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

net/publication/287419973

Is modesty a subdomain of humility?

Article in The Journal of Positive Psychology · December 2015

DOI: 10.1080/17439760.2015.1117130

CITATIONS READS

23 430

11 authors, including:

Don Davis Stacey McElroy-Heltzel

Georgia State University University of Iowa
286 PUBLICATIONS 8,675 CITATIONS 46 PUBLICATIONS 916 CITATIONS

SEE PROFILE SEE PROFILE

Elise Choe Joshua Hook

Boston University University of North Texas
28 PUBLICATIONS 844 CITATIONS 297 PUBLICATIONS 10,106 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Stacey McElroy-Heltzel on 13 January 2020.

The user has requested enhancement of the downloaded file.

 The Journal of Positive Psychology
Dedicated to furthering research and promoting good practice

ISSN: 1743-9760 (Print) 1743-9779 (Online) Journal homepage: https://www.tandfonline.com/loi/rpos20

Is modesty a subdomain of humility?

Don E. Davis, Stacey E. McElroy, Kenneth G. Rice, Elise Choe, Charles

Westbrook, Joshua N. Hook, Daryl R. Van Tongeren, Cirleen DeBlaere, Peter
Hill, Vanessa Placares & Everett L. Worthington Jr.

To cite this article: Don E. Davis, Stacey E. McElroy, Kenneth G. Rice, Elise Choe, Charles
Westbrook, Joshua N. Hook, Daryl R. Van Tongeren, Cirleen DeBlaere, Peter Hill, Vanessa
Placares & Everett L. Worthington Jr. (2016) Is modesty a subdomain of humility?, The Journal of
Positive Psychology, 11:4, 439-446, DOI: 10.1080/17439760.2015.1117130

To link to this article: https://doi.org/10.1080/17439760.2015.1117130

View supplementary material

Published online: 18 Dec 2015.

Submit your article to this journal

Article views: 554

View related articles

View Crossmark data

Citing articles: 8 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rpos20
The Journal of Positive Psychology, 2016
Vol. 11, No. 4, 439–446, http://dx.doi.org/10.1080/17439760.2015.1117130

Is modesty a subdomain of humility?

Don E. Davis *, Stacey E. McElroya, Kenneth G. Ricea, Elise Choea, Charles Westbrooka, Joshua N. Hookb,
a

Daryl R. Van Tongerenc, Cirleen DeBlaerea, Peter Hilld, Vanessa Placaresa and Everett L. Worthington Jr.e
a
Counseling and Psychology Services, Georgia State University, Atlanta, GA, USA; bPsychology departments, University of North
Texas, Denton, TX, USA; cPsychology departments, Hope College, Hope, MI, USA; dPsychology departments, Biola University,
LaMirada, CA, USA; ePsychology departments, Virginia Commonwealth University, Richmond, VA, USA
(Received 29 June 2015; accepted 29 October 2015)

Although many scholars have sharply distinguished humility from modesty, several research teams recently conceptual-
ized modesty as a subdomain of humility. In the present study, we compare these competing conceptualizations using
confirmatory factor analyses. In Study 1, a model with general humility (i.e. second-order factor) predicting the subscales
of the Relational Humility Scale (RHS) and the Modesty facet of the HEXACO-PI showed adequate fit; and modesty
showed one of the stronger factor loadings on the higher order factor. In Study 2, we used a set of items designed to
sharpen the potential distinction between modesty (i.e. interpersonal behavior involving moderating attention to self) and
humility. Then, we tested a model with a general humility factor (i.e. second-order factor) predicting subscales of the
RHS and modesty items. Once again, modesty items loaded strongly on the higher order factor.
Keywords: humility; measurement; modesty

