Intellectual Humility: Scale Development and Theoretical Elaborations in the Context of Religious Leadership.

McElroy, Stacey E. ; Rice, Kenneth G. ; Davis, Don E. 等

This article focused on how perceptions of intellectual humility (IH)--humility regarding one's knowledge or influence over ideas--affect relationships with religious leaders. We developed an informant report measure of IH perceptions using exploratory (Study 1; N = 213) and confirmatory (Study 2; N = 213) factor analyses, as well as a basic manipulation of LH (Study 3; N = 139). Then in Study 4 (N = 105), we examined IH in the context of a major betrayal by a religious leader (i.e., aligning with several factors theorized to strain the practice of IH). The results provide preliminary evidence for the psychometric soundness of the scale, including reliability and content validity of the scores. The scale was able to distinguish between IH and other constructs. Furthermore, the results provide initial evidence that IH is related to social bonds, as perceptions of IH were related to trust and higher forgiveness toward the religious leader, and positive attitudes towards the Sacred.

We would like to acknowledge the generous financial support of a grant from the Fuller Theological Seminary / Thrive Center in concert with the John Templeton Foundation, Grant No. 108 (Intellectual Humility in Religious Leaders), as well as the John Templeton Foundation (Grant No. 29630, The Development, Validation, and Dissemination of Measures of Intellectual Humility and Humility; Grant No. 14979, Relational Humility: An Interdisciplinary Approach to the Study of Humility). The opinions expressed in this publication are those of the author(s) and do not necessarily reflect the views of the Fuller Thrive Center or the John Templeton Foundation.

Whereas philosophy and theology have studied virtue for thousands of years, psychology has only recently joined the conversation. The positive psychology movement has begun to close this gap for some virtues (Peterson & Seligman, 2004). Although humility has been historically understudied, researchers have begun investigating this construct (e.g., Emmons, 2013; Exline & Hill, 2012; Nielsen, Marrone, & Slay, 2010). However, intellectual humility (IH), one of several epistemic virtues being studied by philosophers (e.g., Roberts & Wood, 2007), has not been studied in psychology. Thus, in the present article, we define IH and extend a model of humility to describe its function within relationships, and then attempt to adduce initial evidence for the key proposition that IH helps re ate social bonds.

Definitions

Humility involves (a) an accurate or moderate view of one's strengths and weaknesses as well as being (b) interpersonally other-oriented rather than self-focused, marked by the ability to restrain egotism (i.e., self-oriented emotions such as pride or shame) in ways that maintain social acceptance (Davis et at, 2011). Whereas humility refers to a variety of domains, intellectual humility (IH) pertains to one's knowledge or intellectual influence. Namely, IH involves having (a) insight about the limits of one's knowledge, marked by openness to new ideas; and (b) regulating arrogance, marked by the ability to present one's ideas in a non-offensive manner and receive contrary ideas without taking offense, even when confronted with alternative viewpoints. The difference between humility and IH is the specificity of IH, much like verbal intelligence is theorized to be a sub-domain of general intelligence. Both humility and IH are fundamentally relational in nature because they involve regulating interactions with others. However, IH specifically refers to humility regarding one's beliefs and

. For example, many religions encourage humility as a virtue. Although one may present as humble in most aspects of their life, one might become particularly opinionated and offensive when discussing religion with a member of another denomination or faith. We posit that intellectual humility is especially pertinent anytime there is a competition or negotiation of ideas in a relationship or group. Intellectually humble people regulate their concern for being "right."

They remain open to new ideas, and incorporate various sources of information. They base their beliefs on sound evidence, and they adjust their beliefs according to new evidence. Interpersonally, they also present evidence for their ideas fairly, rather than using manipulative strategies to influence decisions. Roberts and Wood (2003) posited that IH is especially important for individuals viewed by their communities as intellectually talented, accomplished, and skilled, such as leaders. Their power allows them to exert greater influence over ideas within a group, and requires IH for them to maintain fair exchange of ideas within the community.

Intellectual Humility in Relationships

Extending a model of relational humility (Davis et al., 2011), we conceptualize IH as a trait that helps people predict how they will be treated by a target person. Personality judgments store and communicate information about a target person's reputation. Being known as intellectually arrogant hurts one's opportunities for cooperation, whereas being known as intellectually humble can allow one to leverage trust in order to form and strengthen relationships (Baumeister, Bauer, & Lloyd, 2010).

To judge a trait accurately, one must have an opportunity to observe trait relevant behavior (Funder, 1995). Davis et A. (2013) theorized that humility is most accurately judged in situations that evoke egotism (e.g., involvement in escalating conflict, or engagement in a power struggle). That is, humility and its sub-domains are best assessed under strain. One avenue to judge IH well is to see how someone responds to situations that generally provoke intellectually arrogant behavior. This conceptualization builds on theorizing that virtues involve strengthening one's capacity for self-regulation. Like a "muscle" that is temporarily weakened after use but can be strengthened over time (Baumeister et al., 2010; Muraven & Baumeister, 2000), one can develop habits that promote intellectually humble behavior, even in the face of situational factors that cause most people to arrogantly advance their own perspective without sensitivity to alternative viewpoints.

There are numerous contexts that may make IH difficult to practice, and we briefly highlight four here. The first involves discussion of ideas that are linked to identity--used by groups as a signal of loyalty. The second is the experience of negative moral emotions (e.g., contempt or disgust), which narrow thinking and action tendencies, amplify cognitive biases, cause arrogant or stubborn behavior, and promote a sense of moral superiority that decreases empathy and facilitates aggression (Fredrickson, 2001; Westen, Bla-gov, Harenslci, Kilts, & Hamann, 2006). The third is negotiation of a meaningful decision, especially when individuals disagree with one another and are both emotionally invested in the outcome. The fourth is imbalance of power in relationships in which a leader holds influence over ideas and their exchange (Roberts &Wood, 2003). Religious leaders are especially notable examples because they not only possess public influence but are also trusted to explain "God's will."

