See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/232868751

Humility: Review of measurement strategies and conceptualization as

personality judgment

Article in The Journal of Positive Psychology · July 2010

DOI: 10.1080/17439761003791672

CITATIONS READS

263 7,051

3 authors:

Don Davis Everett L. Worthington

Georgia State University Virginia Commonwealth University
286 PUBLICATIONS 8,675 CITATIONS 463 PUBLICATIONS 22,919 CITATIONS

SEE PROFILE SEE PROFILE

Joshua Hook
University of North Texas
297 PUBLICATIONS 10,106 CITATIONS

SEE PROFILE

All content following this page was uploaded by Don Davis on 28 July 2015.

The user has requested enhancement of the downloaded file.

7KLVDUWLFOHZDVGRZQORDGHGE\>9LUJLQLD&RPPRQZHDOWK8QLYHUVLW\@
2Q-XQH
$FFHVVGHWDLOV$FFHVV'HWDLOV>VXEVFULSWLRQQXPEHU@
3XEOLVKHU5RXWOHGJH
,QIRUPD/WG5HJLVWHUHGLQ(QJODQGDQG:DOHV5HJLVWHUHG1XPEHU5HJLVWHUHGRIILFH0RUWLPHU+RXVH
0RUWLPHU6WUHHW/RQGRQ:7-+8.

-RXUQDORI3HUVRQDOLW\$VVHVVPHQW
3XEOLFDWLRQGHWDLOVLQFOXGLQJLQVWUXFWLRQVIRUDXWKRUVDQGVXEVFULSWLRQLQIRUPDWLRQ
KWWSZZZLQIRUPDZRUOGFRPVPSSWLWOHaFRQWHQW W

5HODWLRQDO+XPLOLW\&RQFHSWXDOL]LQJDQG0HDVXULQJ+XPLOLW\DVD
3HUVRQDOLW\-XGJPHQW
'RQ('DYLVD-RVKXD1+RRNE(YHUHWW/:RUWKLQJWRQ-UD'DU\O59DQ7RQJHUHQD$XEUH\/
*DUWQHUD'DYLG--HQQLQJV,,D5REHUW$(PPRQVF
D
'HSDUWPHQWRI3V\FKRORJ\9LUJLQLD&RPPRQZHDOWK8QLYHUVLW\E'HSDUWPHQWRI3V\FKRORJ\
8QLYHUVLW\RI1RUWK7H[DVF'HSDUWPHQWRI3V\FKRORJ\8QLYHUVLW\RI&DOLIRUQLD'DYLV

2QOLQHSXEOLFDWLRQGDWH$SULO

7RFLWHWKLV$UWLFOH'DYLV'RQ(+RRN-RVKXD1:RUWKLQJWRQ-U(YHUHWW/9DQ7RQJHUHQ'DU\O5*DUWQHU
$XEUH\/-HQQLQJV,,'DYLG-DQG(PPRQV5REHUW$   5HODWLRQDO+XPLOLW\&RQFHSWXDOL]LQJDQG0HDVXULQJ
+XPLOLW\DVD3HUVRQDOLW\-XGJPHQW -RXUQDORI3HUVRQDOLW\$VVHVVPHQW٢
7ROLQNWRWKLV$UWLFOH'2,
85/KWWSG[GRLRUJ

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
 Journal of Personality Assessment, 93(3), 225–234, 2011
Copyright !C Taylor & Francis Group, LLC
ISSN: 0022-3891 print / 1532-7752 online
DOI: 10.1080/00223891.2011.558871

Relational Humility: Conceptualizing and Measuring Humility as

a Personality Judgment
DON E. DAVIS,1 JOSHUA N. HOOK,2 EVERETT L. WORTHINGTON JR.,1 DARYL R. VAN TONGEREN,1 AUBREY L. GARTNER,1
DAVID J. JENNINGS II,1 AND ROBERT A. EMMONS3
1
Department of Psychology, Virginia Commonwealth University
2
Department of Psychology, University of North Texas
3
Department of Psychology, University of California, Davis

The study of humility has progressed slowly due to measurement problems. We describe a model of relational humility that conceptualizes
humility as a personality judgment. In this set of 5 studies, we developed the 16-item Relational Humility Scale (RHS) and offered initial evidence
for the theoretical model. In Study 1 (N = 300), we developed the RHS and its subscales—Global Humility, Superiority, and Accurate View of
Self. In Study 2, we confirmed the factor structure of the scale in an independent sample (N = 196). In Study 3, we provided initial evidence
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

supporting construct validity using an experimental design (N = 200). In Study 4 (N = 150), we provided additional evidence of construct validity
by examining the relationships between humility and empathy, forgiveness, and other virtues. In Study 5 (N = 163), we adduced evidence of
discriminant and incremental validity of the RHS compared with the Honesty-Humility subscale of the HEXACO–PI (Lee & Ashton, 2004).

With the rise of positive psychology, the study of virtue is of belief, behavior, and motivation (Davis, Worthington, et al.,
thriving (Lopez & Snyder, 2009). Literatures on virtues such 2010; Tangney, 2009). Such multifaceted definitions have made
as forgiveness (Worthington, 2005), gratitude (Emmons & Mc- humility difficult to measure.
Cullough, 2004), hope (Snyder, Rand, & Sigmon, 2002), and Given that humility has been defined in such aspirational
optimism (Carver & Scheier, 1994) have expanded rapidly. In- terms, humility researchers have doubted the validity of self-
deed, even the study of modesty is flourishing (Sedikides, Gregg, reports of humility. That is, people who claim to be very humble
& Hart, 2007). In stark contrast, the study of humility has been would seem to be bragging about their humility, something truly
stagnant, mainly due to measurement problems (Tangney, 2009). humble people would not do. People tend to self-enhance on val-
One reason that modesty has been studied more rigorously ued traits such as humility (John & Robins, 1993). Researchers
than humility is that definitions of modesty have led to generally have hypothesized that to the degree that people are truly hum-
accepted measurement strategies. Namely, two types of mod- ble, they will more modestly report their own humility on self-
esty have been defined (Gregg, Hart, Sedikedes, & Kumashiro, report measures (Davis, Worthington, et al., 2010). Therefore,
2008). Intrapersonal modesty refers to having an accurate view truly humble people might modestly underreport their humility,
of self—not too high or low. This has been measured as self- moderately humble people might overestimate their humility
enhancement, which is the degree to which a person has an some, and people low in humility (e.g., narcissists) might over-
overly positive view of self. Interpersonal modesty refers to the estimate their humility a great deal. Although this modesty
tendency to moderate praise or recognition in socially acceptable effect for self-reports of humility has not been tested empiri-
ways, particularly in public settings. This has been measured by cally, humility researchers have assumed that it undermines the
observing how participants act in experimental studies. For ex- validity of self-reports. This problem led Tangney (2005) to con-
ample, Tice, Butler, Muraven, and Stillwell (1995) found that clude that humility might be “one construct that is simply not
people tend to portray themselves more modestly with friends amenable to self-reports” (p. 415). Given the problems with self-
than strangers. These two modesty constructs might not neces- reports, researchers have sought other strategies of measuring
sarily overlap. humility.
In contrast, humility has been more difficult to define. Def-
initions of humility include, but are not limited to, both types FOUR APPROACHES TO MEASURING HUMILITY
of modesty. Humility also includes other interpersonal qual- Davis, Worthington, et al. (2010) reviewed and critiqued four
ities, such as respect and empathy during conflict, openness approaches to measuring humility: self-reports, social compar-
toward different cultures or worldviews, and acceptance of self isons, implicit measures, and informant ratings. We briefly de-
as subordinate to God or the transcendent (for a review of def- scribe each approach and its limitations.
initions, see Davis, Worthington, & Hook, 2010). Furthermore,
humility involves thinking and acting in these ways for the right
reasons (i.e., other-oriented motivations). It involves integrity Self-Reports
Despite the strong warnings of humility researchers, self-
report measures of humility have been created. Arguably,
Received April 26, 2010; Revised September 8, 2010.
Address correspondence to Don E. Davis, Department of Psychology, Vir-
the strongest published measure of humility is the Honesty-
ginia Commonwealth University, 806 West Franklin Street, Richmond, VA Humility (HH) subscale of the HEXACO–PI (Lee & Ashton,
23284–2018; Email: [email protected] 2004). The HH was derived through factor analysis. Whereas
225
 226 DAVIS ET AL.