Although the field of positive psychology has grown First, many lay people conflate modesty and humility.
considerably in recent years, psychological research on Indeed, some lay people consider modesty and humility
humility got off to a slow start, presumably due to defi- to be synonyms (Exline & Geyer, 2004; Gregg et al.,
nitional and measurement problems (Davis, Worthington, 2008), so scholars cannot count on participants to distin-
& Hook, 2010; Tangney, 2000). Namely, researchers guish the two constructs clearly. Moreover, some schol-
have disagreed on the core characteristics of humility, ars conflate the two in their definitions. For example,
and some have viewed self-reports of humility as invalid modesty has been defined as having intrapersonal and
due to a presumed modesty effect (i.e. higher actual interpersonal elements (Gregg et al., 2008). Intrapersonal
humility leads to more modest responses when rating modesty has been defined as having an accurate or mod-
one’s humility; Davis et al., 2010). However, in the last erate view of one’s strengths and limitations (Gregg
several years, psychological research on humility has et al.), which is nearly identical to how investigators
increased substantially. Fifteen years after Tangney’s have defined the intrapersonal aspect of humility (Davis
(2000) seminal position paper, there are over 150 empiri- et al., 2010). The interpersonal dimension of modesty
cal samples on humility, nearly 20 measures of humility has been described more narrowly than has the interper-
in circulation, but limited evidence for the presumed sonal dimension of humility. Namely, interpersonal mod-
‘modesty effect’ of humble individuals underreporting esty involves behaviors that mitigate attention to the self
their own humility (Davis et al., 2010). in order to reduce envy and jealousy in groups. One can
Despite clear progress, some definitional issues imagine a much broader range of situations in which
remain. One lingering issue is the need to clarify the humility is relevant, such as how someone handles con-
relationship between humility and modesty. Some early flict, power struggles, cultural differences, scholarly dis-
conceptual work sharply distinguished the two (Exline agreement, and so forth. Therefore, interpersonally,
et al., 2004; Tangney, 2000; cf. Gregg, Hart, Sedikides, humility refers to a broad range of situations, whereas
& Kumashiro, 2008). For example, Exline and col- modesty refers to a more circumscribed set of situations
leagues (2004) defined modesty as a ‘socially oriented involving appropriately seeking social affirmation and
virtue, a self-presentational stance that can be consistent attention without being off-putting.
with an inner sense of humility, but can also arise for Second, many humility measures include content
other reasons such as a presence of situational pressures’ associated with modesty (for a review, see Davis &
(p. 463). There are several reasons to evaluate this Hook, 2014). For example, consider three instruments
distinction empirically. purportedly measuring humility.

*Corresponding author. Email: [email protected]

