Protestant Millennials, Religious Doubt, & the Local Church

Abstract

Millennials are the most analyzed and populous generation in the United States. Collectively, they have been slowly re-shaping the American culture. Protestant Millennials, a subset of this generation, have been ruffling feathers in their local churches. Many, who once regularly attended, are leaving. Unwise responses by local church leaders to their young parishioners’ doubting habits significantly contributed to the departure. This study pursued a sample of college-aged Protestant Millennials to know them in a psychological sense. The intentions were twofold: to discover social personality traits that predict their doubting practices and to develop practical and proactive relational strategies for local church leaders. Self-report data on personality features and doubt phenomena were obtained from 532 religiously committed undergraduates in the United States. Results from multivariate regression procedures revealed three social personality dispositions contributed to the prediction of the doubt constructs. Implications and applications of the findings are discussed.

1. Introduction

Millennials are the most analyzed and populous generation in the United States1 (Fry 2016). Collectively, they have been slowly transforming American culture. Oxford Dictionaries designated Millennials’ trademark word, ‘selfie,’ as the Word of the Year in 2013 (Brumfield 2013). Several fast-food companies (e.g., Wendy’s, Hardee’s) have started serving sandwiches on non-traditional breads, a response to Millennials’ foodie preferences (Horowitz 2013). Even, Fiat Chrysler is wooing this generation with a concept car for the family. The semi-autonomous, electric Portal, was designed “by Millennials for Millennials” (Snider and Snavely 2017, p. 5B).

Protestant Millennials, a subset of this influential generation, have been ruffling feathers in their local churches2. Many are leaving. Survey research on this population revealed 59% to 70% have stopped attending church in the United States (Kinnaman 2011; McConnell 2007). The unwise responses by local church leaders to the doubting habits of these young parishioners is a significant reason for the departure. Interview data indicated the Protestant Millennials had several complaints. The freedom to voice questions about the Christian faith was limited in their local churches. Disclosures of doubts were met with trite responses by older Christians (e.g., Baby Boomers) who also seemed doubtless and judgmental (Kinnaman 2011).

How might Protestant church leaders in the United States respond to this feedback? Would ‘knowing’ Millennial parishioners in a psychological sense help? Can knowledge of social personality traits that predict doubting practices assist? If so, what practical applications from the psychological information can aid local church leaders who seek to practically and proactively engage their Millennial attenders and members?

The following discussion intends to answer the above inquiries. Pertinent information about religious doubt and the doubt phenomena in Batson’s quest religious orientation (QRO) are explicated (Batson 1976). A rationale for multivariate regression analyses of doubt and personality constructs from a sample of religiously committed undergraduates, American Protestant Millennials, is delineated. Moreover, practical applications derived from the predictive associations form a practical and proactive strategy for local church leaders.

2. Religious Doubt & Quest Religious Orientation

Religious Americans doubt; it is a common habit. Smith reported from a national survey that 72% of religious individuals’ faith contains doubt (Smith 1998). Among religious Millennials, Smith and Snell found 52% in mainline Protestant churches, 46% in conservative Protestant churches, and 34% in Black Protestant churches confess having ‘some to many’ doubts (Smith and Snell 2009). Yet, religious doubt remains an uncomfortable phenomenon, often misunderstood. When religious people disclosed views on doubt, they described it as: traumatic, shameful (Strobel 2000), unsafe (Moreland and Issler 2008), scary (Patton 2010), sinful (Kinnaman 2011), and the opposite of faith (Habermus 2013).

According to Laurie, doubt is often mistaken for unbelief (Laurie 2016). Boshart, Jr. wrote, “The Bible mentions two kinds of unbelief. One kind is doubt” (Boshart 2008, p. 1). Buchanan discussed Thomas’ dismissal of his peers’ account of Christ’s resurrection (John 20). He argued, “[Thomas] doubted, not to excuse his unbelief, but to establish robust belief” (Buchanan 2000, p. 64). Unfortunately, it appears the authors interchanged doubt and unbelief as synonyms.

Others appraise and discuss doubt more positively and more precisely. Allport, a psychologist, maintained people who understand their religious doubting experience are in a better position to determine the vibrancy of their own beliefs and disbeliefs (Allport 1950). Fowler, a psychologist, tagged doubt as a necessity for faith development (Fowler 1996). Osborne, a theologian, considered it a natural effect of human finitude (Osborne 2010). Hecht, a historian, argued doubt has had an inspirational effect throughout the ages (Hecht 2003). Moreover, several authors proffer ideas that have potential to reduce the misunderstanding surrounding religious doubt. These efforts include: defining doubt from an interdisciplinary vantage, considering religious doubt in the midst of a complex process, and suggesting how religious people juggle doubt and belief.

Psychologists and theologians have studied and written extensively on the topic of religious doubt for several decades. Puffer opined for a definition of doubt, synthesized from the two disciplines (Puffer 2013). An interdisciplinary definition has the potential to afford a more comprehensive conceptualization. A portion of his suggested integrative definition was:

Allport, a psychologist, defined religious doubt as a hesitant reaction (Allport 1950). It is a state of uncertainty or a questioning of religious tenants according to psychologists, Hunsberger, McKenzie, Pratt, and Pancer (Hunsberger et al. 1993). Beck, a psychologist and theologian, noted the commitment to certain religious beliefs is suspended when people religiously doubt (Beck 1990). One of the biblical Greek words for doubt, diakrino, illumines the ‘hesitant reaction’ with the notions of self-disputation and separation. Moo, a New Testament scholar, explained a doubter is “disputing with oneself; it is a basic conflict of loyalties” (Moo 2015, p. 87). Zodhiates, a biblical Greek scholar, added diakrino entails a separation. An individual is “separating [ideas or options]—one from the other” such as the possible from the impossible3.

(Zodhiates 1966, p. 38)

Concisely summarized, religious doubt, as a cognitive phenomenon, is a state of hesitation. A religious doubter is uncertain and questioning religious tenants. Commitment to religious conviction is suspended. Moreover, in the hesitant reaction, a religious doubter is self-disputing or separating options.

Krause and Ellison regarded religious doubt as a “part of a larger process that unfolds over time” (Krause and Ellison 2009, p. 293). The cognitive construct is not a stand-alone phenomenon. It is a part in a complex series with three interrelated experiential aspects. First, there is a precipitant (i.e., a cause). Triggers of doubt are often intrapersonal or interpersonal influences. Allport argued religious doubt emerges when new information collides with old knowledge or beliefs (Allport 1950). Krause and Ellison researched social conflicts (e.g., religious friends being overly critical) precipitating doubt. Second, there is the aforementioned cognitive state of hesitation or questioning (Allport 1950; Hunsberger et al. 1993). Third, there is a coping reaction. Religious people respond to doubt. Some cope by resolving it (e.g., seeking spiritual growth) or suppressing it (i.e., denial). Last, there is an outcome. Krause and Ellison noted health and psychological effects (i.e., noxious or salubrious). These relate to the harmful or helpful outcomes experienced by religious people who doubt. For instance, youth tend to have more distress related to religious doubting than the elderly (Krause et al. 1999).

