current study. However, the Exploratory subscale seemed measure apologetics (to some extent) rather than a pure exploratory questing search of other faiths that one may expect to see in the Quest construct. The Exploratory subscale was, therefore, excluded from the higher order factor analysis, and the final eight-factor solution was subjected to a higher order factor analysis.

The Higher Order Structure of Quest. While previous researchers have discussed the manner in which these factors may cluster together in both a theoretical and a statistical sense (i.e., often through bivariate correlations), their hypotheses on the nature of Hard and Soft Quest have not been assessed through factor analytic methods. For the current study, Quest's multidimensionality was assessed via the MQOS (i.e., Becks and Jessup's [2004] conceptualization of Quests multidimensional nature). While the final eight-factor solution does have both theoretical and statistical merit, it is somewhat cumbersome. The results of a higher-order factor analysis (of the eight MQOS factors) indicated a three-factor solution. Upon closer examination of the content of each of the factors, it was apparent that the first factor contained elements that previous authors (e.g., Beck & Jessup, 2004; Edwards et al., 2011) have labeled (collectively) as Hard Quest (i.e., Change, Existential Motives, Religious Angst, and Complexity).

Based on theory and the previously reviewed research, Moralistic Interpretation, Ecumenism, Tentativeness, and Universality were each expected to load into a single Soft Quest dimension. However, two Soft Quest measures seemed to emerge: one with Christian-specific content and one with content that was more universal (i.e., open to and applicable to religions other than Christianity). Despite the differences in the targets of the items (i.e., Christian versus universal), it should be noted that each of these dimensions consisted of items that indicate a general religious openness.

MQOS Factors and Higher Order Dimensions: Correlations with Other Measures of Religion

The results of this study, as well as those of Beck and colleagues (i.e., Beck, Baker, Robbins, & Dow, 2001; Beck & Jessup, 2004) and Edwards et al. (2011) support the multidimensional conceptualizations of Quest. However, factor analytic results alone are not sufficient to demonstrate the validity of an instrument or theory. Therefore, each of the eight factors of the MQOS was subjected to correlational analyses with intrinsic and extrinsic religiosity, religious commitment, and defensive theology.

These data support Becks and Jessup's (2004) speculation that those higher in Hard Quest would be more likely to change their religious views, as the Hard Quest scales tended to be negatively correlated to religious commitment. Potentially, these tendencies for lower religious commitment could be a result of their existential (i.e., searching) motives for religion. Additionally, three of the four Hard Quest scales were negatively related to defensiveness (Complexity was unrelated to defensiveness). Therefore, those high in Hard Quest tend to be less dependent on clear-cut answers to anxiety-provoking existential questions. However, this finding should not be construed to indicate that those high in Hard Quest do not experience existential anxiety. Beck and Jessup (2004) speculated that those high in Hard Quest would struggle with isolation and anxiety. Further, in addition to the inclusion of the Religious Angst scale within the overall Hard Quest dimension, previous research has indicated that Hard Quest seems to be negatively associated with spiritual well-being (Beck & Jessup, 2004; Edwards et al., 2011). It seems that the existential search for religious meaning, lower levels of defensiveness, and lower levels of religious commitment yield a dimension of Quest (i.e., Hard Quest) that is existentially distressing. However, additional research is certainly necessary to more definitively ascertain the emotional correlates of Hard Quest.

Further, and consistent with Beck and Jessup (2004), a pattern emerged within the Hard Quest dimension in which the scales were negatively related to intrinsic religiosity and positively related to extrinsic religiosity. It is reasonable to speculate (as did Beck and Jessup [2004], who called Hard Quest a "lonely enterprise" [p. 290]), that those high in Hard Quest may also have higher scores in extrinsic religiosity because they are attempting to reduce their feelings of isolation by associating with a religious group. Their free-roaming search, however, remains religiously non-committal.