© 2015 Taylor & Francis

440 D.E. Davis et al.

• The Honesty–Humility (HH) subscale of the navigate social norms for seeking attention. Respect and
HEXACO-PI (Lee & Ashton, 2004) includes two esteem from others is important, but people have to
subscales related to humility, but one is labeled negotiate for attention in socially appropriate ways, or
‘Modesty’ (e.g. the reverse-scored items, ‘I am their attempts may backfire and damage their reputation.
entitled to more respect than the average person is’ Modest behaviors attenuate the potential for jealousy and
or ‘I want people to know that I am an important escalation of conflict when someone receives attention,
person of high status’). These items appear to align such as receiving honor, getting a work promotion, or
with intrapersonal modesty, which would be very achieving high levels of success. The idea that humility
difficult to distinguish from the intrapersonal aspect involves several subdomains is promising, but it has not
of humility. been empirically evaluated.
• As the name implies, the Humility–Modesty sub-
scale of the Values in Action Strengths Inventory
(Peterson & Seligman, 2004) includes content that Overview and hypotheses
refers explicitly to interpersonal modesty (e.g. ‘I The purpose of the present study is to use confirmatory
never brag about my accomplishments’; ‘I rarely factor analyses (CFAs) to evaluate the plausibility of
call attention to myself’; ‘I have been told that McElroy et al.’s (2014) conceptualization of humility.
modesty is one of my most notable characteris- The model implies a second-order factor in which
tics’). The Humility–Modesty subscale is scored as general humility predicts several subdomains. In Study
a single factor. 1, participants rated a target person using the Relational
• Likewise, the Modesty facet of the NEO PI-R Humility Scale (RHS) (Davis et al., 2011), which has
Agreeableness scale (Costa & McCrae, 1992) had three subscales (i.e. Global Humility, Superiority
an item that explicitly refers to humility. Even [reverse scored], and Accurate View of Self). In addi-
though this item was dropped from the NEO PI-3, tion, participants also completed the Modesty facet of
it is telling that it loaded along with other modesty the HEXACO-PI-R (Lee & Ashton, 2004). We hypoth-
content on the prior version of the Modesty facet. esized that a second-order model would show adequate
fit. We also expected to see that the factor loading for
Thus, three of the most widely used measures of modesty in the second-order factor would be similar or
humility include modesty content. Yet they do not distin- greater in magnitude to the other loadings; a weak
guish modesty from humility. This suggests that those loading would not support McElroy’s et al.’s (2014)
who have constructed measures and adduced evidence of conceptualization.
psychometric adequacy of their measures have treated We started with the HEXACO-PI-R in Study 1
modesty as a part of humility. because many studies on humility use this measure. The
In an attempt to consolidate definitions of humility, HEXACO-PI-R has been critiqued as an assessment of
McElroy et al. (2014) proposed that, analogous to self- humility because items assess constructs such as modesty
efficacy or intelligence, which both involve subdomains, and honesty (e.g. Owens, Johnson, & Mitchell, 2013)
humility involve a variety of situations (i.e. subdomains) and not humility per se. Thus, arguably, the Modesty
that strain egotism, and make it difficult to remain other- subscale fails to do justice to Exline et al.’s (2004) con-
oriented (i.e. attentive and responsive to the needs of ceptual distinction – that is, that modesty refers to an
others) rather than selfish. Indeed, whereas virtues such interpersonal strategy of impression management that
as forgiveness imply a rather circumscribed context (i.e. one can easily fake. In Study 2, we sought to provide a
dealing with an offense), one can imagine a variety of stronger test of the distinction. We pilot tested a set of
contexts and behaviors associated with humility. Thus, items focused specifically on interpersonal modesty
there may be several types or subdomains of humility, based on the Exline et al. distinction. Then, we con-
and it is an empirical question whether behaviors in one ducted a similar set of model tests as in Study 1.
context correlate sufficiently with behavior in other con- Namely, we examined a model with general humility
texts to support the presence of a second-order factor (i.e. second-order factor) predicting the subscales of the
associated with general humility. In fact, scholarship on RHS (i.e. Global Humility, Superiority, and Accurate
humility has already identified several subdomains. For View of Self) and interpersonal modesty (i.e. newly cre-
example, intellectual humility refers to negotiation of ated items). Again, we expected the second-order model
ideas (McElroy et al., 2014); cultural humility refers to to show adequate fit. Most importantly, we expected to
one’s ability to navigate relationships involving cultural see interpersonal modesty show similar or stronger factor
differences that might make cooperation difficult (Hook, loadings as the subscales of the RHS; a weak loading
Davis, & Van Tongeren, in press). So as a potential sub- would provide evidence against the McElroy et al.
domain, modesty might refer to the ability to effectively (2014) conceptualization.
 The Journal of Positive Psychology 441