prior lexical studies of personality inventories have generally created scale that contrasted humility and arrogance words), but
found five personality factors (Costa & McCrae, 1992), Lee and not with established self-report or other-report measures of hu-
Ashton extracted this sixth factor, which is most closely aligned mility or modesty (Powers et al., 2007; Rowatt et al., 2006).
with Agreeableness from the traditional five-factor model. The Given that the IAT–HA was designed to address problems with
HH has four subscales: Fairness, Sincerity, Greed-Avoidance, self-reports, it is especially important to provide evidence of
and Modesty. With the addition of the HH, the HEXACO–PI criterion validity, showing that the measure can predict humble
has demonstrated evidence of incremental predictive validity for behaviors (e.g., doing charitable work without apparent recog-
constructs related to low sociability, such as psychopathy, lack nition or other ulterior motives) or whether one is viewed as
of sexual restraint, Machiavellianism, boredom proneness, and humble by others.
social adroitness (for a review, see Ashton & Lee, 2007). The
HH subscale has made a substantial contribution to the person- Informant Ratings
ality literature. However, a limitation of the scale is that it only
aligns with one aspect of humility—namely, modesty as a lack Finally, researchers have used informant ratings to measure
of superiority (e.g., I am an ordinary person who is no better humility. Namely, items from self-report measures of humility
than others; I wouldn’t want people to treat me as though I were were adapted for use as other-reports. The approach has gen-
superior to them). The other subscales of the HH (Sincerity, erally been used as evidence of criterion-related validity for
Fairness, and Greed-Avoidance) do not align as well with how another measurement strategy (e.g., self-report). Some studies
humility has been defined. have failed to find a relationship between informant ratings
and the humility measure being studied (e.g., self-report, Park
& Peterson, 2006; IAT–HA, Rowatt et al., 2006). Other stud-
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

Social Comparisons ies have found moderate agreement. For example, de Vries,
Rowatt and his colleagues adapted a social-comparison Lee, and Ashton (2008) reported that self–other agreement on
method from the self-enhancement literature to assess humil- the HEXACO–HH was as high as .60 by romantic partners,
ity. They defined humility as having an accurate view of self but ratings by casual acquaintances, coworkers, or friends were
(Rowatt, Ottenbreit, Nesselroade, & Cunningham, 2002). Par- substantially lower (r = .22, .28, and .30, respectively).
ticipants compared themselves to others (e.g., the average col- One advantage of using other-reports to study humility is that
lege student) on how well they adhered to 12 biblical command- it avoids the paradox of having people self-report their own
ments. One limitation of this method is that it cannot discern humility. Whereas claiming to be humble might be immodest,
whether scores reflect attitudes of superiority or accurate per- describing someone else as humble is not. Thus, a straightfor-
ceptions of true differences. For example, Rowatt et al. (2002) ward measure of humility judgments, completed by informants,
found that people with high intrinsic religiosity viewed them- might be used to assess humility, at least within the context of a
selves as better at adhering to biblical commandments than oth- specific relationship.
ers. Perhaps people who were intrinsically religious (i.e., reli- Still, the approach has several limitations. First, as we dis-
gious for its own sake rather than for extrinsic reasons, such as cussed earlier, existing measures of humility do not align fully
social contact or prestige) actually were more likely to adhere with how humility has been defined. Second, using informants
to religious teachings than those who were not as intrinsically can be a time-intensive and expensive method of research. The
religious. Thus, an individual’s actual trait was not estimated Internet, however, has greatly increased the feasibility of such
or accounted for, which is a confound that threatens the inter- research (Vazire, 2006). Third, until recently, researchers had
nal validity of this approach (for a discussion, see Kwan, John, not theoretically elaborated on what it might mean to use other-
Kenny, Bond, & Robins, 2004; Kwan, John, Robins, & Kuang, reports as a primary method of studying humility.
2008).
A MODEL OF RELATIONAL HUMILITY
Implicit Measures To address the need for theoretical elaboration, Davis, Wor-
Rowatt and other colleagues also developed the Implicit As- thington, et al. (2010) proposed a model of relational humility.
sociations Test of Humility versus Arrogance (IAT–HA; Powers, The model aligns the study of humility with a large literature on
Nam, Rowatt, & Hill, 2007; Rowatt et al., 2006). Implicit mea- personality judgements (for reviews, see Funder, 1995; Kenny,
sures have been used to study other constructs that are prone to 2004; Kwan et al., 2004; McCrae & Weiss, 2007).
impression management or socially desirable responding (e.g., Relational humility was defined as an observer’s judgment
attitudes of racial prejudice or self-esteem; Greenwald & Ba- that a target person (a) is interpersonally other-oriented rather
naji, 1995). For the IAT–HA, a computer program flashes words than self-focused, marked by a lack of superiority; and (b) has
with either a humble or arrogant connotation on the monitor. Par- an accurate view of self—not too inflated or too low. Drawing
ticipants pair each word with self or other by pressing a corre- on recent theorizing on moral emotions (Tangney, Stuewig, &
sponding key (e.g., S for self, O for other) as quickly as possible. Mashek, 2007), we suggested that people might judge humil-
Participants’ reaction times are used to infer their level of hu- ity by focusing on the target person’s ability to cultivate pos-
mility. The logic behind the IAT is that people should perform itive, other-oriented emotions, as well as regulate self-focused
cognitively similar pairings more quickly than dissimilar pair- emotions in modest and socially acceptable ways. Furthermore,
ings. For example, a humble person should pair humble words an observer might infer someone’s view of self from his or
with self more quickly than arrogant words. The approach is her behavior, such as facial expressions, language, or abil-
a promising innovation in the measurement of humility. How- ity to learn from mistakes. For example, people who consis-
ever, the IAT–HA needs additional evidence of construct and tently overcommit and thus fail to meet their obligations might
criterion-related validity. It was found to correlate with self- be viewed as having an inflated view of self. Accordingly,
report measures of low humility (e.g., narcissism or an author- relational humility is measured by having an informant, who is
 RELATIONAL HUMILITY 227

in a relationship with a person, rate that person’s humility. This (CFA). In Study 3, we conducted an experiment. Undergraduate
subjective judgment—including distortions caused by relation- participants nominated the most or least humble person they
ship factors, such as characteristics of the judge, the target, or the have known and then rated that person’s degree of humility. In
information available to the judge (Funder, 1995)—is the con- Study 4, we examined evidence for the construct- and criterion-
struct of interest. related validity of the RHS. We also examined evidence for our
Davis, Worthington, et al. (2010) also proposed that judg- model of relational humility. Namely, we explored how judg-
ments of humility generally lead people to deepen their rela- ments of humility affect relationships undergoing conflict (i.e.,
tionship with a target person. According to a theory of emotion one of three situations in which humility is most readily observ-
by Baumeister, Vohs, DeWall, and Zhang (2007), conscious able). In Study 5, we examined evidence of discriminant and
emotions do not cause behavior; rather, they cause a person incremental validity, comparing the RHS to the HH subscale of
to think about what just happened. Specifically, negative emo- the HEXACO–PI (Lee & Ashton, 2004). As in Study 4, Study 5
tions cause people to examine details to prevent an undesired examined how humility judgments were related to forgiveness
event, whereas positive emotions provide a global stamp of ap- of an offense. We used both the RHS and the HH as measures
proval, allowing a person to turn his or her attention toward the of relational humility.
future (Fredrickson, 1998). When a judge perceives positive,
other-oriented emotions (e.g., empathy, sympathy, or gratitude) STUDY 1
being expressed, the judge might interpret this as a sign of The purposes of Study 1 were to (a) use EFA to determine
other-oriented respect and value. Accordingly, seeing the tar- the factor structure of the RHS–71; (b) winnow items to create a
get person as other-oriented should evoke positive emotions brief, face-valid measure of humility judgments; and (c) provide
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

that cause the judge to view the target person as more humble. initial evidence of estimated internal consistency of the scale and
In contrast, when a judge perceives self-focused emotions (e.g., subscales.
shame, pride, or contempt) being expressed in socially offensive
ways, this could signal selfishness and disrespect, which should Method
evoke anxiety in the judge (Tangney et al., 2007), causing the
judge to view the target person as less humble. Participants. Participants were 300 undergraduate students
To accurately judge humility, a person must actually ob- (166 female) from a large urban university. Ages ranged from
serve behaviors that are relevant to humility (Funder, 1995). 18 to 34 (M = 19.1, SD = 2.49). The sample was ethnically
We posit that humility is best observed in three situations that diverse (57.3% White/Caucasian, 15.3% Black/African Amer-
make being other-oriented difficult: (a) honor or recognition, ican, 18.0% Asian/Asian American, 4.0% Latino/Latina, and
(b) hierarchical roles, and (c) conflict. Each of these situations 4.7% other or did not report).
tends to threaten the hierarchy of relationships. For example,
when a person receives an honor or recognition, other people Measure: Humility Judgments (RHS–71). We generated
might feel jealous or question whether their relationship with a list of 71 face-valid items—based on various definitions of
the person being honored will change. In hierarchical roles (e.g., humility—that corresponded with our theoretical conceptual-
parent–child, manager–subordinate, or teacher–student), subor- ization of relational humility. Participants rated the humility of a
dinates who want greater status or fear exploitation might rebel parent by indicating their agreement with items using a 5-point
against the leader’s authority. During conflict, people might ex- rating scale ranging from 1 (strongly disagree) to 5 (strongly
perience negative emotions (e.g., disgust or contempt) that lead agree). Example items include “He/she has a humble charac-
them to feel morally superior to others, making empathy and re- ter,” “He/she has a big ego,” and “He/she knows him/herself
spect difficult. Thus, in all three situations, humility counteracts well.”
the naturally tendency of these situations to cause instability in
relationships. Procedure. Participants were recruited from undergraduate
classes and participated in exchange for partial course credit. Af-
PRESENT STUDIES ter giving consent, participants rated the humility of one of their
The purposes of this set of studies were to (a) provide initial parents (75% chose to rate their mother) using the RHS–71
evidence for the model of relational humility, (b) develop a through a secure online medium. After completing question-
face-valid measure of humility judgments that aligned with our naires, participants were debriefed and given the contact infor-
definition of relational humility, and (c) provide initial evidence mation of the researcher should they have any questions.
that both the HH and the Relational Humility Scale (RHS) can
be used to assess judgments of humility. Results and Discussion
We created the 71-item RHS–71 by developing items to as- The correlation matrix for all humility items was analyzed us-
sess a range of qualities that have been theorized to be part of ing an EFA with maximum likelihood estimation (Fabrigar, We-
humility, including modesty, lack of superiority, accurate view gener, MacCallum, & Strahan, 1999) with an oblique (OBLIM)
of self, respect and value for others, as well as a number of items rotation. To determine the number of factors to extract, we con-
that directly assess humility (e.g., He or she is the most humble ducted a Scree test (Cattell, 1966) as well as a parallel analysis
person I know). (see Steger, 2006). The Scree test suggested a three-factor solu-
In Study 1, we administered the RHS–71 to a sample of un- tion, whereas parallel analysis suggested a six-factor solution.
dergraduate students and used exploratory factor analysis (EFA) We decided to only retain the first three factors for several rea-
to examine the structure of the scale and evaluate items. The fi- sons: (a) only the three factors had enough items to constitute a
nal version yielded a three-factor scale with 16 items, called the reliable subscale; (b) items on the fourth factor were related to
RHS. In Study 2, the factor structure of the RHS was replicated moral character (e.g., he or she is a moral person), but not neces-
on an independent sample using confirmatory factor analysis sarily humility; and (c) items on the fifth and sixth factor shared
 228 DAVIS ET AL.

TABLE 1.—Factor loadings for Relational Humility Scale (RHS) in Study 1. three-factor structure of the RHS using a different sample and
(b) provide additional evidence of the estimated internal consis-
M SD GH S AVS
tency of the scale and subscales. Accordingly, we used CFA to
Global Humility test the structure of the RHS.
He/she has a humble character. 2.78 0.98 .90 −.04 .00
He or she is truly a humble person. 2.98 0.99 .81 −.07 −.02 Method
Most people would consider 2.76 1.06 .70 .02 .01
him/her a humble person. Participants. Participants were 196 undergraduate students
His or her close friends would 3.01 0.92 .63 −.03 −.05 (122 female) from a large urban university. Ages ranged from
consider him/her humble. 18 to 42 (M = 19.2, SD = 2.95). The sample was ethnically
Even strangers would consider 2.68 0.99 .62 .04 −.03
him/her humble.
diverse (55.1% White/Caucasian, 16.8% Black/African Amer-
Superiority ican, 17.3% Asian/Asian American, 4.1% Latino/Latina, and
He/she thinks of him/herself too 0.76 0.98 −.14 −.86 −.03 6.7% other or did not report).
highly.
He/she has a big ego. 0.72 0.98 −.18 −.78 −.06 Measure: Humility judgments (RHS). Humility was as-
He/she thinks of him/herself as 0.88 1.01 −.11 −.74 .00
overly important. sessed with the 16-item RHS, developed in Study 1. For this
Certain tasks are beneath him/her. 0.91 1.05 .06 −.67 .03 sample, the Cronbach’s alpha coefficients were .90 for the Full
I feel inferior when I am with 0.98 1.11 .10 −.60 −.02 Scale, .92 for the Global Humility subscale, .82 for the Superi-
him/her. ority subscale, and .79 for the Accurate View of Self subscale.
He/she strikes me as self-righteous. 1.16 1.23 .06 −.57 .05
The intercorrelations among the subscales were r = −.54 for
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