The Soft Quest higher-order dimensions (i.e., both Christian-and Universal) were also uniquely correlated to each of the religious variables assessed. Interestingly, while each of these Soft Quest dimensions is theoretically linked, they seem to be related to other constructs in opposing ways. While the Universal Soft Quest subscales were negatively related to religious commitment and defensiveness, Moralistic Interpretation (a Christian-Specific subscale) was positively related to religious commitment and defensiveness. Those who are committed to their Christian faith tend to be more accepting of Moralistic Interpretation (i.e., considering the moral meaning of the Biblical text as more important than scientific or historical accuracy), while the inverse is true for Universality (i.e., those that think of other religions as equally valid pursuits of the Truth tend to be less religiously committed). It is, therefore, plausible that Universal Soft Quest is a more open version of Soft Quest than one in which persons may be more committed to a particular religion (i.e., Christian Soft Quest).

Regarding the other dimensions of religious orientation, Moralistic Interpretation (a Christian Soft Quest subscale) was positively related to intrinsic religiosity, whereas each one of the Universal Soft Quest subscales (Universality and Tentativeness) were negatively related to intrinsic religiosity. The Christian Soft Quest scales (Moralistic Interpretation and Ecumenism) were positively related to personal extrinsic religious motivation (i.e., the utilization of religion for peace, comfort, and relief), and unrelated to social extrinsic motivation (i.e., the utilization of religion to make friends or connect with others). Therefore, within this predominantly Christian sample, it seems that those scoring highly on the Christian Soft Quest subscales tend to be those who "stay close to home" (Beck & Jessup, 2004; p. 290) while being open to growth and religious or spiritual change within a prescribed territory. Further, those high in Christian Soft Quest tend to use religion for personal comfort. Persons higher in Universal Soft Quest are less religiously committed, less defensive, less intrinsically motivated, and (in the case of Universality) less likely to utilize religion for social fulfillment. Despite their differential relations to other religious constructs, each of the subscales within these Soft Quest dimensions (as Edwards et al., 2011 stated) seems to allow for faith as well as doubt and openness.

General Commentary of the Higher Order Structure of Quest

The results of this study support the multidimensionality of Quest, as well as a previously posited higher order dimensionality. However, the place of the higher order dimensionality in the future of Quest research remains somewhat unclear. Interestingly, both Beck and Jessup (2004) and Edwards et al. (2011) seem to generally agree on what types of measures (i.e., subscales or factors) constitute Hard and Soft Quest, as well as what other religious variables are related to each of these higher-order dimensions. However, based on their respective discussions of results, they appear to disagree on which of these dimensions is most consistent with the original conceptualization of Quest. These efforts to align just one higher-order dimension with Batson's original conceptualization of Quest seem unnecessary for two primary reasons.

First, they are both correct. The first proposed iteration of the Interactional Scale (Batson & Ventis, 1982) and its revised successor (Batson & Schoenrade, 1991b) were designed to assess three aspects of Quest. The first (readiness to face existential questions without reducing their complexity) and third (openness to change) aspects seem quite consistent with Beck and Jessup's (2004) conceptualization of Hard Quest, which Beck and Jessup felt was most representative of Batson's conceptualization of Quest. However, the second (self-criticism and perception of religious doubts as positive) and third (openness to change) aspects are quite similar to Edwards et al.'s (2011) conceptualization of Soft Quest, which Edwards et al. felt was most consistent with Batson's conceptualization of Quest. Therefore, in terms of construct coverage, both Beck and Jessup (2004) and Edwards et al. (2011) promoted specific higher-order dimensions of Quest that overlapped, but did not fully encompass what Batson apparently intended to measure. Both Hard and Soft Quest, therefore, seem to have merit. These higher-order dimensions help us to understand, in a meta-dimensional sense, how Quest may manifest itself. Therefore, rather than attempting to juggle eight (or nine) factors, our understanding is that the factors share variance, and may be understood dimensionally (i.e., as Hard and Soft Quest).

Second, I am reminded of the (psychometrically apt) adage, that: "the whole is only as good as the sum of its parts." While consistency with the original conceptualization of Quest is important, we are not only concerned with the scope of the construct (i.e., whether researchers are designing measures that adequately encompass Batson's original conceptualization of Quest). With examinations of the multidimensionality of Quest, we are also chiefly concerned with its depth (i.e., whether existing measures are able to adequately assess the unique dimensions that have been posited for Quest). As can be seen above, the higher order dimensions of Hard and Soft Quest seem to generally coincide with all three aspects of Quest specified by Batson (Batson & Ventis, 1982; Batson & Schoenrade, 1991). These dimensions, however, also have a depth that illuminates our understanding of the manifestation of Quest in the lives of people.