Study 1 Honesty–Humility
The purpose of Study 1 was to use CFA to evaluate the Participants rated a target person on the four-item Mod-
plausibility of McElroy et al.’s (2014) theorizing regard- esty facet of the HH subscale of the HEXACO-PI-100
ing general humility. We hypothesized that a second- (Lee & Ashton, 2004). Items are evaluated on a five-
order model, with general humility predicting several point rating (ranging from 1 = strongly disagree to
subdomains, would show similar fit, and thus constitute 5 = strongly agree). An example item is, ‘He or she
a more parsimonious model, than a four-factor model thinks that he or she is an ordinary person who is no
that does not include a second-order factor. We also better than others.’ The HH subscale scores have shown
hypothesized that the modesty subdomain would load evidence of estimated reliability, as well as of incremen-
strongly on the general humility factor. tal predictive validity above and beyond the Big Five
(for a review, see Ashton & Lee, 2007). One item per-
formed particularly poorly (i.e. He/she wouldn’t want
Method people to treat him/her as though he/she was superior to
Participants and procedure them), so we dropped this item and used the other three
items. (We also replicated all analyses using all four
Participants were 163 undergraduate students (103
items, and there were no substantive differences in our
women; 60 men) from a large urban university in the
findings.) The Cronbach’s alpha for the current sample
southeastern United States. Age was not reported in this
was .74.
sample due to an omission in the survey. The sample
was ethnically diverse (55.8% white/Caucasian, 20.2%
black/African American, 11.0% Asian/Asian American, Results and discussion
3.7% Latino/Latina, and 9.2% Other). Participants were
Means, standard deviations, and Cronbach’s alphas for
recruited from undergraduate classes and participated for
all measures are reported in Table 1. Full information
partial course credit. Given theorizing that humility is
maximum likelihood is the Mplus default used to address
especially important when strained (Davis et al., 2011),
missing data and generate unbiased parameter estimates.
we specifically recruited participants asked for individu-
Covariance coverage ranged from .982 to 1.00 and indi-
als who identified their father as very devoted to his
cated that item-level missingness was inconsequential.
work. We reasoned that this would lead to considerable
The covariance matrix of RHS, Modesty was analyzed
variability in perceptions of humility and related
with Maximum Likelihood (MLR) estimation using
subdomains, because father–child relationships involve
Mplus 7.11 (Muthén & Muthén, 2008–2012). To deter-
hierarchy that ought to challenge the practice of humility
mine overall model fit, we examined several fit indices:
(Davis et al., 2010). After completing questionnaires,
χ2, the comparative fit index (CFI), the standardized root
participants were debriefed and given the contact
mean square residual (SRMR), and the root mean square
information of the researcher should they have any
error of approximation (RMSEA). As a rule of thumb, a
questions.
CFI of approximately .95, an SRMR equal to or less
than .08, and an RMSEA equal to or less than .06
Measures suggest good fit (Hu & Bentler, 1999). For model com-
parisons, we interpreted declines in CFI of .01 or greater
Relational Humility Scale
as indication of degradation in fit (Cheung & Rensvold,
Participants rated a target person with the 16-item RHS 2002).
(Davis et al., 2011). Items are completed on a five-point We first examined a four-factor model in which
rating (ranging from 1 = completely disagree to 5 = com- items for each subscale were constrained to load on their
pletely agree). The RHS has three subscales: Global
Humility (e.g. ‘He/she is humble’), Superiority (e.g. ‘He/
she has a big ego’), and Self-Awareness (e.g. ‘He/she
Table 1. Correlations among factor scores in Study 1.
knows him/herself well’). Cronbach’s alphas for full-
scale and subscale scores ranged from .90 to .95. In 1 2 3 4
addition, the RHS has shown evidence supporting its
1. Global humility .95
construct validity. It was found to correlate with empathy 2. Superiority .59** .91
and forgiveness of an offender and positive relationship 3. Accurate view of self .49** .29** .86
characteristics with a parent, such as closeness and 4. Modesty .46** .67** .33** .74
positive and negative affect (Davis et al., 2011). M 3.66 3.54 3.87 3.58
Cronbach’s coefficient alpha for the current sample was SD 1.17 1.02 .87 .70
.89 for Global Humility, .91 for Superiority, and .86 for Note: Alphas are on the diagonal in bold.
Accurate View of Self. **p < .001.
442 D.E. Davis et al.