He/she does not like doing menial 0.98 1.09 −.01 −.52 .03
tasks for others. Global Humility and Superiority, r = .40 for Global Humil-
Accurate View of Self ity and Accurate View of Self, and −.34 for Superiority and
He/she knows him/herself well. 2.85 0.94 .05 −.05 .87 Accurate View of Self.
He/she knows his/her strengths. 2.83 0.98 .02 .09 .74
He/she knows his/her weaknesses. 2.58 1.07 .02 −.07 .72
He/she is self-aware. 2.33 1.12 −.02 .00 .57 Procedure. Participants were recruited from undergradu-
ate classes and participated as part of a course requirement or
Eigenvalue 6.31 2.50 1.79 in exchange for partial course credit. After giving consent, par-
Variance accounted 36.99 12.88 9.11
ticipants rated the humility of a parent (77% rated their mother)
Note. N = 300. Values shown in bold loaded on primary factor; GH = Global Humility using the RHS through a secure online medium. After com-
subscale; S = Superiority subscale; AVS = Accurate View of Self subscale. pleting questionnaires, participants were debriefed and given
the contact information of the researcher should they have any
questions.
a common word (e.g., admire), but they did not align clearly
with a quality of humility, as defined in previous literature.
Results and Discussion
After examining the content of items, the factors were named
Global Humility (e.g., He/she has a humble character), Supe- The covariance matrix was analyzed with maximum likeli-
riority (e.g., He/she has a big ego), and Accurate View of Self hood estimation using Mplus 8.1 (Muthén & Muthén, 2010).
(e.g., He/she knows him/herself well). Items were dropped that Items of the RHS were used as indicators of the Global Hu-
did not load at least .50 on their primary factor, or that loaded mility, Superiority, and Accurate View of Self factors, which
over .25 on any secondary factor. In addition, two items were were modeled as correlated factors. Several fit indexes were ex-
dropped from the Global Humility factor because they were amined to evaluate the overall fit of the model—the chi-square
highly redundant with other items. value, the comparative fit index (CFI), the square root mean
The final version of the RHS consisted of 16 items, with three residual (SRMR), and the root mean square error of approxi-
factors that are theoretically consistent with our definition of mation (RMSEA). As a rule of thumb, a CFI around .95, an
relational humility (Global Humility—5 items; Superiority—7 SRMR equal to or less than .08, and an RMSEA equal to or
items; Accurate View of Self—4 items). Descriptive statistics less than .06 suggest good fit (Hu & Bentler, 1999). We found
and factor loadings for RHS are listed in Table 1. The three that the three-factor model showed good fit, χ 2(101, N = 196)
factors accounted for 63.9% of the variance in items. The Cron- = 168.21, p <.001, CFI = .96, SRMR = .05, RMSEA = .06.
bach’s alphas for the full scale and subscales were .89 for the We also compared the three-factor model to a one-factor model
full scale, .92 for Global Humility, .87 for Superiority, and .82 using a chi-square difference test. Fit worsened substantially for
for Accurate View of Self. The intercorrelations among the sub- the one-factor model, "χ 2(3) = 574.59, p <.001. Therefore,
scales were r = −.45 for Global Humility and Superiority; r = the three-factor solution was retained. Thus, the results of Study
.40 for Global Humility and Accurate View of Self; and r = 2 provide additional evidence for the three-factor structure and
−.26 for Superiority and Accurate View of Self. These findings internal consistency of the RHS.
provide initial evidence that the RHS has a three-factor struc-
ture. Because items were winnowed based on the characteristics STUDY 3
of one sample, we sought to confirm the factor structure in a In Study 3, we manipulated humility experimentally. In a
different sample in Study 2. between-subject design, we randomly assigned participants to
nominate the most or least humble person they have known. We
STUDY 2 hypothesized that the RHS subscales would detect differences
Study 1 yielded a three-factor solution for a 16-item scale between the two groups, even after controlling for characteristics
of humility. The purposes of Study 2 were to (a) replicate the of the relationship between the participant and the target person.
 RELATIONAL HUMILITY 229

Method Results and Discussion

Participants. Participants were 200 undergraduate students To test whether conditions would differ on the RHS sub-
(112 female) from a large urban university. Ages ranged from scales, four univariate analyses of variance were conducted that
18 to 39 (M = 19.2, SD = 2.9). The sample was ethnically compared RHS total scores and subscale scores between the two
diverse (55.0% White/Caucasian, 17.0% Black/African Amer- experimental conditions. To adjust for Type I error, a Bonferroni
ican, 17.5% Asian/Asian American, 4.0% Latino/Latina, and correction was used. As predicted, Humility Judgments (RHS
6.0% other or did not report). Total) were higher in the most humble condition (M = 68.96,
SD = 6.44) than in the least humble condition (M = 38.78,
Measures. SD = 12.02), F(1, 194) = 470.87, p < .001, d = 3.13. For the
subscales, participants in the most humble condition (relative
Humility Judgments: Humility was assessed using the to the least humble condition) reported higher scores on Global
RHS. Cronbach’s alphas for the Full Scale and for the Global Humility (M = 22.17, SD = 2.59 relative to M = 9.26, SD =
Humility, Superiority, and Accurate View of Self subscales were 4.89, p < .001, d = 3.30); lower scores on Superiority (M =
.95, .97, .90, and .90, respectively. 17.03, SD = 6.84 relative to M = 30.07, SD = 4.11, p < .001,
d = –2.31); and higher scores on Accurate View of Self (M =
Similarity and Closeness: Similarity and closeness to the 16.65, SD = 2.87 relative to M = 12.48, SD = 4.31, p < .001,
target person were assessed using single-item measures (i.e., d = 1.14).
“how similar are you to him or her”; “how close do you feel We also tested whether the RHS detected differences between
to him or her”). Items were completed on a 5-point rating scale groups, after controlling for other relational characteristics (i.e.,
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

from 1 (not at all similar [close]) to 5 (very similar [close]). similarity, closeness, positive and negative emotion, and empa-
thy). A hierarchical multiple regression analysis was conducted,
Positive and Negative Affect: Positive and negative affect with humility judgments as the criterion variable, relational
toward the target person were assessed using the 20-item Posi- characteristics entered in Step 1, and experimental condition
tive and Negative Affect Schedule (PANAS; Watson, Clark, & entered in Step 2. Relational characteristics accounted for 70%
Tellegen, 1988). Items were completed using a 5-point rating of the variance in RHS scores, F(5, 167) = 78.59, p < .001. Con-
scale from 1 (very slightly or not at all) to 5 (extremely). Watson trolling for relational characteristics, condition still accounted
et al. (1988) reported Cronbach’s alphas ranging from .84 to .90 for an additional 10% of the variance in RHS scores, "R2 =
for the two subscales and temporal stability estimates ranging .10, F(1, 166) = 90.54, p < .001.
from .39 to .71. Watson et al. also reported evidence of con- The results of Study 3 provided initial evidence supporting
struct validity. Scores on the scale were found to be related to the construct validity of the RHS. The RHS was sensitive to
measures of general distress and dysfunction, depression, and participants’ nominations of people with high or low humility.
anxiety. For this sample, Cronbach’s alphas for the positive and Even after controlling for other relational differences, the RHS
negative factors were .90 and .94, respectively. detected differences between the two conditions.
State Empathy: Empathy toward the target person was
assessed using the 8-item Batson Empathy Adjectives (BEA; STUDY 4
Coke, Batson, & McDavis, 1978). Participants rated each ad- The purpose of Study 4 was to provide evidence of the con-
jective (e.g., concerned) on a 6-point scale from 0 (not at all) struct validity for the RHS and to test one of the key propositions
to 5 (extremely). Coke et al. (1978) reported Cronbach’s alphas of our model of relational humility. Namely, according to our
ranging from .79 to .95. Scores on the scale show evidence of model of relational humility, conflict is one of the situations
construct validity and were found to be positively correlated in which humility is most observable. Conflict tends to cause
with empathic concern, perspective taking, and helping behav- anger, contempt, disgust, and other negative emotions that gen-
ior. The Cronbach’s alpha for the current sample was .94. erally evoke stubborn and arrogant behaviors. Thus, we wanted
to show that humility judgments would predict positive rela-
Procedure. Participants were recruited from undergraduate tionship characteristics in the context of conflict. We examined
classes and participated in exchange for partial course credit. humility in the context of conflict in two relationships.
After giving consent, participants completed an online survey. First, we examined the participants’ relationship with some-
Participants were alternately assigned to one of two conditions. one who had hurt or offended them (i.e., called “the offender”).
In the most humble condition, participants described their re- Our model of relational humility suggests that humility should
lationship with a particular person they saw as most humble be associated with other-oriented emotions toward the target.
(40% rated a parent, 37.4% a friend, 16.2% a relative other than Perceiving that a target person is humble involves seeing that
a parent, and 6% some other relationship). In the least hum- person as interpersonally other-oriented, lacking superiority, and
ble condition, participants described their relationship with the having an accurate view of self. Furthermore, we theorized that
person they saw as least humble (8.9% rated a parent, 54.5% humility is important in situations that challenge the hierarchy
a friend, 7.9% a relative other than a parent, and 28.7% some of a relationship such as conflict. Thus, we hypothesized that
other relationship). Participants assessed their relationship with humility would be positively correlated with forgiveness and
the target person, including humility, similarity, closeness, and empathy.
emotions toward the person (e.g., positive, negative, empathy). Second, we examined participants’ relationship with a par-
After completing questionnaires, participants were debriefed ent with whom they tend to have the most conflict. Gener-
and given the contact information of the researcher should they ally, experiencing conflict in a relationship causes a judge to
have any questions. develop increasingly negative views about the target person’s
 230 DAVIS ET AL.

character (Baumeister et al., 2007). Thus, we hypothesized that Witvliet, & Kiefer, 2005). Items were rated with a 10-point rat-
humility judgments would be related to current emotions toward ing scale ranging from 1 (not at all) to 10 (very much). The
the parent and to perceptions of other virtues. Specifically, we VOS has two subscales: warmth-based virtues (e.g., empathy or
hypothesized that participants who perceived their parent to be forgiveness) and conscientiousness-based virtues (e.g., justice
high in humility would report higher levels of positive emotion or self-control). The psychometric properties of this scale have
and lower levels of negative emotion toward that parent. Also, been reported in previous research (Berry et al., 2004; Berry et
we hypothesized that participants who rated their parent higher al., 2005). Cronbach’s alpha coefficients have ranged from .71
in humility would also rate their parent higher on other virtues. to .94. Furthermore, the items of each scale have demonstrated
adequate fit to a Rasch model for rating scales, demonstrating
Method that the items are forming adequate unidimensional scales. In
Participants. Participants were 150 undergraduate students this sample, Cronbach’s alphas for Warmth-Based Virtues and
(95 female) from a large urban university. Ages ranged from Conscientious-Based Virtues were .94 and .95, respectively.
18 to 28 (M = 18.9, SD = 1.5). The sample was ethnically
diverse (53.9% White/Caucasian, 17.1% Black/African Amer- Procedure. Participants were recruited from undergraduate
ican, 19.1% Asian/Asian American, 2.6% Latino/Latina, and classes and participated for partial course credit. After giving
7.2% other or did not report). consent, participants first thought about a person who had hurt
or offended them, wrote a summary of the transgression, and
Measures Related to an Offender. completed a set of questionnaires about their relationship with
the offender. Second, participants thought about the parent with
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

Humility Judgments: Participants completed the RHS for whom they experienced the highest level of conflict (62.6%
the offender. Cronbach’s alphas for the full scale and for the chose their mother), wrote a description of this relationship,
Global Humility, Superiority, and Accurate View of Self sub- and completed a second set of questionnaires about their rela-
scales were .89, .94, .85, and .89, respectively. tionship with the target parent. After completing questionnaires,
participants were debriefed and given the contact information
Unforgiveness: Unforgiveness toward the offender was as- of the researcher should they have any questions.
sessed with the 12-item Transgression Related Interpersonal
Motivations (TRIM; McCullough et al., 1998). The TRIM has Results and Discussion
two subscales: Revenge (5 items; e.g., “I’ll make him/her pay”) Means, standard deviations, alphas, and intercorrelations for
and Avoidance (7 Items; e.g., “I keep as much distance between all scales are reported in Table 2. To show evidence of construct
us as possible”). Items were completed on a 5-point rating scale validity, we examined the correlations of the RHS with other
from 1 (strongly disagree) to 5 (strongly agree) Lower scores relationship characteristics.
indicate higher forgiveness (i.e., lower unforgiveness). McCul- First, we hypothesized that perceived humility of an offender
lough et al. (1998) reported Cronbach’s alpha coefficients for would be negatively related to unforgiveness and positively re-
scores on both subscales ranging from .85 to .93 and estimates lated to empathy toward that offender. This hypothesis was sup-
of 3-week temporal stability ranging from .44 to .65. Scores on ported. The total RHS, Global Humility, and Accurate View of
the scale showed evidence of construct validity and were found Self subscale scores were related to less total unforgiveness,
to be negatively correlated with other measures of forgiveness, as well as avoidance and revenge, and were related to greater
relationship satisfaction, and commitment. In this sample, Cron- empathy; likewise, the Superiority subscale was related to more
bach’s alphas were .90 for Revenge and .94 for Avoidance. total unforgiveness, as well as avoidance and revenge, and was
related to less empathy (see Tabl 2).
State Empathy: Empathy toward the target offender was Second, we hypothesized that humility judgments of a par-
assessed with the 8-item BEA (Coke et al., 1978; described in ent would be positively related to perceptions of virtue and
Study 3). In the current sample, the Cronbach’s alpha was .94. current experiences of emotion toward that parent. This hy-
pothesis was supported. Higher levels of perceived humility
Measures Related to a Parent. were related to higher perceptions of both warmth-based and
conscientiousness-based virtues (see Table 2). Higher levels of
Humility Judgments: Participants completed the RHS for
perceived humility were also related to higher positive emo-
the parent with whom they have had the most conflict. Cron-
tion and lower negative emotion toward the target parent (see
bach’s alphas for the full scale and for the Global Humility,
Table 2).
Superiority, and Accurate View of Self subscales were .92, .97,
In Study 4, we provided additional evidence supporting the
.85, and .90, respectively.
construct validity of the RHS. Greater levels of perceived humil-
Positive and Negative Affect in Relationship with Parent: ity were related to (a) greater forgiveness of and empathy toward
Positive and negative affect toward the parent were assessed with an offender, and (b) greater perceptions of warmth-based and
the 20-item PANAS (Watson et al., 1988; described in Study 3). conscientious-based virtues, greater positive emotions, and less
In this sample, Cronbach’s alphas for Positive Emotions and negative emotions toward a target parent.
Negative Emotions were .88 and .92, respectively.
STUDY 5
Virtue Judgments: The degree to which parents were The purposes of Study 5 were to (a) provide additional ev-
viewed as possessing various virtues was assessed with the idence that humility judgments are related to the judge’s rela-
18-Item Virtue Orientation Scale (VOS; Berry, Worthington, tionship with an offender, and (b) to examine evidence of the
O’Connor, Parrott, & Wade, 2004; Berry, Worthington, Wade, discriminant and incremental validity of the RHS compared with
 RELATIONAL HUMILITY 231

TABLE 2.——Intercorrelations of the Relational Humility Scale (RHS) with other constructs (Study 4).

Total Subscales
a M SD RHS GH S AVS

Relationship with offender

∗∗ ∗∗ ∗∗ ∗∗
Unforgiveness .94 29.63 12.63 −.49 −.45 .33 −.37
∗∗ ∗∗ ∗∗ ∗∗
Avoidance .90 19.77 9.01 −.50 −.43 .32 −.40
∗∗ ∗∗ ∗∗ ∗∗
Revenge .94 9.87 5.14 −.35 −.35 .25 −.22
∗∗ ∗∗ ∗∗ ∗∗
Empathy .94 19.60 8.16 .49 .48 −.30 .37
Relationship with parent
∗∗ ∗∗ ∗∗ ∗∗
Warmth virtues .94 55.12 20.31 .63 .59 −.48 .49
∗∗ ∗∗ ∗∗
Conscientiousness virtues .95 49.09 18.48 .60 .56 −.46∗ .50
∗∗ ∗∗ ∗∗ ∗∗
Positive emotions .88 30.64 9.30 .57 .54 −.41 .54
∗∗ ∗∗ ∗∗ ∗
Negative emotions .92 22.07 7.80 −.41 −.25 .44 −.20

Note. N = 142. GH = Global Humility subscale; S = Superiority subscale; AVS = Accurate View of Self.
∗
p < .05; ∗∗ p < .01.

the HH subscale of the HEXACO–PI (Lee & Ashton, 2004). In father. After completing questionnaires, participants were de-
this study, we recruited participants who had been hurt by a briefed and given the contact information of the researcher
father’s dedication to his work. This offense aligns with two of should they have any questions.
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

the three situations in which humility should be observable (i.e.,

conflict in a relationship with an authority figure). We predicted Results and Discussion
that judgments of humility—measured with both the RHS and
Means, standard deviations, alphas, and intercorrelations for
the HH—would affect the extent to which participants had ex-
all scales are reported in Table 3. First, we hypothesized that
perienced forgiveness toward their father.
the RHS subscales would show evidence of discriminant valid-
Method ity from the HH subscales. The Global Humility subscale was
moderately correlated with HH; the Superiority subscale was
Participants. Participants were 163 undergraduate students strongly (and negatively) related; and the Accurate View of Self
(95 female) from a large urban university. Ages ranged from was weakly related (see Table 3).
18 to 25 (M = 20.0, SD = 2.6). The sample was ethnically Second, we hypothesized that the RHS would predict unfor-
diverse (54.0% White/Caucasian, 20.2% Black/African Amer- giving motivations above and beyond the HH subscales. To test
ican, 9.2% Asian/Asian American, 3.1% Latino/Latina, and this hypothesis, we conducted two hierarchical regression anal-
13.5% other or did not report). yses with avoidance or revenge motivations as the dependent
variables. For each analysis, the HH subscales were entered in
Measures. the first step and the RHS subscales were entered into the second
Humility Judgments: Participants completed the RHS to step. The HH subscales predicted 32% of the variance in revenge
describe the humility of the target person. Cronbach’s alphas for scores in the first step, and the RHS predicted an additional 8%
Global Humility, Superiority, and Accurate View of Self were of the variance in revenge scores in the second step, R2" = .08,
.93, .89, and .89, respectively. They also completed the 16-item F(3, 140) = 4.63, p = .004. The HH subscales predicted 43%
HH subscale of the HEXACO–PI–100 (Lee & Ashton, 2004) of the variance in avoidance scores in the first step, and the RHS
for the target. Participants rated their father on each item (e.g., predicted an additional 5% of the variance in avoidance scores
“He/she is an ordinary person who is no better than others”) on in the second step, R2" = .05, F(3, 138) = 3.29, p = .022.
a 5-point scale from 1 (strongly disagree) to 5 (strongly agree) We also analyzed the reverse models, examining whether
as applied to the target. The HH has four facet scales: Sincerity, the HH predicted unforgiving motivations after controlling for
Fairness, Greed-Avoidance, and Modesty. The Cronbach’s alpha the subscales of the RHS. The RHS subscales predicted 41% of
for the full scale was .81; the facet scales ranged from .47 to .74 the variance in revenge scores in the first step, and the HH did not
(see Table 3). Thus, the alpha was adequate for the full scale predict significant additional variance in revenge scores in the
score, but several of the facet scores had low reliability. The HH second step, R2" = .01, F(1, 143) = 2.05, p = .154. However,
subscale has considerable evidence of incremental validity, over the RHS subscales predicted 38% of the variance in avoidance
and above the five-factor model (Ashton & Lee, 2007). scores in the first step, and the HH predicted an additional 7%
of the variance in avoidance scores in the second step, R2" =
Unforgiveness: Unforgiveness was assessed using the 12- .07, F(1, 141) = 11.71, p = .001.
item TRIM (McCullough et al., 1998; see description in Study We conducted ancillary analyses to determine whether a
4). Cronbach’s alphas for the avoidance and revenge subscales brief version of the RHS—the Global Humility subscale
were .95 and .87, respectively. alone—would also explain additional variance in revenge and
avoidance motivations, above and beyond the HH subscales. If
Procedure. Participants were recruited from undergraduate it did so, some researchers might wish to use this brief version of
classes and participated for partial course credit. After giving the scale. Thus, we conducted two hierarchical regression anal-
consent, participants thought of a time when they had been yses with avoidance or revenge motivations as the dependent
hurt by their father’s devotion to work. They then completed variable, the HH entered in a first step, and the Global Humility
measures regarding the offense and their relationship with their subscale entered in a second step. The Global Humility subscale
 232 DAVIS ET AL.

TABLE 3.—Intercorrelations of the Relational Humility Scale (RHS) with the HH and forgiveness (Study 5).

M SD N α RHS GH S AVS 4 5 6 7 8 9

1. RHS Total 59.04 11.73 154 .91 —

∗∗
2. RHS Global Humility 17.48 5.78 160 .93 .83
∗∗ ∗∗
3. RHS Superiority 12.15 8.90 160 .89 −.83 −.49
∗∗ ∗∗ ∗∗
4. RHS Accurate View of Self 14.61 4.48 160 .89 .62 .43 −.21
∗∗ ∗ ∗∗
5. HH Sincerity 12.50 3.28 161 .47 .27 .16 −.27 .10
∗∗ ∗∗ ∗∗ ∗∗ ∗∗
6. HH Fairness 15.23 3.43 163 .61 .41 .31 −.39 .21 .33
∗∗ ∗∗ ∗∗ ∗∗ ∗∗
7. HH Greed Avoidance 11.90 4.07 162 .74 .38 .27 −.37 .10 .42 .24
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗
8. HH Modesty 13.65 3.58 162 .66 .60 .47 −.53 .33 .39 .44 .58
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗
9. HH Full Scale Score 53.33 10.64 159 .81 .56 .42 −.53 .25 .71 .66 .78 .82
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗
10. TRIM Avoidance 12.67 6.98 159 .87 .37 −.31 .21 −.25 −.36 −.27 −.24 −.29 −.40
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗
11. TRIM Revenge 6.49 2.87 161 .95 .40 −.29 .35 −.22 −.21 −.22 −.22 −.27 −.32 .67

Note. RHS = Relational Humility Scale; HH = Honesty-Humility; TRIM = Transgression-related interpersonal motivations; GH = Global Humility subscale; S = Superiority
subscale; AVS = Accurate View of Self.
∗
p < .05; ∗ ∗ p < .01.

predicted an additional 3% of the variance in revenge scores, with other positive relationship characteristics and dynamics,
R2" = .029, F(1, 148) = 4.97, p = .027, and an additional 3.7% such as forgiveness, empathy and other emotions, and percep-
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

of the variance in avoidance scores, R2" = .037, F(1, 146) = tion of virtues.
6.82, p = .01, over and above the HH. Studies 4 and 5 were the first to focus on humility of an
Thus, in Study 5, we provided additional evidence for the offender in the context of forgiveness. Previous theory and re-
model of relational humility, finding that both measures of re- search has focused on humility of the victim, proposing that
lational humility (i.e., the RHS and the HH) were related to the humility can promote empathy and dissipate feelings of dis-
degree to which participants had forgiven an offense related to gust and moral superiority toward the offender (e.g., Sandage,
a father’s devotion to work. We also showed evidence of dis- 1999; Worthington, 1998). For example, self-reports of humility
criminant validity for the RHS, showing that the RHS subscales have been positively associated with trait forgivingness (Rowatt
differed from the HH and its subscales. Furthermore, measur- et al., 2006; Shepherd & Belicki, 2008). Moreover, victims have
ing humility judgments with the RHS incrementally predicted been found to be more likely to forgive when they considered
forgiveness, above and beyond the HH, and we also provided themselves capable of committing a similar offense (Exline,
evidence that a briefer version of the RHS—administering only Baumeister, Zell, Kraft, & Witvliet, 2008; Exline, Zell, Mal-
the Global Humility scale—explained unique variance in unfor- colm, DeCourville, & Belicki, 2008). Whereas these studies
giving motivations. focused on humility in the victim, we focused on the degree to
which the victim judged the offender as possessing humility.
In both Studies 4 and 5, perceiving an offender as humble was
GENERAL DISCUSSION positively related with greater forgiveness.
The main purpose of this research was to describe a model With the articulation of a model of relational humility, as
of relational humility and to provide evidence for that model. well as the development of the RHS, researchers have a choice
Second, we developed a face-valid measure of humility judg- of measures to begin studying relational humility. An advantage
ments, which allows continued progress in measuring humility of the RHS is that its items and subscales were evaluated and de-
through other-reports. Thus, in a series of five studies, we de- veloped using both theory and empirical methods. Its subscales
veloped the 16-item RHS (Study 1). We replicated its factor align with how humility has been defined (Davis, Worthington,
structure (Study 2), which provided construct and content valid- et al., 2010). In addition, the RHS is the only measure that has
ity for the construct of relational humility. We demonstrated the been developed primarily to assess the degree an observer judges
sensitivity of the RHS to rating people nominated as high or low a target person to possess humility. Furthermore, the measure is
in humility, even after controlling for relational characteristics brief. Indeed, the Global Humility subscale (5 items) provides a
(Study 3). We also provided evidence supporting the construct brief snapshot of relational humility (in Study 5, we found that
validity of the RHS scale (Study 4). Finally, we provided evi- the 5-item measure predicted unique variance in avoidance and
dence that the RHS subscales account for unique variance, above revenge motivations, above and beyond the HH—albeit only a
and beyond the 16-item HH subscale of the HEXACO–PI–100 small amount of additional variance was accounted for). The
(Lee & Ashton, 2004; Study 5). In psychological science, defi- other two subscales (11 items) provide a more nuanced assess-
nition, concept, and measurement are intertwined. In these five ment of humility.
studies, we provided evidence for the way of defining, concep-
tualizing, and measuring relational humility, as described by our
theoretical model. Limitations
The RHS was found to have three subscales—Global Humil- This set of studies had several limitations. First, only one
ity, Superiority, and Accurate View of Self—that correspond method of measurement was used (i.e., other-report). In the fu-
with our definition of relational humility. The factor structure ture, it will be important to examine consensus among raters and
of the RHS was replicated in an independent sample, and the agreement with self-report. In addition, researchers should ex-
subscales showed acceptable estimates of internal consistency. amine how humility judgments are related to physiological (e.g.,
Furthermore, the RHS and its subscales showed initial evidence are RHS ratings related to more emotional calmness during, for
of construct validity. The subscales were found to be correlated example, discussions of relational trust?), behavioral (e.g., are
 RELATIONAL HUMILITY 233

RHS ratings related to fewer uses of personal pronouns, fewer Fourth, researchers should examine our claim that humility
defensive verbalizations, and fewer nonverbal defensive behav- is most readily observed in three situations: receiving praise
iors?), or cognitive (e.g., are RHS ratings related to scores on or recognition, hierarchical roles, or conflict. For example, re-
the IAT–HA or some other test of cognition) measures. Second, searchers might experimentally manipulate these factors that
this set of studies used cross-sectional rather than longitudinal make humility more difficult to practice and compare how self-
designs, and thus we were unable to assess how humility judg- reports and other-reports predict differences in behavior across
ments might change over time. For example, humility judgments conditions. Similarly, researchers should begin to study humil-
might be sensitive to conflict in a relationship. Similarly, high ity in samples for which humility is potentially important, such
fluctuations in relational humility judgments might be associ- as business leaders, teachers, elected public officials, parents, or
ated with negative relational qualities, such as low commitment couples.
or satisfaction. Third, only college students were studied (see Fifth, researchers should examine evidence for several nu-
Henrich, Heine, & Norenzayan, 2010, who critique the sole use anced hypotheses implied by our theoretical model. For ex-
of undergraduates). Thus, it is important to evaluate the RHS, ample, our model suggests that humility is subjectively eas-
as well as the model of relational humility, in other samples, es- ier (and less ambiguous) to assess in others than in oneself.
pecially community samples in which participants are involved Modesty norms are likely more salient when describing one’s
in hierarchical roles (e.g., mentorship, supervisor–subordinate, own humility than when judging someone else’s humility. Also,
or teacher–student relationships). defensive processes of maintaining a particular conceptualiza-
tion of oneself could be lessened. Moreover, researchers might
examine whether relationship factors moderate how humility
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

is attributed within relationships (see Funder, 1995, for a re-

Future Research view). For example, convergence between self- and observer-
With the development of the RHS, researchers now have at reports should be higher in relationships between people who
least two strong options for measuring relational humility, in- know each other well than in relationships between strangers
cluding the RHS and the HH. Given that humility research is (Kenny, 2004). Also, people might consider different qualities
in an early phase, we suspect that other researchers will put when judging the humility of close friends versus acquaintances
forth alternative definitions of humility and develop associated (McAdams & Pals, 2006). Furthermore, cultural or spiritual fac-
measures. To provide assurance in measurement, it is neces- tors might affect how people judge humility (Davis, Hook, et
sary to continue to evaluate the construct and criterion-related al., 2010; Emmons & Kneezel, 2005; Sandage, 1999). Finally,
validity of the RHS, as well as other measures. Researchers we hypothesize that humility should be less stable over time
should explore the unique correlates of the three subscales. For than other traits such as extraversion. For example, only a few
example, the Accurate View of Self subscale might uniquely immodest acts could squander a reputation of humility.
predict self-monitoring, whereas the Superiority subscale might
uniquely predict Machiavellianism. Furthermore, it is important CONCLUSION
to explore some of the constructs, behaviors, and physiological For some time, humility researchers have been unsatisfied
markers that are associated with judgments of high humility. We with measures of humility (e.g., Tangney, 2009). Whereas mod-
have theorized that restraining expressions of low humility, as esty researchers have been satisfied with separately studying
measured by the Superiority subscale, might have different cor- different aspects of modesty, such as accurate view of self or
relates than expressing high humility, as measured by the Global modest self-presentation, humility researchers have insisted on
Humility and Accurate View of Self subscales. However, it is defining humility as integrity among thought, action, and mo-
also possible that the factor structure of the RHS reflects whether tivation. Such noble conceptualizations of humility have made
items were positively or negatively keyed. humility difficult to assess. Only a few measures of humility
Now that initial measures of relational humility exist, re- have been published in peer-reviewed journals (e.g., Lee & Ash-
searchers should turn to testing some of the many implications ton, 2004; Rowatt et al., 2006). We have proposed a model of
of our theoretical model. First, one might be able to derive an relational humility that contextualizes humility within actual
estimate of trait humility using consensus among several in- relationships and studies humility from the perspective of a
formants (e.g., Davis, Worthington, et al., 2010; Kenny, 2004). third-party observer. The approach aligns the study of humility
However, if consensus between raters is particularly low, sug- with an expansive body of research on personality judgments
gesting that ratings of humility are relationship-dependent, it and circumvents the paradox of using self-reports of humility.
might not make sense to derive a trait estimate. Furthermore, we developed a scale of humility that aligns with
Second, researchers might explore the self–other agreement our proposed definition and approach for studying humility. We
of humility ratings. For instance, this method could be used to hope the introduction of this model and scale will be an impor-
actually examine the degree to which humble people (as rated tant step in catalyzing research on humility.
by others) modestly report their own humility. Researchers have
yet to provide evidence for this hypothesis, which is assumed to ACKNOWLEDGMENT
undermine the validity of self-reports of humility. Gratitude is expressed to the John Templeton Foundation for
Third, RHS scores could also be treated as a criterion variable funding the current project.
to offer evidence of validity for other measures, such as the
IAT–HA (Rowatt et al., 2006). Showing that the IAT–HA is REFERENCES
strongly related to the RHS subscales would provide important Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advan-
evidence that the measure predicts whether people are actually tages of the HEXACO model of personality structure. Personality and Social
seen as humble across their relationships. Psychology Review, 11, 150–166.
 234 DAVIS ET AL.

Baumeister, R. F., Vohs, K. D., DeWall, C. N., & Zhang, L. (2007). How Kwan, V. S. Y., John, O. P., Robins, R. W., & Kuang, L. L. (2008). Conceptu-
emotion shapes behavior: Feedback, anticipation, and reflection, rather than alizing and assessing self-enhancement 835 bias: A componential approach.
direct causation. Personality and Social Psychology Review, 11, 167–203. Journal of Personality and Social Psychology, 94, 1062–1077.
Berry, J. W., Worthington, E. L., Jr., O’Connor, L., Parrott, L., III, & Wade, Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO
N. G. (2004). Forgiveness, vengeful rumination, and affective traits. Journal personality inventory. Multivariate Behavioral Research, 39, 329–358.
of Personality, 73, 1–43. Lopez, S. J., & Snyder, C. R. (Eds.). (2009). Oxford handbook of positive
Berry, J. W., Worthington, E. L., Jr., Wade, N. G., Witvliet, C. v. O., & Kiefer, psychology. New York, NY: Oxford University Press.
R. (2005). Forgiveness, moral identity, and perceived justice in crime victims McAdams, D. P., & Pals, J. L. (2006). A new Big Five: Fundamental principles
and their supporters. Humboldt Journal of Social Relations, 29, 136–162. for an integrative science of personality. American Psychologist, 61, 204–217.
Carver, C. S., & Scheier, M. F. (1994). Optimism and health-related cognition: McCrae, R. R., & Weiss, A. (2007). Observer ratings of personality. In R. W.
What variables actually matter? Psychology and Health, 9, 191–195. Robins, R. C. Fraley, & R. F. Krueger (Eds.), Handbook of research methods
Cattell, R. B. (1966). The Scree test for the number of factors. Multivariate in personality psychology (pp. 259–272). New York, NY: Guilford.
Behavioral Research, 1, 245–276. McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L., Jr., Brown,
Coke, J. S., Batson, C. D., & McDavis, K. (1978). Empathic mediation of S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships: II.
helping: A two-stage model. Journal of Personality and Social Psychology, Theoretical elaboration and measurement. Journal of Personality and Social
36, 752–766. Psychology, 75, 1586–1603.
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory Muthén, L. K., & Muthén, B. O. (2010). Mplus user’s guide (6th ed.). Los
(NEO–PI–R) professional manual. Odessa, FL: Psychological Assessment Angeles, CA: Muthén & Muthén.
Resources. Park, N., & Peterson, C. (2006). Moral competence and character strengths
Davis, D. E., Hook, J. N., Worthington, E. L., Jr., Van Tongeren, D. R., Gartner, among adolescents: The development and validation of the Values in Action
A. L., & Jennings, D. J., II. (2010). Relational spirituality and forgiveness: Inventory of Strengths for Youth. Journal of Adolescence, 29, 891–909.
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011

Development of the Spiritual Humility Scale (SHS). Journal of Psychology Powers, C., Nam, R. K., Rowatt, W. C., & Hill, P. C. (2007). Associations
and Theology, 38, 91–100. between humility, spiritual transcendence, and forgiveness. Research in the
Davis, D. E., Worthington, E. L., Jr., & Hook, J. N. (2010). Relational humil- Social Scientific Study of Religion, 18, 75–94.
ity: A review of definitions and measurement strategies. Journal of Positive Rowatt, W. C., Ottenbreit, A., Nesselroade, K. P., Jr., & Cunningham,
Psychology, 5, 243–252. P. A. (2002). On being holier-than-thou or humbler-than-thee: A social-
de Vries, R. E., Lee, K., & Ashton, M. C. (2008). The Dutch HEXACO Person- psychological perspective on religiousness and humility. Journal for the Sci-
ality Inventory: Psychometric properties, self–other agreement, and relations entific Study of Religion, 41, 227–237.
with psychopathy among low and high acquaintanceship dyads. Journal of Rowatt, W. C., Powers, C., Targhetta, V., Comer, J., Kennedy, S., & Labouff, J.
Personality Assessment, 90, 142–151. (2006). Development and initial validation of an implicit measure of humility
Emmons, R. A., & Kneezel, T. T. (2005). Giving thanks: Spiritual and religious relative to arrogance. Journal of Positive Psychology, 1, 198–211.
correlates of gratitude. Journal of Psychology and Christianity, 24, 140–148. Sandage, S. J. (1999). An ego-humility model of forgiveness: A theory-driven
Emmons, R. A., & McCullough, M. E. (Eds.). (2004). The psychology of grati- empirical test of group interventions. Unpublished doctoral dissertation, Vir-
tude. New York, NY: Oxford University Press. ginia Commonwealth University, Richmond, VA.
Exline, J. J., Baumeister, R. F., Zell, A. L., Kraft, A. J., & Witvliet, C. V. O. Sedikides, C., Gregg, A. P., & Hart, C. M. (2007). The importance of being
(2008). Not so innocent: Does seeing one’s own capability for wrongdo- modest. In C. Sedikides & S. J. Spencer (Eds.), The self (pp. 163–184). New
ing predict forgiveness? Journal of Personality and Social Psychology, 94, York, NY: Psychology Press.
495–515. Shepherd, S., & Belicki, K. (2008). Trait forgiveness and traitedness within the
Exline, J. J., Zell, A. L., Malcolm, W., DeCourville, N., & Belicki, K. (2008). HEXACO model of personality. Personality and Individual Differences, 45,
Does a humble attitude promote forgiveness? Challenges, caveats, and sex 389–394.
differences. In Women’s reflections on the complexities of forgiveness (pp. Snyder, C. R., Rand, K. L., & Sigmon, D. R. (2002). Hope theory: A member
235–251). New York, NY: Routledge/Taylor & Francis Group. of the positive psychology family. In C. R. Snyder & S. J. Lopez (Eds.),
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C. & Strahan, E. J. (1999). Handbook of positive psychology (pp. 411–419). New York, NY: Oxford
Evaluating the use of exploratory factor analysis in psychological research. University Press.
Psychological Methods, 4, 272–299. Steger, M. F. (2006). An illustration of issues in factor extraction and identifica-
Fredrickson, B. L. (1998). What good are positive emotions? Review of General tion of dimensionality in psychological assessment data. Journal of Person-
Psychology, 2, 300–319. ality Assessment, 86, 263–272.
Funder, D. C. (1995). On the accuracy of personality judgment: A realistic Tangney, J. P. (2005). Humility. In C. R. Snyder & S. J. Lopez (Eds.), Handbook
approach. Psychological Review, 102, 652–670. of positive psychology (pp. 411–419). New York, NY: Oxford University
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, Press.
self-esteem, and stereotypes. Psychological Review, 102, 4–27. Tangney, J. P. (2009). Humility. In S. J. Lopez & C. R. Snyder (Eds.), Oxford
Gregg, A. P., Hart, C. M., Sedikides, C., & Kumashiro, M. (2008). Every- handbook of positive psychology (pp. 483–490). New York, NY: Oxford
day conceptions of modesty: A prototype analysis. Personality and Social University Press.
Psychology Bulletin, 34, 978–992. Tangney, J. P., Stuewig, J., & Mashek, D. J. (2007). Moral emotions and moral
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the behavior. Annual Review of Psychology, 58, 345–372.
world? Behavioral and Brain Sciences, 33, 61–83. Tice, D. M., Butler, J. L., Muraven, M. B., & Stillwell, A. M. (1995). When
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance modesty prevails: Differential favorability of self-presentation to friends and
structure analysis: Conventional criteria versus new alternatives. Structural strangers. Journal of Personality and Social Psychology, 69, 1120–1138.
Equation Modeling, 6, 1–55. Vazire, S. (2006). Informant reports: A cheap, fast, and easy method for person-
John, O. P., & Robins, R. W. (1993). Determinants of interjudge agreement on ality assessment. Journal of Research in Personality, 40, 472–481.
personality traits: The Big Five 820 domains, observability, evaluativeness, Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation
and the unique perspective of the self. Journal of Personality, 61, 521–551. of brief measures of positive and negative affect: The PANAS scales. Journal
Kenny, D. A. (2004). PERSON: A general model of interpersonal perception. of Personality and Social Psychology, 54, 1063–1070.
Personality and Social Psychology Review, 8, 265–280. Worthington, E. L. (1998). An empathy-humility-commitment model of for-
Kwan, V. S. Y., John, O. P., Kenny, D. A., Bond, M. H., & Robins, R. W. giveness applied within family dyads. Journal of Family Therapy, 20, 59–76.
(2004). Reconceptualizing individual differences in self-enhancement bias: Worthington, E. L., Jr. (Ed.). (2005). Handbook of forgiveness. New York, NY:
An interpersonal approach. Psychological Review, 111, 94–110. BrunnerRoutledge.

View publication stats