Intellectual Humility in Leaders

Davis et al. (2013) proposed that humility judgments regulate the strength of social bonds (commitment and trust are close proxies for the strength of a social bond). Social bonds cause people in interdependent relationships to react to each other's needs as their own (Brown & Brown, 2006). Strong social bonds cause people to not only sacrifice, but to enjoy sacrificing (for a review, see Stanley, Rhoades, & Whitton, 2010). Thus, bonds require precise regulation, otherwise one will be exploited by those who draw upon strong bonds but do not reciprocate.

Perceptions of IH may be especially important in leader-subordinate relationships. Leadership researchers have contrasted transactional leadership (i.e., offering rewards for effort) with transformational leadership (i.e., providing a larger vision so that people develop a social bond with their organization, and serving their group becomes a key part of their identity). Instead of simply being paid with money, they are paid with meaning and belonging. Transformational leadership has been linked with various positive outcomes such as trust (Goodwin, Whittington, Murray, & Nichols, 2011) and organizational commitment (McMurray, Pirola-Merlo, Sarros, & Islam, 2010). Accordingly, subordinates may be vulnerable to a leader who can draw upon strong bonds without reciprocating. An example might be a narcissistic leader who is adept at drawing admiration from followers, but leaves a trail of relationship problems in her or his wake. We posit that IH helps balance the leader's power. When subordinates perceive a leader as an unfair negotiator of ideas, they will communicate offenses to each other and spread negative information that will affect the leader's reputation. However, if leaders are able to display IH, they will likely be perceived as more democratic and fair, thus protecting their reputation.

Accordingly, we theorize that a key function of IH is to prevent relational wear-and-tear, like oil prevents an engine from overheating. We call this the social oil hypothesis. Leaders are selected based on traits that promote competition, such as intelligence, assertiveness, or work ethic (Waldron, 2012). However, these qualities do not necessarily lead to high quality close relationships. For example, CEOs are more likely than others to have marital problems (Meers & Strober, 2009). Collins (2001) identified "great leaders" who not only led their companies into a period of enhanced productivity, but the companies thrived after they left their post. These leaders had a unique blend of drive and humility. Thus, IH may provide a healthy climate where ideas are freely and fairly exchanged. Indeed, Owens and Hekman (2012) linked humility with a variety of positive organizational outcomes. However, there has been no work specifically on LH in the context of subordinate-leader relationships.

The Present Study

Given the lack of research on IH in leaders, the purpose of the present set of studies is to examine preliminary evidence for the hypothesis that perceptions of IH regulate other relational constructs. There are no existing measures of intellectual humility, so in the first three studies we developed a measure of perceptions of IH, using exploratory (Study 1) and confirmatory (Study 2) factor analyses, as well as a basic experimental manipulation (Study 3). In Study 4 we examined perceptions of IH of a religious leader following a major betrayal. Religious leadership is a relationship particularly prone to exploitation, given that such individuals are often viewed as having a privileged relationship with the Sacred (which is important for followers' identity). Given this status, their perspective often carries more weight than other members of the congregation, and it can be especially disruptive when they violate trust. For this reason, we expect evaluations of IH to be associated with appraisals of relational spiritualty such as desecration and anger towards God (Exline, Park, Smyth, & Carey, 2011; Mahoney, Rye, & Par-gament, 2005). Finally, to the degree that judgments of IH regulate social bonds, we expect to see 1H negatively associated with levels of unforgiveness.

Study 1: Factor Structure of the Intellectual Humility Scale

The purposes of Study 1 were to (a) determine the factor structure of the Intellectual Humility Scale (IHS); (b) winnow items to create a brief, face-valid measure of IH judgments; and (c) provide initial evidence of estimated internal consistency of the scale and subscales.

Method

Participants. Participants (N = 213) were a corn-m unity sample (158 women). Ages ranged from 18-72 (M = 36.15; SD = 13.32). The sample was primarily Caucasian (73.0

), followed by 11.7% Black/African American, 7.3% Asian/Asian American, 4.0% Latino/ Latina, and 4.0% Other or did not report.

Measure. We generated a list of 60 face-valid items based on our conceptualization of IH. We created the items to reflect intrapersonal and interpersonal aspects of IH. Example items included Is interested in alternative viewpoints," "Is open to other's ideas," and "Always has to have the last word in an argument." Participants rated the IH of a parent by indicating their agreement with items using a 5-point rating scale ranging from 1 = strongly disagree to 5 = strongly agree.

Procedure. Participants were recruited through Amazon's MTurk and participated in exchange for a small monetary compensation. Participants were required to be at least 18 years old and reside in the United States. After giving consent, participants were randomly assigned to rate their mother or father (53.2% rated mothcr; 39.5%, father; 6.5%, primary caregiver, if not raised by biological parent) on the IH items. Because we wanted to develop an informant report measure of IH, the participants were asked to rate a caregiver rather than themselves. We chose the parental relationship because we wanted someone that participants not only knew very well, but also might vary widely in terms of degree of relationship quality. The prompt at the beginning of the survey simply instructed the participants to "Rate your agreement with each item regarding the person you are rating." After completing questionnaires, participants were debriefed.

Results and Discussion

The correlation matrix for all humility items was analyzed using Principal Axis Factor Analysis estimation with an oblique (Promax) rotation. Although orthogonal rotation is more widely used, we employed oblique rotation to allow factors to correlate with one another (Reise, Waller, & Comrey, 2000). This is a more appropriate method in social sciences research, because behaviors may not be independent from one another (Costello & Osborne, 2005). Parallel analysis was used to help determine the number of factors to extract (see Steger, 2006), which suggested a three-factor solution. We retained a two-factor solution, however, because the third factor did not have enough items (N = 2) to make a reliable subscale. These two items were dropped rather than incorporated into the two remaining subscales. After examining the content of items, the factors were named Intellectual Openness (I0) (e.g., "Is open to others' ideas") and Intellectual Arrogance (IA) (e.g., "Has little patience for others' beliefs."). Items on the IA subscale were reverse coded. The first and second factors accounted for 56.95% and 6.94% of the variance in items, respectively. Items were dropped that did not load at least .50 on their primary factor, or that loaded over .25 on any secondary factor.