respective factor, and the factors were allowed to

correlate. Model comparisons are reported in Table 2.
Model 1 showed adequate fit, χ2 (146, N = 163)
= 230.75, p < .001, CFI = .946, RMSEA = .06,
SRMR = .06. Factor loadings ranged from .78 to .92 for
the Global subscale, .44 to .92 for the Superiority sub-
scale, .82 to .88 for the Accurate View of Self Subscale,
and .50 to .82 for the Modesty facet. Modesty was mod-
erately associated with global humility (r = .46,
p < .001), Superiority (r = .67; p < .001), Accurate View
of Self (r = .33, p < .001).
We then examined a second-order model with gen-
eral humility predicting the three subscales of the RHS
and the Modesty facet of the HEXACO-PI. This model
showed adequate fit, χ2 (148) = 244.51, p < .001,
CFI = .939, RMSEA = .06, SRMR = .07, and no sub-
stantive difference with the first-order model, Figure 1. CFA from Study 1.
ΔCFI = .007. As illustrated in Figure 1, standardized
estimates revealed that general humility strongly related
to Modesty (ϒ = .75, p < .001), Superiority (ϒ = .83, factor structure. Then, in a second sample, we conducted
p < .001), Global Humility (ϒ = .71, p < .001), and a series of CFAs following the same data analytic strat-
Accurate View of sSelf (ϒ = .46, p < .001). The second- egy used in Study 1. We tested the same hypotheses as
order factor accounted for 50.2% of variance in the first- in Study 1.
order factors.
These results provide initial support for the plausibil- Method
ity of understanding humility as having several subdo-
mains. Modesty was one of the three stronger loading Participants
factors on a higher order general humility factor. Never- Participants in Sample 2a (used for pilot testing items)
theless, due to critiques of the HH scale, we believed were 300 undergraduates (166 women, 134 men) from
this to be a weak test of the conceptual argument the same university as Study 1. The mean age was
advanced by Exline et al. (2004) and others. Therefore, 19.1 years (SD = 2.49). The sample was diverse in terms
in Study 2, we sought a stronger opportunity for our the- of race/ethnicity (57.3% white/Caucasian, 18.0% Asian/
orizing regarding humility and modesty to fail. Asian American, 15.3% black/African American, 4.0%
Latino/Latina, and 4.7% Other or did not report). Partici-
pants in this sample rated the level of modesty of one of
Study 2 their parents.
The purpose of Study 2 was to use CFA to evaluate a Participants in Sample 2b were 175 undergraduate
similar model as in Study 1. Rather than using the Mod- students (106 women; 66 men; 3 did not report sex) from
esty subscale of the HEXICO-PI, we used a scale another university, in the southeastern United States. The
designed specifically to assess interpersonal modesty. mean age was 24.75 years (SD = 5.59). The sample was
First, we pilot tested this measure on an independent ethnically diverse (42.3% African American/black, 55.8
sample in order to refine the items and determine their white/European American, 14.9% Asian/Pacific Islander,

Table 2. Model comparisons in Study 1 and Study 2.

χ2 df CFI RMSEA ΔCFI SRMR

Study 1
1. Four-factor model 230.75 146 .946 .06 .06
2. Four-factor higher order 244.51 148 .939 .06 .007 .07

Study 2
1. Five factor 420.83 242 .932 .07 .05
2. Four-factor method 406.66 235 .935 .07 .003 .04
3. Four-factor method higher order 406.73 237 .935 .07 .000 .04
 The Journal of Positive Psychology 443

8.0% Latino/Latina, 5.7% other, and 2.9% did not report). for the purpose of the present investigation, we first con-
Participants in Sample 2 rated the modesty and humility ducted a CFA to test the factor structure of this scale
of a leader. established in our pilot work. The covariance matrix of