The individual (eight) factors in the current study elucidate the manifestation of Quest in very specific ways. For example, of the religious variables assessed, Complexity (a facet of Hard Quest) was related only to Extrinsic (Personal) religiosity--not Extrinsic (Social). Conversely, Religious Angst (also a facet of Hard Quest) was related to both personal and social extrinsic religiosity. In sum, despite their shared variance within the factor model, each of the factors yields unique information.

Limitations

A few limitations to the current study warrant discussion. First, although this sample is more heterogeneous than others in studies of the multidimensionality of Quest (e.g., Beck & Jessup, 2004), it was mostly Christian, Caucasian, and Female. Of particular interest is the predominance of female participants in the current study. There is evidence that women seem to consider religion more important in their lives than men and tend to engage in religious behaviors more frequently (e.g., prayer and church attendance) (Pew Research Center Forum on Religion & Public Life, 2009). Therefore, the current sample may have yielded above average levels of religiosity than would a strictly gender-balanced sample. A second limitation is that this sample was drawn from a college student population. Regarding the age and context of the sample (primarily young adults attending a university), it is important to note that a significant number of students seem to experience belief change during their college years (e.g., see Edmondson & Park, 2008). Therefore, while this population seems particularly relevant for examinations of Quest, it is unclear how the MQOS might perform in an adult sample of great age and/or outside of the university context. Therefore, based on the results, one cannot claim that the MQOS is inherently comprised of eight (or nine) factors. While responses elicited in validity studies will certainly vary (at least somewhat) across samples, generalization of these results cannot be comfortably drawn beyond college student samples such as those of Beck and Jessup (2004) and the current study.

In addition to this general restriction of age range, some authors (e.g., Donahue, 1985) have questioned the validity of the Quest construct, speculating that Quest is actually sophomoric doubt that would typically be found in young-adult or college-aged populations. While Batson and Schoenrade (1991a) have addressed this concern in their discussion on the validity of the Quest construct (as measured by the Interactional Scale), there seems to be limited research in this area.

Future Directions in Psychometric Research on Quest Motivation

As limited research has been published on the Multidimensional Quest Orientation Scale (MQOS), further investigation is certainly required. However, the current study lends support to the usage of Hard and Soft Quest dimensions in assessing Quest motivation. While individual subscales may yield information pertinent to specific research questions (e.g., assessing religious openness with the Universality subscale), researchers should also further explore the usage of the higher order factors (i.e., the Hard and Soft Quest dimensions) in predicting broad socio-behavioral and emotional outcomes, such as church attendance and psychological well-being.

Confirmatory Factor Analyses should be undertaken in additional samples, including both college-aged participants as well as community samples with a broader range of age representation. In these studies, authors should not only examine the factor structure of the MQOS, but also examine the data for mean differences across samples on the various factors of the MQOS. Further, as has been discussed, the items on the MQOS Exploration subscale seem to be assessing apologetics. Some of the items contain content pertaining to studying one's own faith. It is plausible that some of the participants who responded affirmatively to these items were exploring other religions for defensive, rather than purely exploratory purposes. The correlation of the Exploration subscale with the Defensive Theology Scale (Beck, 2006) seems to lend support to this hypothesis. However, the religious behavior of exploration certainly seems appropriate for inclusion in measures of the Quest construct. Based on the results of the current study alone, the removal of the Exploration subscale from the MQOS cannot be definitively prescribed. Confirmatory factor analytic procedures should be utilized across multiple samples to determine the appropriateness of this scale for inclusion in a multidimensional measure of Quest.

Finally, future research should explore the continued usage of the Christian-specific scales in the MQOS. The Moralistic Interpretation scale, in particular, yielded some unique relationships with other religious variables, including positive correlations with intrinsic religiosity, religious defensiveness, and religious commitment. These relationships are somewhat inconsistent with what would normally be expected from a Quest subscale. Additionally, while Christian-specific subscales may be informative in predominantly Christian samples, their use in more diverse samples is somewhat questionable. Their exclusion would likely result in a more universally useful scale for assessing Quest. As such, future research should explore the factor structure of the MQOS items without the Moralistic Interpretation and Ecumenism items.