Smith suggested a manner in which religious persons grapple with their religious doubts and beliefs. He wrote: “Faith is fraught; confession is haunted by an inescapable sense of its contestability. We don’t believe instead of doubting; we believe while doubting” (Smith 2014, p. 4). People can doubt and believe in a simultaneous fashion. They do not have to stop one, to do the other.

Following Smith’s rationale, it is probable that religious persons handle their disbelief in the same manner as they doubt and believe. For example, this author maintains that Thomas in John 20 struggled with disbelief while believing. In verse 25, he heard his peers’ testimony of a risen Christ. The apostle quickly retorted, “Unless I shall see in His hands the imprint of the nails…and put my hand into His side, I will not believe” (New American Standard). Kostenberger remarked that Thomas’ response was forceful in nature. It can be translated, “I will ‘certainly not’ believe” (Kostenberger 2004, p. 577). Some have argued this follower of Christ was in a state of doubt (MacArthur 2008). If so, why is Thomas’ response void of hesitation or self-disputation? The apostle does not retort with uncertainty; he did not declare that he might or might not believe. If the apostle did not doubt in verse 25, does that mean he stopped believing everything he believed about the Messiah? Not likely. In verse 28, Thomas encountered Jesus face-to-face as had his peers. He quickly confessed belief in Christ’s resurrection with the statement, “My Lord and My God!” It seems plausible that Thomas believed in Christ’s divinity while simultaneously having disbelief, albeit temporary, in Christ’s resurrection.

In the 1970s, C. Daniel Batson designed the quest religious orientation (QRO) (Batson 1976). A key prompt for the development of the new construct was a complaint. Batson argued that certain Allportian traits of the mature religious sentiment were absent in Allport and Ross’ operationalization of the intrinsic religious orientation (Allport and Ross 1967). In the explanation for QRO, Batson, Schoenrade, and Ventis delineated the three missing Allportian traits. The trio became the three components in the new construct (Batson et al. 1993). Important to the purposes of this study are the connections between Batson’s three sub-dimensions of QRO and features in Krause and Ellison’s doubt process and Puffer’s suggested ‘integrative’ definition of religious doubt (Puffer 2013; Krause and Ellison 2009; Batson et al. 1993).

The first segment of QRO is readiness. According to Batson et al., mature religious people are ready to face existential struggles without minimizing the complexity in the problems (Batson et al. 1993). They courageously engage the complicated nature of life by responding to common, everyday hassles along with chronic, unrelenting difficulties.

Readiness appears to operationalize the ’integral trait’ in Allport’s religious sentiment (Allport 1950). Mature religious individuals seek to harmoniously integrate their human experiences into a belief system. In those efforts, they address and attempt to reconcile discrepant situations (e.g., problem of evil). This penchant is often evident through the practice of generating questions (Batson 1976). Such queries are not posed to simply obtain information. Complex, existential concerns about purpose in life, everyday struggles, or tragedies are often reservations, controversies, challenges, and objections related to religious themes4. For instance, a Christian undergraduate might affirm, “God wasn’t very important to her until she began to ask questions about the meaning of her life” (Batson and Schoenrade 1991, p. 436). Moreover, Batson’s readiness resembles the second part in Krause and Ellison’s model, the cognitive state of hesitating, questioning, or self-disputing (Allport 1950; Hunsberger et al. 1993; Moo 2015; Krause and Ellison 2009).

The second segment of QRO is openness. Batson et al. indicated religious persons are open to an ongoing search for religious knowledge (exploration). They are also willing to absorb the new information (change) (Batson et al. 1993). Openness seems to mimic Allport’s ‘comprehensive trait’ of religious maturity. People lack contentment unless they deal “with matters central to all existence” (Allport 1950, p. 69). Hence, they explore new ideas and tweak their doctrine when needed.

Using the schema of Krause and Ellison, the change dimension in Batson’s openness appears to be a reaction to doubt, a coping response (Krause and Ellison 2009). There is a proclivity to do something that is either potentially adaptive or maladaptive. For instance, a quest-oriented individual might affirm, “There are many religious issues on which my views are still changing” (Batson and Schoenrade 1991, p. 436). The exploration aspect mimics the precipitant component in the doubt process (Krause and Ellison 2009). Newly explored information can become a challenge to a current belief system that in turn can potentially cause doubt.

The third segment of QRO is doubt positivism. The feature reveals a willingness “to doubt and be self-critical” (Batson et al. 1993, p. 166). It appears to operationalize Allport’s ‘heuristic trait’ in the religious sentiment. Beliefs can be held tentatively until confirmed. Mature religious people with a quest orientation can “entertain religious propositions [and] act wholeheartedly without absolute certainty” (Allport 1950, p. 74). Batson’s doubt positivism identifies an outcome in the process of doubt. It is a level of comfortability with hesitancy, uncertainty, and self-criticism (Allport 1950; Hunsberger et al. 1993; Batson et al. 1993). It can be considered a psychological effect, a mindset revealing ease with the cognitive phenomenon (Krause and Ellison 2009). For example, a quest oriented individual might affirm, ‘Doubting is an important part of what it means to be religious” (Batson and Schoenrade 1991, p. 436).

When Batson operationalized the quest religious orientation, he created a measure, the Quest Scale (QS), blending readiness, openness, and doubt positivism (Batson and Schoenrade 1991). A factor analysis revealed a one-factor structure as the robust conclusion indicating the three segments were to be a unitary construct (Batson and Schoenrade 1991; Beck et al. 2001; Edwards et al. 2002). Using Krause and Ellison’s schema, the QS appears to assess participants’ self-report of the cognitive state of questioning along with two experiential aspects in the doubt process—a coping reaction to doubt and a psychological outcome of doubt (Krause and Ellison 2009).

Some researchers question the utility of a one-dimensional operationalization of religion as quest (Edwards et al. 2002). Their concerns are specifically with the total score in measures of quest religiosity (e.g., the QS). Most of the correlational data reported in the research literature involves a composite score (Crosby 2013). Leak regarded the QS total score as ambiguous. He suggested the use of the sub-dimensional scores in place of the total score to enhance discriminate validity (Leak 2011). Beck, Baker, Robbins, and Dow were bothered by the loss of predictive validity. The three segments of QRO “may have different relationships with certain religious variables” than the QS total score (Beck et al. 2001, p. 155). Hence, Beck and Jessup pursued a multidimensional operationalization creating the Multidimensional Quest Orientation Scale (MQOS) (Beck and Jessup 2004).