The final version of the Intellectual Humility Scale (IHS) consisted of 16 items, with two factors that are theoretically consistent with our definition of IH. The JO subscale contained items that aligned with the first part of our definition, which focused on having insight into the limits of one's knowledge, evidenced by an openness to new ideas (e.g., "Is good at considering the limitations of their perspective; Is open to competing ideas"). The IA subscale contained items that aligned with the second part of our definition, which focused on being able to regulate arrogant behaviors, evidenced by the ability to present one's ideas in a non-offensive manner even when confronted with alternative viewpoints (e.g., "Often becomes angry when their ideas are not implemented; Becomes angry when their advice is not taken"). We ran a second exploratory factor analysis after dropping items, in order to estimate descriptive statistics and factor loadings (see Table 1). The Cronbach's alpha coefficients were .94 for JO (7 items) and .93 for IA (9 items). The two factors were strongly correlated with each other, r = .74, p < .001. These findings provide initial evidence that the IHS has a two-factor structure. Because items were winnowed based on the characteristics of one sample, we sought to confirm the factor structure in a different sample in Study 2.

Study 2: Cross-Validation of Factor Structure

Study 1 yielded a two-factor solution for a 16-item scale of IH. The purposes of Study 2 were to (a) replicate the two-factor structure of the IHS using a different sample and (b) provide additional evidence of the estimated internal consistency of the scale and subscales.

Method

Participants and procedure. Participants (N = 213) were a community sample (118 women) recruited through Amazon's MTurk. Ages ranged from 19 to 71 (M = 34.78; SD = 12.12). The sample was mostly Caucasian (82.2%); 5.6 % were Black/African American, 3.8% Asian/Asian American, 2.3% Latino/ Latina, and 6.1% Other or did not report. We used the same procedure as in Study 1.

Measure. Participants completed the 16-item IHS developed in Study 1. For the current sample, the Cronbach's alpha coefficients were .93 for the IA subscale and .92 for the 10 subscale. To further assess reliability, we also calculated construct reliability (CR) for the IA (CR = .92) and 10 (CR = .92) subscale scores (CR corrects for attenuation by modeling measurement error; see Fan, 2003). The subscales were strongly correlated with each other, r = .73,p < .001.

Results and Discussion

The covariance matrix was analyzed with MLR estimation using Mplus 6.1 (Muthen & Muthen, 2008). Items of the IHS were used as indicators of the IA and JO factors, which were modeled as correlated. Several fit indices were examined to evaluate the overall fit of the model--the Chi-square value, the comparative fit index (CFI), the square-root-mean-residual (SRMR), and the root-mean-square-error-approximation (RM-SEA). As a rule of thumb, a CFI around .95, an SRMR equal or less than .08, and an RMSEA equal or less than .06 suggest good fit (Hu & Bender, 1999). We found that the two-factor model showed good fit, x2(89) = 106.23, p < .001, CFI= .98, SRMR = .04, RNISEA = .03. Factor loadings ranged from .68 to .88. Furthermore, because the correlations between the subscale scores were high, we calculated average variance extracted (AVE) to examine the discriminate validity of the scores. The AVE of the JO subscale score (.63) and the IA subscale score (.61) were both higher than the squared correlation between the subscales (.53), indicating adequate discriminate validity (see Fornell & Larcker, 1981). Thus, the results of Study 2 provide additional evidence for the two-factor structure and internal consistency of the IHS.

Study 3: Differentiating between Virtues

The primary purpose of Study 3 was to conduct a basic experiment to explore whether the IHS detects differences between known groups. Namely, participants were randomly assigned to think of someone most or least virtuous in one of three areas: IH, modesty, or drive. We hypothesized that the IHS would do a better job distinguishing between nominees in the IH conditions than for the modesty or drive conditions. We also examined how perceptions of IH were related to relational constructs and other judgments of personality. Related to our theorizing that 1H should help regulate social bonds, we hypothesized that IH would be positively related to trust. We also hypothesized that IH would be positively related to agreeableness and openness, consistent with the content of the two subscales of the IHS: Intellectual Arrogance and Intellectual Openness.

Method

Participants. Participants were 139 undergraduates (99 women, 71.2%) from a large urban university in the Southeastern region of the United States. The mean age was 25.11 years (SD = 5.92). The sample was ethnically diverse (32.4% Black/African American; 32.4% White; 15.1% Asian/Pacific Islander; 12.2% Latino/a; 5.8% Multiracial; 0.7% American Indian; and 1.4% did not respond). Most participants identified as Christian (71.2%), followed by Muslim (2.9%), Hindu (4.3%), Buddhist (3.6%), and Jewish (0.7%); 17.2% reported no religious affiliation or did not respond. Finally, most participants identified as heterosexual (86.3%) followed by gay or lesbian (3.0%), bisexual (2.2%), and no response (8.6%).

Measures.

Intellectual humility. Intellectual Humility was assessed with the final version of the IHS (see Table 1).

TABLE 1 Items, Means, Standard Deviations, and Factor
Loading of the IHS (Study 1)

Item                                     M    SD    IA    IO

Often becomes angry when their ideas  2.81  1.29   .85   .07
are not implemented.

Values winning an argument over       2.52  1.29   .84  -.01
maintaining a relationship.

Always has to have the last word in   2.76  1.34   .79   .02
an argument.

Gets defensive if others do not       2.86  1.29   .78  -.07
agree with them.

Becomes angry when their advice is    2.86  1.28   .75  -.03
not taken,

Has little patience for others'       2.69  1.28   .73  -.10
beliefs.

Acts like a know-it-all.              2.44  1.29   .70  -.08

Often points out others' mistakes.    3.04  1.26   .67  -.06

Makes fun of people with different    2.48  1.30   .63  -.15
viewpoints.

Seeks out alternative viewpoints.     2.94  1.12   .13   .89

Encourages others to share their      3.23  1.15  -.04   .71
viewpoints.

Enjoys diverse perspectives.          3.02  1.16  -.05   .82

Is open to competing ideas.           3.18  1.13  -.06   .81

Is good at mediating controversial    2.89  1.17  -.07   .73
topics.

Is good at considering the            3.07  1.14  -.09   .77
limitations of their perspective.