References

Allport, G. W., & Ross, J. M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5, 432-443.

Altmeyer, B., & Hunsberger, B. (1992). Authoritarianism, religious fundamentalism, quest, and prejudice. The International Journal for the Psychology of Religion, 2, 113-133.

Armstrong, K. (2009). The case for god. New York, NY: Anchor.

Batson, C. D. (1976). Religion as prosocial: Agent or double-agent? Journal for the Scientific Study of Religion, IS, 29-45.

Batson, C. D., Flink, C. H., Schoenrade, P. A., Fultz, J., & Pych, V. (1986), Religious orientation and overt versus covert racial prejudice. Journal of Personality and Social Psychology, 50, 175-181.

Batson, C. D., Naifeh, S. J., & Pate, S. (1978). Social desirability, religious orientation, and racial prejudice. Journal for the Scientific Study of Religion, 17, 31-41.

Batson, C. D., & Schoenrade, P. A. (1991a). Measuring religion as quest: 1) Validity concerns. Journal for the Scientific Study of Religion, 30, 416-429.

Batson, C. D., & Schoenrade, P. A. (1991b). Measuring religion as quest: 1) Reliability concerns. Journal for the Scientific Study of Religion, 30, 430-447.

Batson, C. D., & Ventis, W. L. (1982). The religious experience: A social-psychological perspective. New York, NY: Oxford University-Press.

Bcck, R. (2006). Defensive versus existential religion: Is religious defensiveness predictive of worldview defense? Journal of Psychology and Theology, 34, 143-153.

Beck, R., Baker, L., Robbins, M., & Dow, S. (2001). A second look at quest motivation: Is quest unidimensional of multidimensional? Journal of Psychology and Theology, 29, 148-157.

Beck, R., & Jessup, R. K. (2004). The multidimensional nature of quest motivation. Journal of Psychology and Theology, 32, 283-294.

Crosby, J. W., & Varela, J. G. (In Press). Preferences for religious help-seeking: Racial and gender differences, interfaith intolerance, and defensive theology. Mental Health, Religion & Culture, (XX), XX-XX.

Donahue, M.J. (1985). Intrinsic and extrinsic religiousness: Review and meta-analysis. Journal of Personality and Social Psychology, 48, 400-419.

Dudley, R. L., & Cruise, R.J. (1990). Measuring religious maturity: A proposed scale. Review of Religious Research, 37, 97-109.

Edmondson, D, & Park, C. (2009). Shifting foundations: Religious belief change and adjustment in college students. Mental Health, Religion & Culture, 12, 289-302.

Edwards, K. J., Hall, T. W., & Slater, W. (2002, August). The structure of the quest construct. Paper presented at the American Psychology Association Annual convention, Chicago, IL.

Edwards, K. J., Hall, T. W., Slater, W., & Hill, J. (2011). The multidimensional structure of the quest construct. Journal of Psychology and 'Theology, 39, 87-110.

Edwards, K.J., Slater, W., Hall, T. W., Oda, A., & Eck, B. (2001, August). Spiritual change among college students: A national study. Paper presented at the 109th Annual Convention of the American Psychological Association, San Francisco, C A.

Gorsuch, R. L., & McPherson, S. E. (1989). Intrinsic/extrinsic measurement: I/E revised and single-item scales. Journal for the Scientific Study of Religion, 28, 348-354.

Pew Research Center, Pew Forum on Religion & Public Life (2009). The stronger sex--spiritually speaking. Retrieved from http://www.pewforum.org/The-Strongcr-Sex----Spiritually-Speaking.aspx

Pew Research Center. (2012). American Values Survey. Retrieved from http://www.people-press.org/vaIues-Questions/q41d/i-never-doubt-the-existence-of-god/#generation

Worthington, E. L., Wade, N. G., Hight, T. L., Ripley, J. S., McCullough, M. E., Berry, J. W., Schmitt, M. M., Berry, J. T., Bursley, K. H., O'Connor, L. (2003). The Religious Commitment Inventory--10: Development, refinement, and validation of a brief scale for research and counseling. Journal of Counseling Psychology, 50, 84-96.