RMS items was analyzed with Maximum Likelihood
(MLR) estimation using Mplus 7.11 (Muthén & Muthén,
Measures
2008–2012). The eight modesty items were used as indi-
Participants completed the RHS. Cronbach’s coefficients cators of the Modesty and Immodesty factors, which
alpha for Sample 1b were .95 for Global Humility, .89 were modeled as correlated factors. The two-factor
for Superiority, and .91 for Accurate View of Self. model of modesty items showed good fit, χ2 (19,
N = 175) = 43.52, p < .001, CFI = .98, RMSEA = .05,
Perceived modesty items SRMR = .036. Across the different fit indices, the RMS
appeared to replicate the two-factor structure. The two
To assess perceived modesty, we generated a list of 19
subscales were very strongly correlated with each other
face-valid items, drawn from various definitions of mod-
(r = .79, p < .001). Given the strong correlations between
esty (see Appendix in Supplemental data). Participants
factors, we also examined a single-factor model, which
rated the modesty of a target person indicating their
showed substantial degradation in fit, χ2 (20, N = 175)
agreement with items using a five-point rating (ranging
= 79.62, p < .001, CFI = .88, RMSEA = .14,
from 1 = strongly disagree to 5 = strongly agree). We
SRMR = .06, ΔCFI = .10.
sent these items to three experts who have regularly pub-
Next, we examined a five-factor model with the three
lished on humility/modesty. We used this feedback to
RHS and two RMS subscales. This model showed good
revise or drop several items, leading to a pool of 19
fit, χ2 (242, N = 175) = 420.82, p < .001, CFI = .932,
items. The correlation matrix for the 19 modesty items
RMSEA = .07, SRMR = .05. Given the strong correla-
was analyzed using a principal axis factoring estimation
tion between Immodesty and Modesty, we also examined
and promax rotation. A two-factor solution best fit the
a model with a method factor for reverse coded items
data based on the results of parallel analysis. Items were
(i.e. Immodesty and Superiority subscales) and four other
dropped that did not load at least .50 on their primary
factors (i.e. RHS subscales and one factor for the RMS).
factor, or that cross-loaded over .25 on any secondary
Degradation in fit did not occur, χ2 (235) = 406.66,
factor. After examining the content of items, the two fac-
p < .001, CFI = .935, RMSEA = .07, SRMR = .04,
tors were named Modesty (sample item: ‘He/she is the
ΔCFI = .003.
most modest person I know’) and Immodesty (sample
Last, we examined a second-order model involving
item: ‘He/she likes to be the center of attention’). The
a higher order factor (i.e. general humility), four lower
final version (which we called the Relational Modesty
order factors (i.e. Global Humility, Superiority,
Scale; RMS) consisted of eight items (four items per fac-
Accurate View of Self, and Modesty) and a method
tor). Descriptive statistics, factor loadings, eigenvalues,
factor (i.e. reverse scored items). Degradation in fit did
and variance accounted for are listed in Table 3. The fac-
not occur, χ2 (237) = 406.73, p < .001, CFI = .93,
tors were strongly correlated with each other (r = .52,
ΔCFI = .000, RMSEA = .07, SRMR = .04. Results are
p < .001). The Cronbach’s coefficients alpha for Sample
illustrated in Figure 2. The factor loadings were all
2b were .88 for Modesty and .84 for Immodesty.
strong: .99 for Modesty; .93 for Global Humility; .85
for Accurate View of Self; and .91 for Superiority.
Results and discussion Once again, these results provide evidence for the
Means, SDs, and correlations of constructs in Sample 2b plausibility of understanding humility as having several
are reported in Table 4. Because the RMS was created subdomains.