Is open to others' ideas.             3.22  1.17  -.18   .72

Note. Bold = primary factor loading; IO = Intellectual
Openness; IA = Intellectual Arrogance.

Trust. The 8-item Dyadic Trust Scale (DTS; Lar-zelere & Huston, 1980) was used to assess trust in relationships. Items (e.g. "He/she is primarily interested in his (her) own welfare.") were completed using a 5-point Likert-type scale (1 = strongly disagree, 5 = strongly agree). The DTS had a Cronbach's alpha of .87 in a previous study using undergraduates (Conley, Moors, Ziegler, & Feltner, 2011). The scores on the scale have shown evidence of construct validity, correlating with measures of love and depth of self-disclosure (Larzelere & Huston, 1980).

Personality. The 44-item Big Five Inventory (BFI; John, Donahue, & Kentle, 1991) was used to assess personality judgments of the target person's openness, agreeableness, neuroticism, conscientiousness, and extraversion. Items (e.g., "Is inventive") were completed on a 5-point Likert-type scale (1 = strongly disagree; 5 = strongly agree). The BFI had Cronbach's alphas for the subscales ranging from .75-.80 (John et al., 1991). The scores on the scale have shown evidence of construct validity, being correlated with other measures of personality such as the NEO (Costa & McCrae, 1992).

Procedure. Participants were recruited from an undergraduate research pool and received partial course credit for completing the study; they indicated consent and completed all measures online. Participants were randomly assigned to one of six conditions, in a 2 (most or least virtuous) x 3 (IH, modesty, drive) design. Participants were prompted by the following question:

Who is the most [least] intellectually humble [modest; driven] person that you have known personally (please do not list this person by name or give us enough information to specifically identify that person. You can just describe your relationship with the person. For example, parent, close friend, family member, etc.)?

We intentionally refrained from defining these constructs for participants because we are especially interested in how people understand and perceive humility within relationships, and were interested in whether lay perceptions of these three constructs would show predictable differences on our new measure. After selecting someone, they rated the target person's level of IH as well as other measures of relationship dynamics and personality.

Results and Discussion

Alphas, means, standard deviations, and intercor-relations are reported in Table 2. Since correlations between the IHS subscales, trust, and the BFI were only negligibly different, we proceeded with further analyses using the full scale IHS (see Table 2). Our first hypothesis was that the IHS would do a better job distinguishing between most and least intellectually humble conditions than it would between most and least modest or driven conditions. To test this hypothesis, we conducted a 2 (most or least virtuous) x 3 (1H, modesty, drive) between groups ANOVA. The predicted interaction was found, F(6, 115) = 12.29, p < .001. To interpret the interaction, we conducted follow-up t-tests, calculated Cohen's d, and the 95% confidence intervals of the effect sizes. Calculating the confidence intervals of the effect sizes helped us determine if the IHS was able to meaningfully distinguish between the constructs. For all three virtues (i.e., IH, modesty, drive), participants in the most or least virtuous groups differed from each other on IHS scores, but the magnitude of that difference was greater for the IH condition (442] = 6.14, p < .001; d= 2.16) than for the modesty (427] = 2.66,p = .014; d = 1.06) or drive (453] = 2.70, p = .010; d = .74) conditions. The confidence interval for the IHS scores in the modesty condition (20.74 + 16.07) overlapped with both the confidence intervals for the IHS scores in the IH (28.57 + 8.25) and drive (9.43 + 7.03) conditions. However, the confidence intervals for the IHS scores in the IH and drive conditions did not overlap with each other, suggesting that the IHS was clearly able to distinguish IH from the less closely related construct of drive.

TABLE 2 Descriptive Statistics and Intercorrelations

Construct               M     SD  [alpha]  r(IHS-FS)    r(lA)    r(IO)

Study 3

Intellectual        51.92  17.43      .96
Humility

Relationship with
exemplar

Trust               28.09  10.08      .95     .74***   .70***   .68***

Personality
judgment

Agrccablencss       30.57   9.66      .93     .78***   .73***   .70***

Openness            31.62   6.58      .80     .54***   .45***   .57***

Conscientiousness   31.72   8.32      .90     .58***   .53***   .55***

Extravcrsion        27.95   5.58      .75      -.070    -.097    -.022

Neuroticism         22.41   7.15      .87    -.58***  -.54***  -.54***

Study 4

Intellectual        46.59  15.96      .94
Humility

Spiritual
appraisals

Viewing the         19.04   9.96      .96       -.07     -.03     -.11
offense as a
desecration

Anger towards God   22.43  18.52      .84     -.35**   -.35**   -.29**

Positive Attitudes  19.24   7.20      .99      .31**   .37***     .23*
towards God

Repair of social
bonds

Benevolence         18.76   5.00      .80     .41***   .45***   .35***
motives

Avoidance           21.22   7.57      .88    -.54***  -.51***  -.51***

Revenge              8.94   4.09      .85        -20     -.14    -.22*

Note. IHS-FS = Full Scale; IA = Intellectual Arrogance;
IO = Intellectual Openness.

*p<.05, **p<.01, ***p< .001.

Our second hypothesis was that IH would be positively related to trust, as well as openness and agreeableness. This hypothesis was supported. Intellectual Humility was strongly and positively related to trust (r = .74,p < .001). As predicted, IH was also strongly and positively related to perceptions of agreeableness (r = .78,p <.001) and openness (r= .54,p <.001). Finally, IH was also related to conscientiousness (r = .58, p < .001) and neuroticism (r =--.58,p < .001), but was not significantly related to extraversion (r=--.07,p= .311).

Thus, the results of Study 3 provided preliminary evidence of construct validity for the IHS scores. Namely, we created six known groups by randomly assigning people to think of someone most or least virtuous on three different constructs (i.e., IH, modesty, drive), and found the expected interaction that suggested the IHS did a better job ofdistinguishing groups for IH than for drive. We also found preliminary evidence to support our hypothesis that 1H is related to the strength of social bonds, with IH being positively related to trust. Also, consistent with our theorizing, perceptions of IH were strongly related to perceptions of higher agreeableness and openness.

Study 4: Intellectual Humility and Betrayal by a Religious Leader

We previously described several factors that make IH especially difficult to practice. Based on that logic, the present study focused on participants who had experienced a major betrayal by a religious leader--a context that involves a "perfect storm" in terms of straining intellectual humility in leaders. Namely, when victims experience betrayal by a religious leader, core aspects of their religious identity may be challenged, which may also evoke negative moral emotions (e.g., contempt) associated with viewing the offense as a desecration (Mahoney et al., 2005). We did not systematically manipulate these various factors theorized to strain IH, but we did strategically recruit our sample based on this logic.

We predicted that victims' spiritual appraisals of the context of the betrayal would be related to perceptions of IH. Davis, Hook, and Worthington (2008) described several ways a victim may interpret an offense in a spiritual way. In the present study, we focused on how such victims may experience (a) the destruction of something sacred (i.e., desecration; Mahoney et at, 2005) and (b) attitudes towards the sacred (Exline et al., 2011). Negative spiritual appraisals tend to intensify people's negative reactions to an offense (Mahoney et at, 2005); thus, we expected such experiences to be negatively related to perceptions of the leader as IH.

We have also theorized that IH regulates social bonds. An implication is that IH should help promote the repair of relationships after a betrayal. Alternatively, when people ruminate angrily and experience less forgiveness towards the leader, this should also erode their view of the leader's character, including IH. For example, victims and offenders view transgressions in predictably different ways (Baumeister, Stillwell, & Wotman, 1990). Conflict often involves negotiations (i.e., exchanging ideas and reasons) over a shared story of the offense. The more entrenched offenders are in justifying their view of the offense, the more victims are likely to perceive them as intellectually arrogant. Thus, we hypothesized that IH would be negatively related to unforgiveness and positively related to intentions to repair the relationship.

Method

Participants. Participants were 105 undergraduates (69 women) from a large urban university in the Southeastern region of the United States, with a mean age of 25.56 years (SD = 6.68). The sample was diverse in terms of ethnicity, with 42.9% self-identifying as Black/African American; 25.7%, White; 15.2% Asian/ Pacific Islander; 7.6% Multiracial; 5.7% Latino/a; 1% American Indian; and 1.9%, no response. In terms of religious affiliation, the majority were Christian (50.96%); 3.8% were Muslim, 3.8% Hindu, 2.9% Buddhist, 1.9% Jewish, 1% New Age, 25% other, and 10.6%, no religious affiliation or did not respond. Finally, most participants identified as heterosexual (87.6%); 3.9% were gay or lesbian, 2.9, bisexual, and 5.7% did not respond.

Measures. Participants rated the IH of the religious leader with the IHS. They also completed measures of how they perceived the offense in a spiritual way, including desecration and anger towards the sacred, as well as measures related to forgiveness.

Viewing the offense as a desecration. The 23-item Sacred Loss and Desecration Scale (SLDS; Pargament, Magyar, Benore, & Mahoney, 2005) was used to measure desecration. It has two subscales: Sacred Loss (e.g., "This event involved losing a gift from God.") and Desecration (e.g., "A sacred part of my life was violated."). Items were rated on a 5-point Likert-type scale (1 = not at all; 5 = very much). The SLDS had Cronbach's alphas for the subscales ranging from .92-.93 (Parga-ment et al., 2005). The scores on the scale have shown evidence of construct validity, being correlated with religious coping and spiritual change (Pargament et al., 2005).

Attitudes towards God. The 9-item Attitudes towards God Scale (ATGS; Wood et al., 2010) was used to assess participants' feelings towards God. It consists of two subscales: Positive Attitudes (e.g. "Feel loved by God) and Anger/Disappointment (e.g, "Feel abandoned by God"). Items were rated on a 10-point Likert-type scale (1 = not at all; 10 = extremely). The ATGS had Cronbach's alphas for the subscales ranging from .64--.98 (Wood et al., 2010). The scores on the scale have shown evidence of construct validity, being related to trait anger and dispositional forgiveness (Wood et al., 2010).

Unforgiveness. The 18-item Transgression-Related Interpersonal Motivations Inventory (TRIM; McCullough et al, 1998) was used to measure unthrgive-ness towards the offender. It consists of three subscales: Avoidance (e.g., "I am avoiding him/her"), Revenge (e.g., "I'll make him/her pay."), and Benevolence Motivations (e.g., "Even though his/her actions hurt me, I have goodwill for him/her."). Items were rated on a 5-point Liken type scale (1 = strongly disagree; 5 = strongly agree.) The TRIM had Cronbach's alphas for the subscales ranging from .84--.96 (McCullough et al., 1998). Scores on the scale have shown evidence of construct validity, being correlated with other measures of forgiveness, relationship satisfaction, and commitment (McCullough et al., 1998).

Procedure. A sample of undergraduate participants completed the study in exchange for partial course credit. Participants indicated consent and completed all measures online through a Qi ... laltrics survey. To confirm eligibility for the study, participants answered the following question: "To be in this study, you said you have experienced an offense by a religious/spiri-tual leader. Please describe that offense and how you learned about it."

Examples of participants' responses include: "I was kicked out of my small group for asking questions;" "The pastor at my girlfriend's church was stealing money;" and "A religious leader was accused of sexually molesting boys." Afterwards they completed the questionnaire.

Results and Discussion

Means, standard deviations, and Cronbach's alphas for the scores are reported in Table 2. As in Study 3, the subscales for the IHS preformed comparably for each of the associations, so only the full scale correlations are discussed here. The hypothesis that IH would be related to viewing the offense in a spiritual way was partially supported. Intellectual Humility was positively related to positive attitudes towards God (r = .31, p = .010) and negatively related to anger towards God (r =--.35,p = .010), but not to viewing the offense as a desecration (r =--.07,p = .475). We further hypothesized that IH would be related to forgiveness-related constructs, and this hypothesis was also supported. Intellectual Humility was negatively related to avoidance on the TRIM (r =--.54,p < .001) and was positively related to benevolence motivations (r = <.001). IH was marginally negatively related to revenge motivations on the TRIM (r =--.20,p = .054).

General Discussion

Theorizing within philosophy has suxested the importance of humility in religious leaders (Roberts & Wood, 2003; 2007). In the current paper, we sought to elaborate a model of relational humility (Davis et al., 2013) to incorporate the construct of IH and then to examine initial evidence for this model in relationships between subordinates and religious leaders. We conceptualized IH as a sub-domain of general humility that involves how people hold and negotiate cherished beliefs. We also theorized that IH is particularly important and evident when strained by convictions related to identity, negative moral emotions (e.g., contempt), and competition and power struggle over ideas. Furthermore, we posited that IH helps regulate the formation and repair of social bonds. To examine initial evidence for these ideas, we focused on betrayals by religious leaders, which involve several factors that can strain IH. Then we examined whether perceptions of 1H were related to signs of relational repair.

Given that there was no published measure of IH, our first three studies focused on developing a measure of perceptions of IH. In Study 1, exploratory factor analysis was used to winnow items and determine the structure of the scale. Results of exploratory factor analyses revealed that the IHS had a two-factor structure. One factor measures one's ability to regulate one's emotions in the face of conflicting viewpoints, while the other factor measures one's general interest and openness to different ideas. Although items on the Intellectual Openness (10) subscale were positively worded and items on the Intellectual Arrogance (IA) subscale were negatively worded, the content of the items aligned with the two components of our definition of IH. For example, items on the TO subscale often explicitly mentioned openness or alternative perspectives (see Table 1, items 10-13 and 16), while items on the IA subscale often mentioned affective responses (see Table 1, items 1 and 4-6). In Study 2, this two factor structure replicated well in an independent sample. The IHS scores showed evidence of reliability, with Cronbach's alphas ranging from .90 to .97.

We also began the process of situating the construct of IH within the literatures on personality, positive psychology, and leadership. Namely, Study 3 examined the construct validity of scores on the IHS, which were shown to adequately distinguish between IH and drive. Furthermore, IH was related to openness and agreeableness, providing additional evidence for construct validity of the scores. In terms of discriminant validity, we might have expected IH to be more strongly related to openness than to agreeableness, but this was not the case. In fact, the general pattern was very similar to other measures of general humility (Davis et al., 2011), so an important next step is to examine evidence that IH is distinct from general humility. Although we generally expect these two constructs to correlate strongly, researchers will need to theoretically examine specific contexts when someone might demonstrate low intellectual humility. This work may require further clarification of other sub-domains of general humility. Interestingly, IH was not related to extraversion, indicating that the IHS is not simply tapping into one's social competence. IH and neuroticism were negatively related, and it would therefore be worthwhile to examine the relationship between IH and stress, perfectionism, depression, and anxiety in future studies.

We also presented preliminary evidence that IH is related to the strength of social bonds. In Study 3, IH was related to higher trust. In Study 4, involving participants who had experienced a major betrayal by a religious leader, IH was also related to lower unfor-giveness and higher conciliatory motivations. These findings are consistent with our theorizing that IH may help repair relationships after conflict, making offenders more willing to negotiate their perception of the offense. However, a plausible alternative explanation is that unforgiveness causes rumination about the offense and changes in how victims view the offender's character, including intellectual humility. They also may motivate the victim to spread negative information in order to harm the offender's reputation. Thus, the offense may have caused declines in how victims perceived the leaders' IH. Longitudinal studies will help clarify the causal links between these constructs.

Finally, we also examined how spiritual appraisals are linked to perceptions of IH. Interestingly, viewing the offense as a desecration was not related to perceived IH. However, perceptions of IH were related to positive and negative attitudes towards God, suggesting that many people's experience of God was associated with their ability to maintain a positive view of their religious leader's moral character. Previous work has examined factors that promote forgiveness of a religious leader after a transgression, although victim rather than offender characteristics have tended to be the focus of study (Sutton & Thomas, 2006). Factors such as gender, dispositional forgiveness, and religious faith have been examined (Sutton, McLeland, Weaks, Cogswell, & Miphouvieng, 2007; Thomas, White, & Sutton, 2008). Interestingly, one study found that apology was associated with lower forgiveness, and it was theorized that an apology may be seen as weak (Thomas et al., 2008). This is notable in light of our findings that perceptions of IH were associated lower unforgiveness. Awareness of one's limitations is often a prerequisite for offering a sincere apology. Our findings are generally consistent with Sutton et al., (2007) which found that that the engagement of spiritual resources when considering a pastor's transgression promoted forgiveness, and that an avoidant attachment style towards God was associated with less willingness to forgive. The current research makes an important step towards examining more closely offender characteristics that may be associated with forgiveness.

Limitations and Future Research

The present studies had several limitations. First, participants provided the only source of ratings, and in every case, were rating another individual. Using a single source of data is subject to biases such as rater bias, person perception heuristic, and assumed similarity between self and other (Lee et al., 2009). Furthermore, each target was only rated by one individual and we were therefore unable to calculate inter-rater reliability. The gold standard for research on personality judgments involves triangulation of self-report, other-report, and behavioral observation (Roberts & Ilardi, 2003). A behavioral observation measure of humility or IH has not been developed. An important next step is to examine the relative validity of various sources of information about IH.

Second, all of the present studies used cross-sectional designs, with the exception of Study 3, which was experimental. Thus, although we found evidence consistent with the idea that IH helps strengthen and repair social bonds, the relationship might be bi-directional (e.g., offenses and unforgiveness cause changes in how the judge views the target person's character). To examine this question, future research will need to examine how perceptions of IH change over time and estimate trait IH prospectively using other sources of information.

Third, items on the 10 subscale were all positively worded, while items on the IA subscale were all negatively worded. An alternative explanation is that these may represent positive and negative valences of IH rather than openness and arrogance. Finally, future research might explore how IH converges with and diverges from general humility, post-formal operational thought (Kramer, 1983), reflective judgment (King & Kitchener, 2004), wisdom (Sternberg, 2004), quest (Batson & Schoenrade, 1991) and spiritual openness (Genia, 1996).

Conclusion

In the case of virtue, psychologists have regularly arrived late (i.e., decades, centuries, or even millennia) to the conversation. In this article, we integrated theorizing on IH with recent theorizing on humility, and we sought to provide initial evidence for the importance of IH in leaders, especially religious leaders. We hope this will be the start of a lively conversation about intellectual humility and its importance for relationships.

References

Batson, C., & Schocnrade, P. A. (1991). Measuring religion as quest: 2) Reliability concerns. Journaifir the Scientific Study ofReligion, 30, 430-447.

Baumeister, R. F., Bauer, I. M., & Lloyd, S. A. (2010). Choice, free will, and religion. Psychology ofReligion and Spirituality, 2,67-82.

Baumeister, It F., Stillwell, A., & Woman, S. R. (1990). Victim and perpetrator accounts of interpersonal conflict: Autobiographical narratives about anger. Journal of Personality and Social Psychology, 59,994-1005.

Brown, S. L., & Brown, R. M. (2006). Selective investment theory: Recasting the functional significance of dose relationships. Psychological Inquiry, 17, 1-29.

Collins, J. (2001). Good to great: Why some companies make the leap ... and others don't. New York, NY: HarperBusiness.

Conley, T. D., Moors, A. C., Ziegler, A., & Feltner, M. R. (2011). Trust and satisfaction in adult child-mother (and other) relationships. Basic & Applied Social Psychology, 33, 239-254.

Costa Jr, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) Professional Manual. Odessa, FL: Psychological Assessment Resources.

Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research, and Evaluation, 10, 1-9.

Davis, D. E., Hook, J. N., & Worthington, E. L., Jr., (2008). Relational spirituality and forgiveness: The roles of attachment to God, religious coping, and viewing the transaction as a desecration. Journal of Psychoiogy and Christianity, 27, 293-301.

Davis, D. E., Hook, J. N., Worthington, E. L., Jr., Van Tongeren, D. R., Gartner, A. I-, Jennings, D. J., & Emmons, R. A. (2011). Relational humility: Conceptualizing and measuring humility as a personality judgment. Journal qfPersonality Assessment, 93,225-234.

Davis, D. E.. Worthington, E. L., Jr., Hook, J. N., Emmons, R. A., Hill, P. C., Bollinger, R. A., & Van Tongeren, D. R. (2013). Humility and the development and repair of social bonds: Two longitudinal studies. Sefand Identity, 12, 58-77.

Emmons, R. A. (2013). Humility: Humility, the modest strength. In J. J. Froh, A. C. Parks (Eds.), Activitieslor teachingpositive psychology: A guide fir instructors (pp. 19-22). Washington, DC, US: American Psychological Association.

Exline, J. J., & Hill, P. C. (2012). Humility: A consistent and robust predictor of generosity. TheJournal of Positive Psychology, 7,208-218.

Exline, J. J., Park, C. L., Smyth, J. M., & Carey, M. P. (2011). Anger toward God: Social-cognitive predictors, prevalence, and links with adjustment to bereavement and cancer. Journal if Personality and Social Psychology, 100,129-148.

Fan, X. (2003). Two approaches for correcting correlation attenuation caused by measurement error: Implications for research practice. Educational er Psychological Measurement, 63(6), 915-930.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.

Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56,218.

Funder, D. C. (1995). On the accuracy of personality judgment: A realistic approach. Psychological Review, 10, 652.

Genia, V. (1996). I, E, quest, and fundamentalism as predictors of psychological and spiritual well-be ing. Journal for the Scientific Study yReligion, 35,56-64.

Goodwin, V. L., Whictingron, J., Murray, B., & Nichols, T. (2011). Moderator or mediator? Examining the role of trust in the transformational leadership paradigm. Journal of Managerial Issues, 23, 409-425.

Hu, L. T., & Bender, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6,1-55.

John, 0. P., Donahue, E. M., & Kende. R. L. (1991). The Big Five Inventoiy--Versions 4a and .517. University of California, Berkley. Institute of Personality and Social Research.

King, P. M., & Kitchener, K. (2004). Reflective judgment: Theory and research on the development of epistemic assumptions through adulthood. Educational Psychologist, 39, 5-18.

Kramer, D. A. (1983). Post-formal operations? A need for further conceptualization. Human Development, 26, 91-105.

Larzelere, R. E., & Huston, T. L. (1980). The Dyadic Trust Scale: Toward understanding interpersonal trust in close relationships. Journal of Marriage and the Family, 42, 595-604.

Lee, K., Ashton, M. C., Pozzebon, J. A., Visscr, B. A., Bourdagc, J. S., & Ogunfowora, B. (2009). Similarity and assumed similarity in personality reports of well-acquainted persons. Journal of Personality and Social Psychology, 96, 460-472.

Mahoney, A., Rye, M. S., & Pargament, M. I. (2005). When the sacred is violated: Desecration as a unique challenge to forgiveness. Handbook ofForgiveness, 57-71.

McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L., Jr., Brown, S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships: IL Theoretical elaboration and measurement. Journal of Personality and Social Psychology, 75, 1586-1603.

McMurray, A. J., Pirola-Merlo, A. A., Sarros, J. C., & Islam, M. M. (2010). Leadership, climate, psychological capital, commitment, and wellbeing in a non-profit organization. Leadership er Organization Development Journal, 31,436-457.

Meets, S., & Strober, J. (2009). Getting to 50/50: How working couples can have it all by sharing it alL New York, NY: Bantam Dell.

Muraven, M., & Baumeister R. F. (2000). Self-regulation and depletion of limited resources: Does self-control resemble a muscle? Pg-chologi cal Bulletin, 126, 247-259.

Muthen, L. K., & Muthen, B. (2008). Mplus user's guide. Los Angeles: Muthen & Muthen.

Nielsen, R., Marrone, J. A., & Slay, H. S. (2010). A new look at humility: Exploring the humility concept and its role in socialized charismatic leadership. Journal of Leadership & Organizational Studies, 17,33-43.

Owens, B. P., & Hekman, D. R. (2012). Modeling how to grow: An inductive examination of humble leader behaviors, contingencies, and outcomes. Academy of Management Journal, 55,787-818.

Pargament, K. I., Magyar, G. M., Benore, E., & Mahoney, A. (2005). Sacrilege: A study of sacred loss and desecration and their implications for health and well-being in a community sample. Journal for the Scientific Study of Religion, 44, 59-78.

Peterson, C., & Seligman, M. E. P. (2004). Character strengths and virtues: A handbook and classification. American Psychologist, 51, 1072-1079.

Reise, S. P., Waller, N. G., & Comrey, A. L. (2000). Factor analysis and scale revision. Psychological Assessment, 12,287-297.

Roberts, M. C., & Ilardi, S. S. (Eds.). (2003). Handbook of research methods in clinical psychology. Malden, MA: Blackwell Publishing Ltd.

Roberts, R. C., & Wood, W. J. (2003). Humility and epistemic goods. In M. DePaul and L. Zagzebski (Eds.), Intellectual virtue: Perspectives from ethics and epistemology (pp. 203-226). Oxford: Clarendon Press.

Roberts, R. C., & Wood, W. J. (2007). Intellectual virtues: An essay in regulative epistemology. Oxford: Oxford University Press.

Stanley, S. M., Rhoades, G. K., & Whitton, S. W. (2010). Commitment: Functions, formation, and the securing of romantic attachment. Journal of Family Theory and Review, 2,243-257.

Steger, M. F. (2006). An illustration of issues in factor extraction and identification of dimensionality in psychological assessment data. Journal ofPersonality Assessment, 86,263-272.

Sternberg, R. J. (2004). What is wisdom and how can we develop it? Annals ty'the American Academy qfPolitical and Social Science, 59, 164-174.

Sutton, G. W., McLeland, K. C., Weaks, K. L., Cogswell, P. E.. & Miphouvieng, R. N. (2007). Does gender matter? Relationship of gender, spousal support, spirituality, and dispositional forgiveness to pastoral restoration. Pastoral Psychology, 55, 645-663.

Sutton, G. W., & Thomas, E. K. (2006). Restoring Christian leaders: How conceptualizations of forgiveness and restoration used in empirical studies can influence practice and research. American Journal of Pastoral Counseling, 8, 27-42.

Thomas, E. K., White, K.. & Sutton, G. W. (2008). Clergy apologies following abuse: What makes a difference? Exploring forgiveness, apology, responsibility-taking, gender, and restoration. Journal of' Psychology and Christianity, 27, 16-29.

Waldron, G. (2012). Spousal interviews of applicants for employment: A legal and ethical dilemma. Journal of Leadership, Accountability, and Ethics, 9, 119-128.

Westen, a, Blagov, P. S., Harenski, K., Kilts, C., & Hamann, S. (2006). Neural bases of motivated reasoning: An fMRI study of emotional constraints on partisan political judgment in the 2004 US presidential election. Journa/ ofCosmitive Neuroscience, 18, 1947-1958.

Wood, B. T., Worthington, E. R.. Exline, J., Yali, A., Aten, J. D., & McMinn, M. R. (2010). Development, refinement, and psychometric properties of the Attitudes toward God Scale (ATGS-9). Psychology ofReligion and Spirituality, 2, 148-167.

Author Information

MCELROY, STACEY E. MS. Address: Department of Counseling and Psychological Services, P.O. Box 3980, Atlanta, GA. Title: Doctoral Student. Degrees: MS--Georgia State University; BS--Uni-versity of Georgia; AS--Athens Technical College. Specializations: intellectual humility, positive psychology in couples.

RICE, KENNETH G. PhD. Address: Department of Counseling & Psychological Services Georgia State University 30 Pryor Street P.O. Box 3980 Atlanta, GA 30302-3980. Title:Matheny Endowed Chair and Professor. Degrees: PhD--University of Notre Dame; MA--University of Notre Dame; BS--Universiry of Florida; AS--Daytona Beach Community College. Specializations: personality and mental health.

DAVIS, DON E. PhD. Address: Department of Counseling and Psychological Services, Georgia State University, P.O. Box 3980, Atlanta, GA 30302-3980. Title: Assistant Professor of Counseling and Psychological Services Georgia State University. Degrees: PhD (Counseling Psychology) Virginia Commonwealth University; BA (Psychology) Yale University. Specializations: humility, forgiveness, positive psychology, religion/spirituality.

HOOK, JOSHUA N. Address: University of North Texas, 1155 Union Circle #311280, Demon, TX 76203. Email: joshua.hook@ unt.edu. Title: Assistant Professor of Psychology. Degrees: BS (Psychology) University of Illinois at Urbana-Champaign; MS (Counseling Psychology) Virginia Commonwealth University; PhD (Counseling Psychology) Virginia Commonwealth University. Specializations: humility, forgiveness, religion/spirituality.

HILL, PETER C. PhD. Address: Rosemead School of Psychology, Biola University, 13800 Biola Ave., La Mirada, CA 90639. Title: Professor of Psychology. Degrees: PhD--University of Houston; MA--University of Houston; BA--Nyack College. Specializations: psychology of religion, religious and spiritual measurement, virtue theory, religion, and health.

WORTHINGTON JR., EVERETT L. Ph.D. Address: Department of Psychology, Virginia Commonwealth University, 806 West Franklin Street, PO Box 842018, Richmond, VA 23284-2018. Website: www.EyWorthington-forgiveness.com. Email: eworth@ vcu.edu. Title: Professor of Psychology (Counseling Psychology). Degrees: PhD (Counseling Psychology) University of Missouri-Columbia. Specializations: forgiveness and other virtues, religion and spirituality in counseling and marriage, the Hope-Focused Couple Approach to marriage/couple enrichment.

VAN TONGEREN, DARYL R. Ph.D. Address: Department of Psychology, Hope College, Schaap Science Center, 35 East 12th Street, Holland, MI 49423-3605. Email: vantongeren@hope.cdu. Title: Assistant Professor of Psychology. Degrees: BA (Psychology) Colorado Christian University; MA (Experimental Psychology) University of Colorado, Colorado Springs; PhD (Social Psychology) Virginia Commonwealth University. Specializations: social psychological approaches to meaning, religion, and virtues.

Stacey E. McElroy, Kenneth G. Rice, and Don E. Davis

Georgia State University

Joshua N. Hook

University ofiVorth Texas

Peter C. Hill

Biala University

Everett L. Worthington, Jr.

Virginia Commonwealth University

Daryl R. Van Tongeren

Hope College