A comparison of the QS and MQOS reveals a few important distinctions. The former has 12 questions, four items per sub-dimension and one composite score (Batson and Schoenrade 1991). The MQOS has nine subscales; five directly measure Batson’s QRO with 38 items (Beck and Jessup 2004). Each MQOS subscale has a ‘stand-alone’ total score. For the readiness segment of QRO, there are thirteen items encompassed in two MQOS subscales. There are fifteen questions in the openness segment involving two MQOS subscales. In the doubt positivism segment, there are ten items encompassed in one MQOS subscale. Last, unlike the total score in QS of Batson’s unidimensional model, Beck and Jessup’s multidimensional model allows for the isolation and systematic assessment of each segment in quest religious orientation (Beck and Jessup 2004).

3. Empirical Investigations of QRO & Personality Traits

Halfaer maintained, “The religious belief system is an ideal vehicle for studying complex personality processes” (Helfaer 1972, p. 5). Examination of the research literature on quest religious orientation (i.e., a one-dimensional model) and personality studies reveals three trends. Findings are either not significant, conflicting, or recurring with limited repetition. Using the Myers-Briggs Type Indicator (MBTI), Francis and Ross found no significant differences in QRO scores among the four Jungian preferences (Myers and McCaulley 1985; Francis and Ross 2000). Yet, Ross and Francis reported quest scores were statistically higher among people high in the intuition function than the sensing preference. The three other preferences (i.e., extraversion and introversion, thinking and feeling, judging and perceiving) were unrelated to QRO (Ross and Francis 2010).

Employing the Eysenck Personality Profiler (EPP), Hills, Francis, Argyle, and Jackson stated QRO was unrelated to extraversion and related to low psychoticism (Eysenck et al. 1992; Hills et al. 2004). However, Francis noted quest religion was associated with low extraversion and not associated with psychoticism (Francis 2010). With the Big Five Inventory (BFI), Henningsgaard and Arnau found QRO related to high neuroticism, high openness, and low conscientiousness (John et al. 1991; Henningaard and Arnau 2008). Robbins, Francis, McIlroy, Clark, and Pritchard reported QRO unrelated to all five categories (Robbins et al. 2010).

Neuroticism has been one notable recurring (albeit limited) personality construct. In the EPP, both Hills et al. and Francis found it positively related to QRO (Hills et al. 2004; Francis 2010). In the BFI, Henningsgaard and Arnau noted the same (Henningaard and Arnau 2008). Although the definitions of neuroticism in the EPP and BFI are similar, each measure has different sub-scales comprising neuroticism with the exception of anxiety. Henningsgaard and Arnau underscored that the traits in the BFI explained 8% of the variance in quest religiosity. This outcome was due to the positive associations with neuroticism and openness along with the negative relation with conscientiousness (Henningaard and Arnau 2008).

Some research has been conducted using a multidimensional model of QRO. Constructs studied with Beck and Jessup’s MQOS include: attachment—insecure and secure (Beck 2006), psychological distress—depression, anxiety, stress (Messay 2010), and religious life—intrinsic and extrinsic orientation (Crosby 2013; Beck and Jessup 2004), spiritual well-being, orthodoxy (Beck and Jessup 2004), religious commitment, defensive theology (Crosby 2013), forgiveness (Messay 2010), religious reflection, religious schema, religious fundamentalism, and biblical fundamentalism (Watson et al. 2014). There is scant research examining the subscales of MQOS and personality models. Puffer investigated the ‘exploration’ sub-dimension (i.e., the level of activity focused on expanding one’s religious knowledge) in MQOS and social personality traits (Puffer 2013; Beck and Jessup 2004). He found two dispositions predicting exploration when he used the California Psychological Inventory (CPI) (Gough 1987).

Unfortunately, exploration is only one of five subscales in the MQOS that specifically operationalizes Batson’s three sub-dimensions in quest religiosity (Batson et al. 1993; Beck and Jessup 2004). The other four, existential motives, complexity, change, and tentativeness, have not been investigated with a personality model. The present study addressed this particular gap in the literature by extending Puffer’s study (Puffer 2013). This extension suggests another personality model, CPI, can potentially expand the role that personality traits play in religious orientation variables (Gough 1987; Miller and Worthington 2012). Soto and John underscored the popularity, age (i.e., five decades), and attractive longitudinal data of the CPI (Soto and John 2009). In research of personality and religious orientation, it is often underused. The extension would also afford an opportunity to examine the benefit of a multidimensional model for future studies investigating personality traits and QRO.

With a sample of Protestant Millennial undergraduates in the United States, two research questions were pursued. First, what combination of nine social personality traits (SPT) selected from the CPI predicts existential motives, complexity, change, and tentativeness of the MQOS (Beck and Jessup 2004; Gough 1987)? Second, how can American Protestant church leaders (e.g., elders, mentors) practically and proactively apply the social-personality and doubt information as they relationally engage their Millennial parishioners (Kinnaman 2011)? Based on Puffer’s outcomes, it was anticipated that eight of the nine selected SPTs from the CPI (e.g., tolerance, independence) would have a positive relationship with the four aforementioned MQOS variables. A ninth SPT, good impression, would have a negative association (Puffer 2013).

4. Method

4.1. Participants and Procedure

Participants were undergraduates (n = 532) enrolled in a private Christian university located in the Midwest of the United States. The students comprised of 341 females (64%) and 189 males (36%) with an average age of 20 years5. Self-identified ethnic affiliations include: 93% European American, 2% Hispanic Americans, 1% African Americans, 1% Asian Americans, and 3% other or undesignated.

Concerning religious landscape, 68% of the respondents (n = 532) indicated a high importance for religion and 56% identified being a Protestant. Regrettably, an administrative oversight occurred and these demographic questions were not obtained from participants (n = 167) in the first year of the investigation. Yet, when the questions were included in the second and third years, 99% of the participants who were added (n = 365) stated religion was very important and 81% indicated being a Protestant. All participants signed a Christian lifestyle agreement for the duration of their enrollment at the university. The covenant involves a pledge to adhere to a Christian code of conduct (e.g., social-moral ethics, biblical religious practices, positive conflict resolution, and sexual purity).

Concerns about respondents’ ‘religious conformity’ were explored. Three religious constructs investigated in this study were compared to the same variables in Crosby’s study that used a sample from a public, non-religious institution (Crosby 2013). Selected constructs included religious complexity (i.e., preference for complex religious views, not naïve ones), religious tentativeness (i.e., positivity toward religious doubt), and religious angst (i.e., experience of negative emotions and social isolation in a religious pilgrimage) (Beck and Jessup 2004). The results from independent T-tests revealed this study’s means were significantly higher than the means reported in Crosby’s study6. The finding suggests the participants in this study had a higher preference level for critical religious thinking, had a higher level of comfortability with religious uncertainty, and indicated more negative social-emotional experiences in their religious journey than participants from a public, non-religious university7.

The study transpired over three years. The Human Subjects Review Board of the university granted approval for this investigation8. College students with different majors (i.e., psychology, Christian ministry) were recruited from a variety of classes. Extra credit was offered for voluntary involvement. After a brief overview of the study, each undergraduate received a packet containing the consent form, a demographic questionnaire, and the measures of religious doubt and social personality traits. Completed packets were returned to the author.