(1.) For ease of examination, Beck and Jessup (2004) displayed the 62 items of the MQOS in a sorted fashion, with the items clustered into their respective subscales. The original order of presentation was not published. Therefore, for the current study, the item-order was randomized.

Appendix A

The Multidimensional Quest Orientation Scale: Randomized Order as Presented to Participants

(1.) I feel that the spiritual meaning of Biblical stories are more important than their historical accuracy. (Moralistic Interpretation)

(2.) Although I feel joy and peace in my spiritual life, I also frequently experience feelings of anxiety and loneliness. (Religious Angst)

(3.) I would characterize my religious beliefs as very complex rather than simple and straightforward. (Complexity)

(4.) I feel that I could serve God being a member of many different kinds of Christian churches and congregations (e.g., Catholic, Baptist, Methodist, Assemblies of God). (Ecumenism)

(5.) I believe a central part of spiritual maturity is growing comfortable with doubt. (Tentativeness)

(6.) I anticipate dramatically changing my religious beliefs in the future. (Change)

(7.) I feel that it takes a lot of time and intensive study to even begin to have an informed opinion about religious issues. (Complexity)

(8.) My faith in God is accompanied by anxiety and doubt. (Religious Angst)

(9.) My religious development has often been filled with doubt and has been troubling at times. (Religious Angst)

(10.) I have often felt abandoned by God during my spiritual journey. (Religious Angst)

(11.) I believe that changing one's religious beliefs is a good sign of spiritual development. (Change)

(12.) I feel that reading the Biblical stories in a literal way misses their deeper spiritual meaning. (Moralistic Interpretation)

(13.) To me, church affiliation (e.g. Catholic, Baptist, Methodist, Assemblies of God) is an irrelevant issue in determining one's salvation. (Ecumenism)

(14.) I believe religious doubts play an important role in spiritual development. (Tentativeness)

(15.) I think the major world religions (e.g., Christianity, Judaism, Islam, Buddhism) are equally valid ways to seek God. (Universality)

(16.) My religious questions have led to deeper questions rather than definitive answers. (Did not load on any factor at [greater than or equal to] .40; developed as a Tentativeness item)

(17.) I have spent more time compared with most people I know investigating the foundations of my religious faith. (Exploration)

(18.) In my effort to seek after God I have spent a lot of time studying the teachings of religions around the world. (Exploration)

(19.) I don't feel a need to know all the answers to every religious question I may have. (Tentativeness)

(20.) What seems to have primarily motivated my religious development is a search for meaning in a seemingly random universe. (Existential Motives)

(21.) There are valuable lessons to be learned from Christian faiths (e.g. Catholic, Baptist, Methodist, Assemblies of God) that are different from my own. (Ecumenism)

(22.) I have been placing a lot of effort in exploring religious questions. (Exploration)

(23.) I consistently explore issues that will deepen my religious faith. (Exploration)

(24.) I think that the doctrinal differences between Christian churches and congregations (e.g., Catholic, Baptist, Methodist, Assemblies of God) are largely irrelevant in God's eyes. (Ecumenism)

(25.) My religious journey has primarily been devoted toward finding a meaning or purpose for my life rather than engaging in traditional religious practices. (Existential Motives)

(26.) Being a Christian is not about being a member of any one Christian faith (e.g., Catholic, Baptist, Methodist, Assemblies of God). (Ecumenism)

(27.) I dislike very philosophical answers to my religious questions, (r) (Did not load on any factor at > .40; developed as a Complexity item)

(28.) I frequently assume that my current religious beliefs may be wrong. (Change)

(29.) A deep understanding of the Bible involves looking past the literal meaning to see the deeper spiritual truth being communicated. (Moralistic Interpretation)

(30.) I believe much of the truth of the Bible is primarily found in reading its stories allegorically. (Moralistic Interpretation)

(31.) I feel that most religious questions do not have simple, straightforward answers. (Complexity)

(32.) I feel like most religious questions involve complex answers that take a lifetime to fully understand. (Complexity)

(33.) My religious searching has been primarily devoted toward finding a meaning or purpose for my life rather than the traditional focus of developing a relationship with a personal God. (Existential Motives)

(34.) I am not disturbed by unanswered questions in my religious life. (Tentativeness)