Table 3. Exploratory factor analysis of the modesty items in Sample 2a.

Item Mean SD N 1 2
He/she tends to share the credit with others .70 1.03 298 .88 .06
He/she shares the spotlight 1.28 1.14 297 .77 .07
He/she accepts deserved praise without making a big deal of it 1.15 1.04 295 .69 −.04
He/she is one of the most modest people I know .85 1.04 298 .68 −.03
He/she likes to call attention to his/her accomplishments 2.61 1.00 296 .10 .84
He/she tries to impress others with false modesty 2.77 .95 298 .13 .81
He/she likes to be the center of attention 2.87 .94 298 −.18 .68
He/she likes to talk about him/herself 2.51 1.20 297 −.27 .57
Note: Bold = primary factor loading.
444 D.E. Davis et al.

Table 4. Correlations among factor scores in Sample 2b. This issue is particularly timely, because the most
widely used measures of humility include modest content
1 3 4 5
(Davis & Hook, 2014). If McElroy et al. (2014) are cor-
1. Modesty .88 rect, then things are as they should be, because interper-
2. Global humility .91** .95 sonal modesty may be a central domain associated with
3. Superiority .90** .85** .89
humility. On the other hand, if modesty is not a subdo-
4. Accurate view of self .84** .80** .76** .91
M 3.61 3.62 3.78 3.74 main of humility, then it is a major problem that the
SD .86 .96 .92 .93 prominently used measures of humility have failed to
make this distinction. This would mean that much of the
Note: Alphas are on the diagonal in bold.
**p < .001. literature on humility has used measures with potentially
major problems with construct validity.
Across two samples, our findings certainly support
the promise of the subdomain approach. In both, mod-
esty had one of the strongest loadings on a general
humility factor. Thus, until there is more evidence for
the distinction, we contend that the weight of evidence
(i.e. content of existing measures; results of CFA) sup-
ports the idea that interpersonal modesty is a subdomain
of humility.
This does not preclude the interpretation that there
are several necessary, but not sufficient aspects of humil-
ity. For example, to be viewed as humble, one must cer-
tainly act modestly (i.e. handling attention in socially
acceptable ways), but one may also need to show integ-
rity across situations. For example, we doubt people will
perceive someone as humble who seems to act modestly
only when it seems socially beneficial. This hypothesis
could be easily tested using experimental methods that
Figure 2. CFA Study 2. have observers watch the same behavior but receive dif-
ferent information about the motives of the target person
The results of Study 2 replicated those of Study 1 (Tice, Butler, Muraven, & Stillwell, 1995).
with a different (and conceptually stronger) measure of
modesty that aligns well with the Exline et al. (2004) Limitations and future research
distinction between humility and modesty. Findings from
The present study had several limitations. First, we stud-
Study 1 and 2 converged; in fact, findings were even
ied college students in all three samples. One’s under-
stronger in Study 2 regarding the plausibility of a sec-
standing and practice of humility may develop greater
ond-order general humility factor that includes modesty
depth over time, so it is important to explore whether
as one of the several subdomains.
similar findings occur in middle-aged or older adults, as
well as younger children. In addition, in Study 1, we
General discussion focused on father–child relationships, but future studies
might examine whether results generalize in other con-
The science of humility is at a key stage. Various mea-
texts that might strain the practice of humility.
sures are now available. Refinement and consolidation of
Second, our method was correlational, but experi-
definitions and measures are needed. One issue that has
mental approaches might also shed greater light on the
bedeviled humility scholars has been the wide range of
centrality of modesty and humility perceptions. For
interpersonal contexts that seem relevant to viewing
example, researchers might randomly assign participants
someone as humble. Researchers have generally agreed
to observe responses to situations that strain humility
that, intrapersonally, humility involves having an accu-
(Tice et al., 1995). If modesty is a subdomain of humil-
rate view of self – not too high or two low (Davis et al.,
ity, then dispersing credit or bragging might strongly pre-
2011). However, existing measures include an array of
dict shifts in humility perceptions relative to others
interpersonal behaviors (Davis & Hook, 2014; Hook &
behaviors. Comparing how observation of different
Davis, 2014). We were particularly interested in evaluat-
behaviors affects humility perceptions is a promising line
ing modesty as a potential subdomain of humility
of work that might inform the dialogue regarding how to
(Exline et al., 2004; Tangney, 2000; cf. McElroy et al.,
situate humility and modesty.
2014).
 The Journal of Positive Psychology 445

Third, our study primarily focused on modesty Disclosure statement

content, but measures of humility include a variety of No potential conflict of interest was reported by the authors.
subdomains (Davis & Hook, 2014). Now that a number
of measures have been created, the field is in a critical
stage of honing and refining definitions and measurement
strategies. Researchers may consider having participants Funding
observe a variety of behaviors, rate perceptions of humil- This work was supported by the John Templeton Foundation
ity, and then examine which behaviors are most predic- [grant number 14979].
tive of humility judgments. Similarly, scholars might ask
people to nominate individuals who are high or low in Supplemental data
humility and then give reasons for their nomination.
Supplemental data for this article can be accessed here: http://
These nominations might clarify some of the primary dx.doi.org/10.1080/17439760.2015.1117130.
subdomains of humility.
Fourth, we only assessed perceptions of humility and
modesty in others. From a personality judgment tradition
References
(Vazire, 2010), trait language helps communicate reputa-
Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and
tional information about how someone tends to act
practical advantages of the HEXACO model of personality
within one’s relationships, so it makes sense that we structure. Personality and Social Psychology Review, 11,
started here. However, it is important to explore how 150–166.
self-reports may differ from other-reports. For example, Cheung, G. W., & Rensvold, R. B. (2002). Evaluating good-
humility researchers have generally feared that self-re- ness-of-fit indexes for testing measurement invariance.
Structural Equation Modeling: A Multidisciplinary Journal,
ports may lack validity if self-enhancement decreases as
9, 233–255.
true humility increases. However, prior theory and Costa, P. T., Jr., & McCrae, R. R. (1992). Four ways five
research on person perception has shown that self-reports factors are basic. Personality and Individual Differences,
can have higher predictive validity for intrapsychic 13, 653–665.
qualities (Vazire, 2010; Vazire & Carlson, 2011). Indeed, Davis, D. E., & Hook, J. N. (2014). Humility, religion, and
spirituality: An endpiece. Journal of Psychology and
individuals have access to their own thoughts, feelings,
Theology, 42, 111–117.
and motivations. Perceiving another’s humility involves Davis, D. E., Hook, J. N., Worthington, E. L., Jr., Van
watching what someone says and does and then inferring Tongeren, D. R., Gartner, A. L., Jennings, D. J., II, &
their mental state. A variety of individual differences Emmons, R. A. (2011). Relational humility: Conceptualiz-
may also affect self-reports of humility. For example, if ing and measuring humility as a personality judgment.
Journal of Personality Assessment, 93, 225–234.
one aspect of humility involves unselfish motivations,
Davis, D. E., Worthington, E. L., Jr., & Hook, J. N. (2010).
then some people may hold themselves to particularly Humility: Review of measurement strategies and conceptu-
high standards in that regard. This raises complex ques- alization as personality judgment. The Journal of Positive
tions about how one’s values related to humility affect Psychology, 5, 243–252.
long-term regulation of humility-related behavior. Exline, J. J., Campbell, W. K., Baumeister, R. F., Joiner, T.,
Krueger, J., & Kachorek, L. V. (2004). Humility and
modesty. In C. Peterson & M. Seligman (Eds.), Character
strengths and virtues: A handbook and classification
Conclusion (pp. 461–475). New York, NY: Oxford University Press.
In recent years, the study of humility has clearly shaken Exline, J. J., & Geyer, A. L. (2004). Perceptions of humility: A
free of the preoccupation with so-called intractable mea- preliminary study. Self and Identity, 3, 95–114.
Gregg, A. P., Hart, C. M., Sedikides, C., & Kumashiro, M.
surement problems. A variety of measures exist, and
(2008). Everyday conceptions of modesty: A prototype
now scholars are doing the hard work of consolidating analysis. Personality and Social Psychology Bulletin, 34,
definitions and evaluating which measures (or the combi- 978–992.
nation thereof) have the strongest evidence for construct Hook, J. N., & Davis, D. E. (2014). Humility, religion, and
validity. Furthermore, instead of remaining a rather spirituality: Introduction to the special issue. Journal of
Psychology and Theology, 42, 3–6.
focused specialty field, scholars are situating humility
Hook, J. N., Davis, D. E., & Van Tongeren, D. R. (in press).
within rather expansive literatures in personality and Shifting social motives and religious expression in a
social psychology on self-enhancement, narcissism, and globalized world. Religion, Brain & Behavior, 39–41.
impression management. We hope this study will initiate Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit
additional theory to clarify the subdomains of humility indexes in covariance structure analysis: Conventional crite-
ria versus new alternatives. Structural Equation Modeling:
and use multi-method strategies to evaluate the impor-
A Multidisciplinary Journal, 6, 1–55.
tance of certain subdomains over others, and how some Lee, K., & Ashton, M. C. (2004). Psychometric properties of the
subdomains may have greater primacy in certain contexts HEXACO personality inventory. Multivariate Behavioral
than others. Research, 39, 329–358.
 446 D.E. Davis et al.

McElroy, S. E., Rice, K. G., Davis, D. E., Hook, J. N., Hill, P. Tangney, J. P. (2000). Humility: Theoretical perspectives,
C., Worthington, E. L., Jr., & Van Tongeren, D. R. (2014). empirical findings and directions for future research. Jour-
Intellectual humility: Scale development and theoretical nal of Social and Clinical Psychology, 19, 70–82.
elaborations in the context of religious leadership. Journal Tice, D. M., Butler, J. L., Muraven, M. B., & Stillwell, A. M.
of Psychology and Theology, 42, 19–30. (1995). When modesty prevails: Differential favorability of
Muthén, L. K., & Muthén, B. (2008–2012). Mplus user’s self-presentation to friends and strangers. Journal of
guide. Los Angeles, CA: Author. Personality and Social Psychology, 69, 1120–1138.
Owens, B. P., Johnson, M. D., & Mitchell, T. R. (2013). Vazire, S. (2010). Who knows what about a person? The self-
Expressed humility in organizations: Implications for per- other knowledge asymmetry (SOKA) model. Journal of
formance, teams, and leadership. Organization Science, 24, Personality and Social Psychology, 98, 281–300.
1517–1538. Vazire, S., & Carlson, E. N. (2011). Others sometimes know us
Peterson, C., & Seligman, M. E. P. (2004). Character strengths better than we know ourselves. Current Directions in
and virtues: A handbook and classification. New York, Psychological Science, 20, 104–108.
NY: Oxford University Press.

View publication stats