4.2. Measures

4.2.1. Religious Doubt

To assess doubt phenomena embedded in Batson’s quest religious orientation, the Multidimensional Quest Orientation Scale (MQOS) was employed (Batson 1976; Beck and Jessup 2004). The 62-items have a seven-point Likert scale for response and are organized into nine components. Only four of the five sub-dimensions operationalizing Batson’s QRO were utilized in this study. The other four subscales (i.e., religious angst, ecumenism, universality, and moral interpretation) appraise additional features in people’s experience with religion as quest (Beck and Jessup 2004).

For this study, existential motives (EXM) and complexity (COM) operationalized the readiness segment in Batson’s QRO. These two sub-dimensions in the MQOS also represented the cognitive state of hesitation or questioning in Krause and Ellison’s schema (Krause and Ellison 2009). EXM has five items gauging the level to which religious actions are regarded as a drive to discover purpose in life (e.g., “My religious questions have been primarily devoted to exploring my place in the universe rather than about religious doctrines and belief systems”) (Beck and Jessup 2004, p. 293). COM has eight items indicating the amount of importance for embracing complex views on religious beliefs versus naïve ones (e.g., “I feel that most religious questions do not have simple, straightforward answers”) (Beck and Jessup 2004, p. 293).

Change (CHAN) operationalized the openness segment in Batson’s QRO. The MQOS subscale also represented a coping reaction to doubt as noted in Krause and Ellison’s model. CHAN has nine items that assesses the level of willingness to evaluate and modify present day religious belief (e.g., “Spiritual maturity involves changing one’s religious beliefs over time”) (Beck and Jessup 2004, p. 292). The pursuit of new religious information or exploration in openness was not addressed in this study.

Tentativeness (TEN) operationalized the doubt positivism segment within Batson’s QRO. Ten items appraise the degree of positivity toward doubt (e.g., “I believe a central part of spiritual maturity is growing comfortable with doubt”) (Beck and Jessup 2004, p. 292). TEN, also, represented an outcome of doubt, a psychological effect as mentioned in Krause and Ellison’s schema.

Last, regarding psychometrics, Crosby’s recent factor analysis of the MQOS supported the structure of the measure. He concluded the “subscales are assessing distinct aspects of quest” (Crosby 2013, p. 218). Obtained reliability coefficient for each variable (see Table 1) indicated good internal consistency that is consistent with the reports by Beck and Jessup and by Crosby (Crosby 2013; Beck and Jessup 2004).

Table 1. Intercorrelations, Means, Standard Deviations, and Reliabilities of the Variables for Religious Doubt and Social Personality Traits.

	COM	EXM	TEN	TCH	DO	CS	SY	SP	IN	EM	GI	TO	F/M
COM	--
EXM	0.29	--
TEN	0.28	0.17	--
TCH	0.30	0.30	0.31	--
DO	−0.06	−0.10	−0.03	−0.14	--
CS	0.02	−0.10	0.00	−0.06	0.48	--
SY	−0.01	−0.04	0.02	−0.06	0.61	0.56	--
SP	−0.05	−0.01	0.02	−0.05	0.50	0.50	0.72	--
IN	−0.10	−0.17	−0.80	−0.13	0.64	0.48	0.52	0.53	--
EM	−0.00	−0.07	0.08	−0.04	0.44	0.60	0.56	0.51	0.36	--
GI	−0.14	−0.20	−0.11	−0.19	0.14	0.19	−0.01	−0.13	0.17	0.11	--
TO	−0.08	−0.27	−0.06	−0.15	0.18	0.34	0.15	0.10	0.20	0.32	0.42	--
F/M	0.12	0.07	0.06	0.08	−0.25	−0.06	−0.18	−0.27	−0.32	−0.12	−0.05	0.05	--
Mean	32.72	18.35	46.67	24.51	51.60	47.08	50.16	45.37	47.31	50.43	49.89	48.72	48.71
SD	6.67	5.55	9.05	10.35	11.58	8.88	9.41	12.01	8.81	8.72	8.40	7.68	11.22
Alpha	0.67	0.77	0.76	0.89	0.79	0.58	0.75	0.69	0.70	0.58	0.77	0.68	0.63

Note: n = 532; COM = Complexity; EXM = Existential Motives; TEN = Tentativeness; TCH = Transformed Change (CHAN); DO = Dominance; CS = Capacity for Status; SY= Sociability; SP = Social Presence; IN = Independence; EM = Empathy; GI = Good Impression; TO = Tolerance; F/M = Femininity/Masculinity; Underlined & bold “r” = p ≤ 0.05; bold “r” = p ≤ 0.01 and unmarked “r” = non-significant “r,” alphas for COM, EXM, TEN, and TCH were obtained Cronbachs; and alphas for DO, CS, SY, SP, IN, EM, GI, TO, and F/M are from Gough (Gough 1991).

4.2.2. Social Personality Traits

The California Psychological Inventory (CPI) was utilized to measure social personality traits. This instrument has 462-items with a true-false response format. Items are organized into twenty scales assessing the “processes of interpersonal life” (Gough 1987, p. 1). According to Soto and John, the CPI is “an existing measure widely used in personality research for more than a half century” (Soto and John 2009, p. 25). Reported psychometrics for the CPI are robust (see Table 1).

As in the study by Puffer, nine CPI scales were selected. Dominance (DO) notes a self-assured, assertive leadership. Capacity for status (CS) describes an ambitious, ascendant penchant. Sociability (SY) indicates an extroverted conviviality with others. Social presence (SP) points to an energetic and confident social engagement. Independence (IN) highlights determination blending autonomy and social competency. Empathy (EM) underlines a perceptive, social suaveness capable of detecting thoughts and feelings of others. Good impression (GI) emphasizes a conscientious desire to present well to others. Tolerance (TO) reveals a non-judgmental approach with others that is reasonable and diplomatic. Femininity/masculinity (F/M) refers to a social sensitivity with an androgynous bent (Gough 1987; Puffer 2013).

The aforementioned dispositions were chosen for three reasons. An extension of Puffer’s study requires duplication of his measures. Nine SPTs are more manageable than twenty and these nine traits do not envelop intrapersonal features as in the case with the remaining eleven (e.g., Well-Being stressed self-contentment and comfort) (Gough 1987).

5. Results

Table 1 presents descriptive statistics for the CPI and the MQOS scales. The means for the nine SPTs vary between 45 and 52 T. This range is in the middle of four interpretative categories—low (30–45 T) and high (55–70 T) (McAllister 1988). Each of the means for the four MQOS subscales fall within Beck’s stated range of one standard deviation in his original study with Jessup9. The obtained alpha coefficients for the MQOS subscales and published alphas for the CPI scales indicate good internal consistency (Gough 1987).

Pearson product-moment correlations were conducted for the CPI and MQOS variables. The matrix is also presented in Table 1. In general, the SPTs were negatively related with small coefficients or unrelated to the quest religiosity scales. The exception was the small positive correlation between femininity/masculinity (F/M) and complexity (COM).

Assumptions for multivariate regression procedures were evaluated. Missing data were managed using listwise deletion in SPSS. According to Allison, this technique can yield ample estimates on uncertainty and minimal bias. It is also robust in regression procedures (Allison 2001). Outliers among the thirteen variables were identified and eliminated utilizing published score ranges. In this study, one possible cause for outliers may have resulted from the hand-scoring process. Multicollinearity was checked via the correlation matrix in Table 1. One skewed distribution, change (CHAN—positive), was successfully transformed using a logarithmic procedure to improve normality, linearity, and homoscendasticity (Tabachnick and Fidell 1996).

Standard multiple regression analyses were performed to pursue “optimal predictions” of the influence of social dispositions on each quest religiosity subscale in the MQOS (Allison 1999, p. 3). Table 2 displays the results. Initially, five predictors emerged—two positive and three negative ones. As expected, good impression (GI) negatively predicted complexity (COM) and change (CHAN). Capacity for status (CS) and empathy (EM) positively predicted COM and tentativeness (TEN), respectively. Unexpectedly, tolerance (TO) and independence (IN) were negative predictors for existential motives (EXM).

Table 2. Standard Multiple Regression Analyses for the Influence of Social Personality Traits on Religious Doubt Phenomenon.

Construct	Variable	B	SE B	Beta	p-Value	F	p-Value	R2
Religious Doubt—EXM (n = 501)						6.21	<0.01	0.10
	DO	0.02	0.03	0.04	0.58
	CS	−0.01	0.04	−0.01	0.82
	SY	−0.02	0.04	−0.03	0.71
	SP	0.05	0.03	0.11	0.13
	IN	0.10	0.04	−0.15	0.02
	EM	0.03	0.04	0.04	0.52
	GI	−0.04	0.03	−0.06	0.26
	TO	−0.17	0.04	−0.24	0.00
	F/M	0.03	0.02	0.05	0.26
Religious Doubt—COM (n = 502)						2.95	0.002	0.05
	DO	0.01	0.04	−0.01	0.88
	CS	0.11	0.05	0.15	0.03
	SY	0.03	0.05	−0.04	0.63
	SP	−0.07	0.04	−0.13	0.08
	IN	−0.05	0.05	−0.06	0.36
	EM	0.04	0.05	0.05	0.47
	GI	−0.11	0.04	−0.14	0.01
	TO	−0.05	0.05	−0.06	0.26
	F/M	0.05	0.03	0.09	0.08
Religious Doubt—TCH (n = 509)						3.72	<0.01	0.05
	DO	−0.00	0.00	−0.10	0.14
	CS	0.00	0.00	0.08	0.24
	SY	−1.44	0.00	−0.00	0.99
	SP	0.00	0.00	−0.03	0.69
	IN	−0.00	0.00	−0.04	0.52
	EM	0.00	0.00	0.04	0.50
	GI	−0.00	0.00	−0.16	0.00
	TO	−0.00	0.00	−0.09	0.10
	F/M	0.00	0.00	0.05	0.35
Religious Doubt—TEN (n = 509)						1.94	0.045	0.02
	DO	−0.00	0.05	−0.01	0.93
	CS	−0.00	0.07	−0.01	0.94
	SY	0.01	0.07	0.01	0.86
	SP	−0.01	0.06	−0.01	0.89
	IN	−0.09	0.07	−0.08	0.24
	EM	0.16	0.07	0.15	0.02
	GI	−0.10	0.06	−0.10	0.07
	TO	−0.05	0.06	−0.04	0.45
	F/M	0.05	0.04	0.05	0.27

Note: COM = Complexity; EXM = Existential Motives; TEN = Tentativeness; TCH = Transformed Change; DO = Dominance; CS = Capacity for Status; SY = Sociability; SP = Social Presence; IN = Independence; EM = Empathy; GI = Good Impression; TO = Tolerance; F/M = Femininity/Masculinity.

Table 3 shows outcomes from additional multiple regression procedures with the five aforementioned obtained predictors. First, EXM (existential motives) was regressed on just intolerance (-TO) and dependence (-IN) together. The explained variance was R² = 0.08 relative to R² = 0.10 in the model with nine SPTs in Table 2. In the regression models predicting EXM with the obtained predictors separated, the explained variances were R² = 0.07 for -TO and R² = 0.03 for -IN. A test for interaction was pursued. Due to the multicollinearity among the predictors and the product term, centering methods were employed (Tabachnick and Fidell 2007). Yet, the centered product term was not statistically a significant predictor of EXM (Williams 2015).

Table 3. Additional Standard Multiple Regression Analyses for Obtained Social Personality Traits Predictors on Religious Doubt Phenomenon.

Construct	Variable	B	SE B	Beta	p-Value	F	p-Value	R2
EXM—two SPTs (n = 502)						23.3	<0.01	0.08
	TO	−0.18	0.03	−0.25	0.00
	IN	0.07	0.03	−0.11	0.02
EXM—one SPT (n = 502)						40.59	<0.01	0.07
	TO	−0.20	0.03	−0.25	0.00
EXM—one SPT (n = 503)						14.20	<0.01	0.03
	IN	−0.11	0.03	−0.17	0.00
EXM—interaction test (n = 502)						15.76	<0.01	0.08
	TOcent	−0.18	0.03	−0.25	0.00
	INcent	−0.07	0.03	−0.10	0.02
	TOINprod	0.00	0.00	0.03	0.44
COM—two SPTs (n = 503)						5.37	0.005	0.02
	CS	0.04	0.03	0.05	0.27
	GI	−0.12	0.04	−0.15	0.00
COM—one SPT (n = 503)						0.27	0.602	−0.001
	CS	0.02	0.03	0.02	0.60
COM—one SPT (n = 504)						9.76	0.002	0.02
	GI	−0.11	0.04	−0.14	0.00
TCH—one SPT (n = 511)						20.04	<0.01	0.04
	GI	−0.00	0.00	−0.19	0.00
TEN—one SPT (n = 511)						2.95	0.09	0.004
	EM	0.08	0.05	0.08	0.09

Note: COM = Complexity; EXM = Existential Motives; TEN = Tentativeness; TCH = Transformed Change; CS = Capacity for Status; IN = Independence; EM = Empathy; GI = Good Impression; TO = Tolerance; TOcent = centered Tolerance; INcent = centered Independence; and TOINprod = the centered product term—centered Tolerance and centered Independence.

Second, in the regression model predicting complexity (COM) with only the two independent variables together (+CS & −GI), the explained variance was R² = 0.02 relative to the R² = 0.05 in the model with the nine SPTs in Table 2. In the regression models predicting COM with the SPTs separated, the results revealed R² = 0.02 for −GI and non-significance for +CS. Since +CS dropped as a significant predictor, there was no need to test for interaction.

Third, TCH (change) was regressed on just non-conformity (-GI). In this regression model, the explained variance was R² = 0.04 relative to R² = 0.05 in the model with nine SPTs in Table 2. Fourth, TEN (tentativeness) was regressed on empathy (+EM). In this final regression model, the result revealed EM as non-significant and no longer a viable predictor.

Hence, the multiple regression procedures in this study revealed that three SPTs contribute to prediction of the subscales in the MQOS. All were negative. Intolerance (-TO) and dependence (-IN) predicted existential motives (EXM) and non-conformity (-GI) predicted complexity (COM) and change (TCH).

6. Discussion

Kinnaman encouraged American Christian churches to be proactive with the Millennials in their congregations (Kinnaman 2011). The present study pursued a sample of religiously committed undergraduates, Protestant Millennials. Obtained personality information relative to their religious doubting experiences led to the development of relational strategies for local church leaders. The suggested actions offer practical and proactive ideas on how church leadership can respond to their young parishioners’ doubt. The findings expanded the role of personality traits in research with religious orientation constructs. The use of a multidimensional model of QRO revealed additional insight into the personality-religion relationship.

6.1. Personality Traits Predicting Readiness in QRO

Intolerance, dependence, and non-comformity emerged as predictors of the readiness segment in Batson’s QRO. In this sub-dimension, religious people courageously face existential struggles (EXM) and do not minimize the complexity (COM) in problems (Batson et al. 1993). Unexpected results were intolerant (-TO) and dependent (-IN) traits as predictors of religious behavior driven to discover purpose in life (EXM). The anticipated finding revealed disregard for making a good impression or non-conformity (-GI) predicting the level of importance for complex views on religious knowledge (COM) (Beck and Jessup 2004; Gough 1987).

6.1.1. Implications for Intolerance and Dependence

Intolerance and dependence as predictors of existential motives (EXM) corroborate and conflict with previous research. Watson et al. reported quest predicted intolerance (Watson et al. 1999). Yet, Batson et al. indicated that QRO was negatively related to intolerance (Batson et al. 1993). Nesbit characterized Millennials, in general, as tolerant. This generation tends to agree to disagree and offers freedom for contrary views (Nesbit 2010). In regards to dependence, Hills et al. found this trait positively related to QRO (Hills et al. 2004). Wiley noted QRO positively associating with an external locus of control with powerful persons (Wiley 2006). However, Hall and Baym opined relationships are “ongoing collaborative [dependence] and individual [independence] processes of finding balance within holistic tensions” (Hall and Baym 2011, p. 318).

Of special note, the explained variance (R²) with intolerance and dependence predicting EXM was 8–10% (see Table 2 and Table 3). This finding supports the use of the California Personality Inventory (CPI) and the Multidimensional Quest Orientation Scale (MQOS) in research of personality and religious orientation. First, this particular R² is comparable to Henningaard and Arnau’s results. Neuroticism, openness, and conscientiousness in the Big Five Inventory (BFI) predicted QRO (i.e., the one-dimensional model) with a R² of 8% (Henningaard and Arnau 2008). Second, there is an expansion in the number of personality traits explaining the variance with quest religiosity. The CPI traits add to the list contributed by the aforementioned personality models (MBTI, EPP, & BFI). Intolerance is added along with dominance, empathy, and masculinity/femininity from Puffer’s study (Puffer 2013). Third, clarity and specificity is increased when a multidimensional measure of QRO is employed. The isolation and independent assessment of existential motives (EXM) from the readiness segment in Batson’s QRO clarifies which specific feature in the one-dimensional model associates with the social personality traits.

Intolerance and dependence may appear as odd, almost contrasting traits that contribute to the prediction of doubt phenomena. Yet, the two make sense in a context when American Protestant Millennials are asking existential questions. This is the cognitive state of hesitation or questioning, the second part in Krause and Ellison’s process of doubt (Krause and Ellison 2009). Undergraduates reflect on the meaning of life. They wonder their place in God’s universe (Beck and Jessup 2004). Often, the religious status quo is examined and possibly challenged.

Furthermore, an intolerant social presentation displays as faultfinding, suspicion, narrow-mindedness, judgment, hostility, superiority, and mistrust10 (Gough 1987; McAllister 1988). There is a vexing dissatisfaction with something. The discontent is likely noticeable to others in a religious community and can motivate intolerant individuals to exclude or push people away (Barnes et al. 2017). A dependent social presentation displays as submission, timidity, conciliatoriness, anxiety, quick capitulation, caution, and convention10 (Gough 1987; McAllister 1988). It exposes an intense relational need fraught with insecurities. A passionate reliance on community can draw people close, but it may hinder efforts to obtain valuable insight, direction, and encouragement (Hall and Baym 2011).

In the cognitive state of hesitation, it is possible that intolerance contributes to the prediction of existential questioning by laying bare an unspecified dissatisfaction. The discontent can simultaneously draw attention and push others away. Yet, the trait may fuel a vigorous examination of the status quo—current religious beliefs and practices. For dependence, it is plausible that the trait contributes to the prediction of existential questioning by pursuing relational connections. The strong reliance is riddled with insecurities, but the disposition can help American Protestant Millennials attach to a community that potentially can assist them in addressing and resolving their existential questions.

6.1.2. Applications for Intolerance and Dependence

How can American Protestant local church leaders practically and proactively engage Millennials during existential questioning? Covey, a leadership educator, argued effective leaders make it a priority to understand others first and then seek to be understood (Covey 1989). A potential strategy for understanding and engaging Protestant Millennials includes five responses—pursue, empathize, validate, self-disclose, and probe. Church leaders can intentionally pursue relationships with Millennial parishioners. This generation, in general, wants connections with people including those who are older than they. According to Nesbit, Millennials desire “mentors [who can] speak into their life [being frank and honest]…and can challenge and support [them]” (Nesbit 2010, pp. 130–31). They expect these relationships to be bidirectional and mutually transparent (Sbanotto 2012). When equality and candor are obviously active, the friendship can become a safe and trustworthy engagement prompting commitment between the participants (Hall and Baym 2011).

In conversations with Millennial members or attenders, church leadership (i.e., elders, deacons, and volunteer mentors) need to empathize and validate. Empathy entails listeners grasping the message of a storyteller and then articulating understanding of the message (Egan 2010). To grasp a message, leaders need to be quick to listen to the words (i.e., verbal content) and non-verbal content (e.g., facial expressions, emotions). Then, they express what was heard. They identify the speaker’s concerns and the emotions attached to the complaints. Validation involves listeners genuinely confirming or authenticating the storyteller’s message as an acknowledgment of the worth or sensibility in the opinions, stories, assumptions, and emotions. Listeners do not have to completely agree with the comments or ideas that seem unreal or untrue to them. They do need to communicate the merit in discussing or examining the belief or feeling in a sincere and respectful manner (Rodriguez 2017).

Potentially, the chafing effects of an intolerant social presentation could tempt church leaders to react instead of responding. Intolerant American Protestant Millennials may rudely push people away, demand others to be like them, or control others through power (Barnes et al. 2017). An empathic and validating response offers kindness in the face of seemingly prickly incorrigibleness. Such a reply demonstrates an authentic desire to understand others first (Covey 1989). It also has great potential to generate a strong bond or connection between leadership and young parishioners.

Nesbit mentioned Millennials’ preference for transparency in relationships (Nesbit 2010). This creates an opportunity for church leaders to self-disclose struggles in their own faith pilgrimage. The vulnerability and authenticity can be both attractive and didactic (Sbanotto and Blomberg 2016). Protestant Millennials can vicariously learn from the mistakes/successes of more experienced followers of Christ, which strengthens the importance of leaning into others in a healthy manner (Hall and Baym 2011). The personal stories of leaders can help normalize existential questioning, avoid the appearance of being doubtless, and potentially open opportunities for Millennials to self-disclose more of their doubts (Kinnaman 2011; Sbanotto and Blomberg 2016).

When sufficient trust has developed and in the context of a one-on-one discussion, the senior leaders and volunteer mentors of a Protestant church community can probe for root causes in the vexing dissatisfaction. Possible queries to their Millennials include: Is the discontent personal? Are they wondering about glitches in their religious views/beliefs? Are they pondering better options? Is the dissatisfaction with others? Are they bristling against inauthentic and irrelevant practices/doctrines in the church (Kinnaman 2011)? Have peers, often a source of knowledge, confused them (Kinnaman 2011)? Have others labeled their new theological views falsehoods and are they wondering what to do with those opinions (Vicari 2014)? Such inquiries can model genuine concern for a Protestant Millennial, a willingness to assist them in a struggle, and demonstrate boldness in the pursuit of matters of the heart (Hall and Baym 2011). The answers to the aforementioned questions can create a sense of vulnerability since the young parishioners may be releasing personal secrets. It is important that church leaders confidentially hold these self-disclosures and not misuse the information.

6.1.3. Implications for Non-conformity

Non-conformity (-GI) predicting complexity (COM) may also be considered a surprising outcome. Some support for this finding exists in the research literature. Puffer reported -GI as a predictor of the total score in the Quest Scale (Puffer 2013).

In COM, people prefer complex religious perspectives instead of naïve ones. They value intellectual matters and express attraction to philosophical dimensions in religious beliefs. There is a willingness to invest time to study resources that can provide clarity (Beck and Jessup 2004). Complexity like existential motives, associates with Batson’s readiness in QRO and with Krause and Allison’s cognitive state of hesitation or questioning (Batson et al. 1993; Krause and Ellison 2009).

According to Gough and McAllister, a non-conforming social presentation appears as sarcasm, criticalness, rebellion, mutableness, tactlessness, non-conciliatoriness, and bluntness10 (Gough 1987; McAllister 1988). It reveals a penchant of non-responsiveness to social expectations. Presenting well to others and being cooperative and polite are low priorities (McAllister 1988; Sadler et al. 2010). Efforts to manage others’ impressions through reparative tactics (i.e., being apologetic) are most likely overlooked (Lee et al. 1999).

Yet, a non-conforming social self has another side. This disposition can also empower religious people to disregard imagined or real pressure from others to remain naïve and uninformed in religious matters (Puffer 2013; Gough 1987). Hence, in the cognitive state of hesitation, non-conformity may contribute to the prediction of complexity through the values of genuineness and congruency instead of phoniness (Batson et al. 1993). These particular social values can reduce some of the social distractions that can hinder the nurturance of complex religious views.

6.1.4. Applications for Non-conformity

How can local American Protestant church leaders practically and proactively engage their Millennials in their preference for complex views on religious knowledge? Another possible strategy for understanding and engaging includes three responses—refrain, suggest, and offer (Covey 1989). Older or seasoned Christians often have a habit of correction. They particularly correct youths’ opinions that appear foolish to them (Barnes et al. 2017). In those moments when ‘correction’ seems necessary, church leaders need to refrain from lecturing, giving ultimatums (i.e., demands for agreement), and coming across condescending (Kinnaman 2011). Such restraint builds relational connections and communicates respect, love, and kindness. The restraint follows the Royal Law. That decision can set a foundation for the moments when leaders seek to be understood in a non-manipulative manner.

Pastors and elders can suggest resources for complicated religious topics. They can offer a selection of authors who hold orthodox and unorthodox views. The suggestions can be followed with an invitation to process the content while getting coffee. Each discussion needs to be approached with an authentic and teachable posture. This entails the honesty of leaders about their own flaws and limits in their knowledge base (Sbanotto and Blomberg 2016). The resources offer American Protestant Millennials a way to avoid cursory, rootless religious knowledge. They would be reading credible resources of religious views, critically thinking about the ideas to ascertain validity, and discussing their opinions and reactions with a safe, older, and mature follower of Christ (Kinnaman 2011; Beck and Jessup 2004).

Building on the attraction for religious complexity, church leadership might offer an ‘apprenticeship’ to their Millennials (Batson et al. 1993; Beck and Jessup 2004). According to Willard, a philosopher and theologian, participants journey together for one to three years (Willard 2009). The partners learn together through experience (i.e., events and activities), reason (i.e., reflection and application of the information from authors), and authority (i.e., a study of the Bible). Participants humbly request their partner—to teach them and help them recognize blind spots hindering spiritual growth. This intentional and mutual connection allows both senior and younger parishioners to speak into the other’s life and offer wisdom and encouragement (Sbanotto and Blomberg 2016).

6.2. Personality Trait Predicting Openness in QRO

One social personality trait (SPT) emerged as a predictor of Batson’s openness segment in QRO. Non-conformity (-GI) also predicted change (TCH). This expected finding means disregard for making a good impression was predictive of willingness to evaluate and modify present day religious beliefs (Batson et al. 1993). Moreover, TCH resembles a coping reaction to doubt that has the potential to be adaptive or maladaptive (Krause and Ellison 2009).

6.2.1. Implications for Non-Conformity

As previously mentioned, non-conforming (-GI) persons tend to be disinterested in presenting well to others or maintaining the status quo (Sadler et al. 2010; Grant and Mayer 2009). Nesbit identified Millennials, in general, as possessing a penchant for being unique. This drive encompasses a rejection of their adult caregivers’ bent toward conformity (Nesbit 2010).

The tasks associated with religious change (i.e., evaluation and modification) match well with observed features in a non-conforming social presentation. In the practice of evaluating religious beliefs, observed propensities in non-conformity such as criticalness and rebellion can enable religious individuals to judge the value of current religious tenants and habits (Gough 1987; McAllister 1988). Criticalness can fuel the analytical process. Rebellion can help release any social constraints or pressure to maintain the religious status quo in their religious community.

For modification of religious beliefs, observed personality tendencies in non-conformity such as mutableness and non-conciliatoriness can empower people to boldly transform their religious beliefs and habits in a context of social disapproval (Gough 1987; McAllister 1988). In particular, mutableness can slacken any strain in the decision-making process and offers freedom to switch directions. Non-conciliatoriness can help individuals avoid foreclosure and prioritize personal ownership of religion over social conformity.

When the obtained social personality traits (SPTs) in the California Psychology Inventory (CPI) predicting QRO in the present investigation and Puffer’s study are analyzed together, non-conformity (-GI) emerged three times (Puffer 2013). This trait was a statistically significant predictor for change (TCH), complexity (COM), and the total score in the Quest Scale (R² = 0.04)11. The re-occurrence and statistical significance also support the use of the CPI and the Multidimensional Quest Orientation Scale (MQOS) in research of personality and religious orientation. Although explained variance is small, a discernable pattern has emerged with the CPI measuring personality traits with quest religiosity. A similar pattern is evident with the BFI (i.e., neuroticism) (Henningaard and Arnau 2008). The MQOS provides increased clarity and specificity when investigating QRO with personality traits. The isolation and independent assessment of TCH from the openness segment in Batson’s QRO clarifies which specific feature in the one-dimensional model associates with the social personality traits.

6.2.2. Applications for Non-Conformity

How can the local Protestant church in the United States practically engage their Millennials during a religious change process? Another possible strategy for understanding and engaging entails four responses—remember, extend, assist, and forewarn (Covey 1989). Leadership needs to remember that although non-conformity can appear immature to them, there is an adaptive side. This social self may be intentional, a desired impression (Goleman 2006). Instead of faking good, the Protestant Millennials are working hard at being genuine and real among fellow followers of Christ (Batson et al. 1993).

Barnes, Childerston, and Rosenau, theologically trained psychologists, described God’s connecting grace. Christ received people without conditions and altered His life (i.e., death on the cross) for others to join Him (Barnes et al. 2017). Likewise, church shepherds can extend Christ’s graciousness to non-conforming Protestant Millennials. They can take the time that is needed to understand the young parishioners and learn from them. Leaders also need to be willing to change or alter themselves (e.g., knowledge base, religious habits) and trust God to work in the life of the Millennials independent of their wisdom. Applying God’s connecting grace is risky. There is the potential to be misunderstood by others in all generations—Silent Generation, Boomers, and Millennials (Barnes et al. 2017).

Nesbitt Sbanotto and Blomberg, seminary professors, argued a safe “relational space” is required for the expression of religious doubt (Sbanotto and Blomberg 2016, p. 240). When Millennial parishioners are in the process of change, pastors, elders, and volunteer mentors can assist by asking non-threatening, well-timed questions. Examples include: What beliefs/practices seem wrong? What new beliefs/practices have emerged? Do you need assistance with these beliefs? Are the changes consistent with biblical moorings or different? How do significant others respond to these changes? How are you (the Millennial member/attender) handling those responses? Furthermore, leadership needs to keep answers confidential and not misuse the information.

Finally, senior pastoral directors can forewarn, at the appropriate time, about potential glitches that can emerge in a religious change process. Although Millennials, in general, value change and distaste stagnation (Nesbit 2010), tweaking of doctrine/practices might germinate cognitive dissonance (Festinger 1957). Ample time may be needed to sort through confusion; they may need to order and rank valued preferences and consequences, associated with a religious alteration. There also may be discomfort during cognitive dissonance. Protestant Millennials should be warned in advance to not let this be an impetus to rush through the experience of ordering and ranking preferences.

6.3. Limitations and Future Research

The present investigation was not without limitations. First, the regression procedures preclude cause and effect conclusions. Second, regarding participant selection, individuals in the sample needed to approximate some of the values embraced by local Protestant church leaders (Kinnaman 2011). Millennial undergraduates from a private Christian university would most likely describe themselves as Protestant and affirm a lifestyle prioritizing Christian values (e.g., religion is important) and biblical practices (e.g., local church attendance). Hence, generalities derived from the outcomes are limited to populations with similar characteristics of the sample. Third, only a few features of religious doubt and the process that relate to or encompasses this cognitive phenomenon were studied. Other aspects—Krause and Allison’s precipitants of doubt, Guinness’ different kinds of doubt, or Ellison and Lee’s problematic religious doubting—were not investigated (Krause and Ellison 2009; Ellison and Lee 2010; Guinness 1976). Last, the explained variances associated with the statistically significant social personality predictors of doubt constructs were small (i.e., R² ranged 2–10%). According to Miller and Worthington, the range for variance for religious and spiritual variables is typically ten percent or less (Miller and Worthington 2012). With quest religiosity, the Big Five Inventory accounted for 8% of explained variance (Henningaard and Arnau 2008). Yet, the obtained personality traits in this study still play an important role in these religious variables and can offer useful information for American church leadership in ‘knowing’ Protestant Millennials (Miller and Worthington 2012; Allison 1999).

These limits notwithstanding, the importance of understanding the doubting habits of American Protestant Millennials needs to continue as a research priority. Future studies can pursue ‘everyday’ data on their doubt. Applications of the Ecological Momentary Assessment (EMA) methodology can obtain and analyze such information. It would be important to discover the frequency of daily questing/doubting, the kinds of daily doubts, how religious people juggle the triad of doubt, belief, and unbelief, and the important dimensions in problematic doubting (Runyan et al. 2013). Future investigations can also replicate the present investigation with some of the aforementioned limitations addressed. Replications can check the stability of the results in this investigation. A study could discover how social personality traits and other variables (e.g., social interest, hyper-competitiveness) predict the meta–dimensions of quest religiosity, soft and hard quest, as suggested by Crosby and Beck and Jessup (Crosby 2013; Beck and Jessup 2004).

7. Conclusions

Kinnaman urged American Christian churches to pursue two objectives. First, they need to “cultivate a ‘new mind’ for understanding and discipling” [their Millennial parishioners]. Second, they “need to move from thinking and talking to doing and changing” (Kinnaman 2011, p. 213). The findings and applications from this study aid in these efforts.

Three personality traits—intolerance, dependence, and non-conformity from the California Personality Inventory (CPI)—were discovered to be predictors of doubt phenomena. The trio supported the role that CPI personality dispositions can play in religious orientation variables (Gough 1987; Miller and Worthington 2012). The doubt constructs—existential motives, religious complexity, and religious change—were from Beck and Jessup’s Multidimensional Quest Orientation Scale (Beck and Jessup 2004). The doubt variables associate with Batson’s readiness and openness in quest religiosity and Krause and Allison’s the cognitive state of hesitation and a coping reaction to doubt (Krause and Ellison 2009; Batson et al. 1993). The findings in this study also underlined benefits in using a multidimensional model of QRO in future investigations of personality traits and religious orientation constructs.

Intolerance, dependence, and non-conformity as social personality traits offer a unique view into the doubting habits of American Protestant Millennials who are religiously committed undergraduates. Several suggestions were offered on how to carefully apply this knowledge in authentic relational strategies. Such knowledge has great potential for Protestant church leaders to engage Millennial-aged parishioners in vibrant local church interactions and ministry (Sbanotto and Blomberg 2016).