(35.) I would characterize my religious beliefs as very philosophical in nature. (Did not load on any factor at [greater than or equal to] .40; developed as a Complexity item)

(36.) I don't think one Christian faith (e.g. Catholic, Baptist, Methodist, Assemblies of God) is any more correct when compared to the others. (Ecumenism)

(37.) The major world religions (e.g., Christianity, Judaism, Islam, Buddhism) may take different paths, but each path ultimately leads to God. (Universality)

(38.) Mostly it is spiritually immature people who hold the same religious beliefs for long periods of time. (Change)

(39.) My religious questions have been primarily devoted to exploring my place in the universe rather than about religious doctrines and belief systems. (Existential Motives)

(40.) Heaven is open to people of all world religions (e.g., Christianity, Judaism, Islam, Buddhism). (Universality)

(41.) I am comfortable leaving many of my spiritual questions unanswered. (Tentativeness)

(42.) It would be hard for me to express my religious views in a short amount of time due to the complexity of the arguments I would give. (Complexity)

(43.) I don't believe God approves of any one Christian church or congregation (e.g., Catholic, Baptist, Methodist, Assemblies of God) over another. (Ecumenism)

(44.) I believe spiritual growth requires consistent change in one's religious beliefs. (Change)

(45.) I would characterize most of my religious behavior as a "search for truth." (Exploration)

(46.) I have often felt lost and alone during my spiritual journey. (Religious Angst)

(47.) My religious journey has more been abstract and philosophical than the more traditional religious efforts to develop a relationship with a personal God. (Existential Motives)

(48.) I believe that the more spiritually mature I become I will discover more questions than answers. (Tentativeness)

(49.) I feel that spiritually mature people struggle with doubts. (Tentativeness)

(50.) I would characterize my religious life as one of consistent searching and exploration. (Exploration)

(51.) I feel that most things in religion are clear and easy to understand, (r) (Did not load on any factor at [greater than or equal to] .40; developed as a Complexity item)

(52.) I don't think it really matters what church (e.g. Catholic, Baptist, Methodist, Assemblies of God) a person attends as long as they love and serve God. (Ecumenism)

(53.) I think changing one's religious beliefs over time is a sign of spiritual maturity. (Change)

(54.) I believe that consistently questioning my current religious beliefs will promote spiritual growth. (Change)

(55.) A primarily literal reading of the Bible may miss its deeper truths. (Moralistic Interpretation)

(56.) I feel that it is naive to expect definitive answers to deep religious questions. (Tentativeness)

(57.) I understand that most of my religious questions cannot be answered. (Tentativeness)

(58.) A primarily literal reading of the Bible is an overly simplistic way of understanding the meaning of its stories. (Moralistic Interpretation)

(59.) I often question if some of my most central religious beliefs are wrong. (Change)

(60.) I believe that Christianity is the only way to know God and receive salvation as opposed to other world religions (e.g. Judaism, Islam, Buddhism), (r) (Universality)

(61.) Spiritual maturity involves changing one's religious beliefs over time. (Change)

(62.) I would mostly describe my spiritual journey as a "struggle." (Religious Angst)

James W. Crosby

Sam Houston State University

Correspondence concerning this article should be addressed to James Crosby, Department of Psychology and Philosophy, MS 2447, Sam Houston State University, Huntsville, TX, 77341, crosby@shsu.edu

The author would like to thank Matthew Light for his assistance with this project.

Author Information

CROSBY, JAIMES W. Ph.D. Address: Sam Houston State University, Department of Psychology & Philosophy, MS 2447, Huntsville, TX, 77341. Title: Assistant Professor of Psychology. Degrees: Ph.D., Oklahoma State University; M.S., B.S., Abilene Christian University. Specializations: Psychology of religion, peer victimization across the lifespan, and school psychology.
COPYRIGHT 2013 Rosemead School of Psychology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Author:Crosby, James W.
Publication:Journal of Psychology and Theology
Article Type:Report
Geographic Code:1USA
Date:Sep 22, 2013
Words:10932
Previous Article:Negative views of parents and struggles with God: an exploration of two mediators.
Next Article:Social personality traits as salient predictors of religious doubt phenomena among undergraduates.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters