Content uploaded by Baptiste Barbot
Author content
All content in this area was uploaded by Baptiste Barbot on Aug 06, 2018
Content may be subject to copyright.
BAPTISTE BARBOT
Creativity and Self-esteem in Adolescence: A Study of
Their Domain-Specific, Multivariate Relationships
ABSTRACT
Although empirical investigations on the Creative Self have historically started with a focus on self-
esteem, the literature on its relationship with creative performance remains thin and inconsistent, with esti-
mated relationships ranging from moderate and negative, to strongly positive. Discrepancies may be
explained by the domain-specificity of both creativity and self-esteem that have been widely overlooked in
this line of work. Therefore, this study explores the multivariate relationships between creativity in three
domains (Music, Literary-Verbal, Graphic) and self-esteem in seven domains (e.g., Academic, Emotional)
among 170 adolescents. Creative productions were scored by four raters, and latent consensus in each
domain captured using a multi-informant latent-consensus model in SEM. This model was further extended
in a structural model reveling that (a) creativity is mainly domain-specific, and (b) the contribution of
domain-specific self-esteem on domain-specific creativity greatly varies according to both the domains of
creativity and self-esteem. Up to 30% of the variance in creative performance was explained by “domain-
relevant” self-esteem facets, and a moderate contribution of creative self-esteem across creativity domains.
Results are discussed in light of several important methodological directions for this line of work, as well as
its implications for creativity-based interventions designed to support positive self-esteem development in
adolescence.
Keywords: creativity, adolescence, multidimensional self-esteem, domain-specificity, measurement.
“The greater the artist, the greater the doubt. Perfect confidence is granted to the less talented as a consola-
tion prize.” (Hughes, 1996, p. 72). This quote from art critique Robert Hughes reflects some of the popular
beliefs surrounding the inner self of eminent creators and artists. Despite numerous examples of suicidal
and self-injuring “Big-C” creators (i.e., people recognized for their eminent creative contribution; Kaufman
& Beghetto, 2009) such as Vincent van Gogh, the academic literature on the relationship between creativity
and self-esteem (a critical predictor of suicidality; e.g., Wild, Flisher, & Lombard, 2004) appears far more
optimistic. A review of scholarly work on the topic (Lebuda & Csikszentmihalyi, 2017) concluded that emi-
nent “Big-C” creators may be distinguished from “Pro-c” creators (i.e., professional creators who have not
reached eminent achievements), notably by their higher self-esteem and related “core self-evaluation” (e.g.,
Judge, Erez, Bono, & Thoresen, 2002). By extension, it is generally believed that a high self-esteem is a natu-
ral characteristic of creative individuals in general (“little-c” level), or in other words, that self-esteem con-
tributes positively to creativity (e.g., VanderArk, 1989).
However, to date, this view relies on a very thin and inconsistent body of scientific evidence with only a
shy dozen published empirical studies that have included both measures of creativity and self-esteem (Deng
& Zhang, 2011). Such studies have often turned a blind eye on the very nature of the creativity tasks at
hand, and more importantly, to the domain of creative activity considered. Is self-esteem an equally useful
“ingredient” of creativity when children or adolescents are asked to generate original uses for a brick, or
when prompted to improvise a dance? This simple question points to the more general question of whether
the relationship between self-esteem and task performance varies according to situational circumstances. As
noted by Wang and Wang (2016), self-esteem could be even more engaged in creativity tasks that elicit
heightened evaluation stress, which would otherwise impede creative performance (Silvia & Phillips, 2004).
On this ground, it is likely that distinct task-specific demands differentially engage distinct person-level
resources such as self-esteem. This is consistent with the well-established domain-specific view of creativity
(Baer & Kaufman, 2005; Barbot & Tinio, 2015). Further, much as creativity is understood as partly domain-
specific, self-concept and esteem are also acknowledged as domain-specific entities (Marsh, Craven, &
The Journal of Creative Behavior, Vol. 0, Iss. 0, pp. 1–14 ©2018 by the Creative Education Foundation, Inc. ÓDOI: 10.10 02/jocb.365 1
Debus, 1998). This domain-specificity however, has rarely been accounted for in creativity studies. Concep-
tually, physical self-esteem may be domain-relevant when improvising a dance whereas academic self-esteem
may not be.
Together, the literature on the relationship between creativity and self-esteem is inconclusive perhaps
due to the differential contribution of (domain-specific) self-esteem to (domain-specific) creativity. Yet, this
hypothesis has never been investigated. Therefore, in the work presented here, the domain-specificity of both
creativity (in three production-based domains) and self-esteem (across seven domains of self-evaluation) is
accounted for with the underlying hypothesis that relationships between both constructs are demand-rele-
vant. Clarifying these relationships is not trivial: this line of work could help understand far more than the
characteristics of eminent creators, or the person-level resources contributing to successful creative outputs.
It may, in turn, inform how to tailor creativity-based interventions to help people build or sustain a positive
self-esteem or, reciprocally, how to build self-esteem to facilitate engagement in creative activities. The for-
mer may be particularly relevant for populations with vulnerable self-esteem, specifically adolescents. A brief
overview of adolescents’ developmental context illuminates how creativity and self-esteem relationships may
reasonably be interpreted as more differentiated and domain-specific than previously conceptualized.
GENERALITY-SPECIFICITY OF CREATIVITY AND SELF-ESTEEM: THE DEVELOPMENTAL CONTEXT
OF ADOLESCENCE
Adolescence is a critical period for both the development of identity and creativity (Barbot & Heuser,
2017; Dollinger, Clancy Dollinger, & Centeno, 2005; Rothenberg, 1990). With the onset of puberty, cognitive
development, and new external demands, adolescents engage in the formation of their identities, which is
often associated with confusion regarding the self. It results in typical loss of self-esteem, with the lowest
levels of the whole life-span attained during that period (e.g., Robins, Trzesniewski, Tracy, Gosling, & Potter,
2002). Further, if self-esteem is a rather unified construct in childhood with almost undistinguishable facets
(Jacobs, Lanza, Osgood, Eccles, & Wigfield, 2002), there is a diversification of self-esteem into domain-speci-
fic facets as individuals invest or disinvest specific areas of experience (Harter, 1990; Marsh et al., 1998).
1
Among the commonly identified “non-academic” domains of the self, social, emotional, and physical
self-esteem are the most salient (Shavelson, Hubner, & Stanton, 1976).
2
Often focused on the study of self-
concept, some have argued for the broader consideration of specialized areas of experience including aca-
demic domains (Marsh, Byrne, & Shavelson, 1988) and artistic or creative domains (Kaufman, 2012; Vis-
poel, 1995). The latter makes solid conceptual sense given that adolescence is a period of identification and
crystallization of creative pursuits (Barbot, Lubart, & Besanc
ßon, 2016; Beghetto & Dilley, 2016). By extension
(creative) task-oriented commitment is seen as an organizing principle of both creative self-perceptions and
actual creative ability (Barbot & Tinio, 2015; Plucker & Beghetto, 2004): By concentrating their efforts in
creative tasks that interest them, adolescents’ will progressively “specialize” their creative potential, making it
a more domain-specific ability. This specialization, in turn, leads to the apparent domain-specificity of cre-
ativity and limited support for a “c-factor” (unidimensional and monolithic) of the creativity entity (Barbot,
Besanc
ßon, & Lubart, 2016; Barbot & Tinio, 2015).
ARE CREATIVITY AND SELF-ESTEEM RELATED?
Although there is resurfacing interest in the relationship between dimensions of the self and creativity,
the evaluative component of the self, namely self-esteem, has received limited attention (Barbot & Heuser,
2017). Indeed, the nomological network of the “creative self” has boomed these past few years, with the
introduction of myriads of related concepts investigated as predictors of creative performance, including cre-
ative identity (e.g., Petkus, 1996), creative self-beliefs and self-concepts (e.g., Karwowski, 2016; Karwowski &
Lebuda, 2015), or one of its most enthusiastically embraced variation, creative self-efficacy (e.g., Beghetto,
2006; Royston & Reiter-Palmon, 2017). However, self-efficacy remains an inaccurate proxy for self-esteem
(Bandura, 1997). One’s conviction regarding their ability to perform specific tasks (self-efficacy), may not
lead to pride or feeling of self-worth (self-esteem) from these convictions (e.g., Karwowski & Barbot, 2016).
For example, an adolescent may have a rather high creative self-efficacy in music, but that perception of
1
According to this view and based on empirical findings, global and newly formed domain-specific self-esteem are not organized
hierarchically but have the same “level” in the structure of the Self (Marsh et al.,1998).
2
Shavelson et al. (1976) and further authors have used the terms “self-concept” and “self-esteem” interchangeably. For the sake
of clarity, the term “self-esteem” only is used throughout this work.
2
Creativity and Self-Esteem
competence may be unrelated to his or her otherwise low creative self-esteem, or even his or her global self-
esteem in general. Although the investigation of the relationship between creativity and the Self was histori-
cally initiated by a focus on self-esteem (e.g., Coopersmith, 1967), empirical support for the creativity-self-
esteem relationship remains limited since then. Adding to the scarcity of such investigations, published stud-
ies have all relied on a broad range of creativity measures, including divergent thinking tasks, production-
based tasks, self-reported creativity, or creative achievements, which all tap into distinct aspects of the con-
struct of creativity. As a result, estimates of the relationship between “generic-creativity” and self-esteem
have greatly varied, ranging from a moderate negative relationship (r=.43) when considering peer-rated
creativity among high ability Chinese students (Lau, Li, & Chu, 2004) to a strong and positive relationship
(r=.67) when considering teacher-rated creativity of US preschoolers (Kemple, David, & Wang, 1996).
However, the bulk of this line of work has reported (perhaps due to publication biases), positive relation-
ships of marginal effect sizes, within the r=.15–.30 range (Cantero, Alfonso-Benlliure, & Melero, 2016;
Goldsmith & Matherly, 1988; Jaquish & Ripple, 1981; Pretz & Nelson, 2017; Rank, Nelson, Allen, & Xu,
2009; Wang & Wang, 2016).
In a tentative meta-analysis of 17 studies (Deng & Zhang, 2011),
3
the authors acknowledged that
observed effects may not be fairly comparable due to the range of methods used. This supports more recent
meta-analytical efforts urging creativity researchers to be more conscious of the type of creativity assessment
used when considering the relationship between creativity and personality-related constructs (Puryear, Ket-
tler, & Rinn, 2017). Consistently, Deng and Zhang (2011) observed that self-esteem in relation to product-
based assessments of creativity lead to heterogeneous findings (i.e., unreliable estimates of effect size),
whereas it leads to correlations of about .29 with measures of creative-personality. As outlined above, diver-
gences with respect to the age-groups studied may also account for some of the inconsistencies observed, as
both creativity and self-esteem may progressively become more domain-specific entities throughout adoles-
cence.
PRESENT STUDY
The research presented here was designed to address the main inconsistencies of the creativity and self-
esteem literature by considering creativity in multiple domains (Music, Literary-Verbal, Graphic), as it
relates to self-esteem in multiple domains as well (e.g., Emotional, Academic or Creative self-esteem). Ado-
lescence was chosen as a relevant developmental context that assumes the diversification of creative pursuits
and the differentiation of both subdomains of self-esteem and creativity. A general theory-driven hypothesis
underlying this work was that the association between creative performance and self-esteem varies by
domain, notably based on the specific demands of each creative endeavor. For example, due to the engage-
ment of the body and gestures in musical creative expression, physical self-esteem may be more clearly rele-
vant to musical creativity (e.g., Barbot & Webster, 2018) than to other creative outlets such as drawing or
writing. Finally, because most creativity studies have often limited their focus on one to two domains of cre-
ative manifestations (Kaufman & Baer, 2012; Kaufman, Baer, Cropley, Reiter-Palmon, & Sinnett, 2013), a
secondary, yet, non-trivial objective was to further examine the issue of generality-specificity of creativity in
the investigated domains, representing one of the rare attempts to integrate more than two performance
domains in the same study.
METHOD
PARTICIPANTS
The data presented in this study were assembled from two related studies (see procedure) and involved
170 adolescents (46.5% female, 53.5% male) sampled from two middle-school and two high-schools in a
large metropole in France (both urban and suburban settings, as well as a range of sociodemographic back-
grounds were represented). Participants were schooled in 9th to 12th grade with a large proportion (60.6%)
in 10th grade. Their age ranged 13–19 (M
age
=16.39, SD
age
=1.15) with females and males not differing
statistically with respect to age (F[1, 167] =1.81, p=.18).
3
Some of which including measures of self-efficacy rather than self-esteem, and all studies using different indicators of creativity.
3
Journal of Creative Behavior
MEASURES
Self-esteem measures
Multidimensional self-esteem was measured with the Multidimensional Scale of Self-Esteem (EMES; Bar-
bot, Safont-Mottay, & Oubrayrie-Roussel, 2018), a 48 self-rated item scale (using 5-points Likert-type scales)
designed for adolescents, yielding six domain-specific self-esteem scores in three core “non-academic” areas
(Shavelson et al., 1976): Emotional (e.g., “I feel good with myself”), Physical (e.g., “I am proud of my body”),
and Social (e.g., “Others don’t trust me”); One general Academic domain (e.g., “I am proud of my academic
performance”), as well as a scale for Future self-esteem (e.g., “I trust in my future”), referring to one’s projec-
tion of anticipated value of oneself as an adult. The EMES also has a scale for (general) Artistic-Creative
self-esteem (e.g., “I feel that I am skilled in any kind of artistic outlets”), included to meaningfully contrast
with the more “convergent” domains of Academic self-esteem. This instrument has demonstrated strong
internal consistency, test–retest reliability, and evidence of criterion validity with other multidimensional
measure of self-esteem such as the SPPA (Harter, 1990), Big Five personality, and social desirability (Barbot
et al., 2018). Global self-esteem was measured with the Rosenberg Self-Esteem Scale (RSES; French version
by Vallieres & Vallerand, 1990) a 10 self-rated item measure (e.g., I am able to do things as well as most other
people”) that is well established for its strong psychometric properties across cultural contexts (Gnambs,
Scharl, & Schroeders, 2018).
Creativity measures
Three production-based tasks were employed to assess creative performance in Music, Literary-Verbal,
and Graphic domains. All tasks focused on “convergent-integrative” thinking requiring participants to
combine imposed elements in a creative way, culminating into a unique, elaborated creative production
(Barbot, Besanc
ßon, & Lubart, 2015). For the Music domain, the Musical Expression Test (MET; Barbot &
Lubart, 2012) was used. Structured as an individual, one-hour introduction to music composition work-
shop, the MET yields a 30-second elaborated musical piece that is composed using a computer-based
audio sequencer and 12 playful instruments (e.g., glockenspiel, tambourine) and objects (e.g., comb, plastic
tube). After two activities introducing the sound-production set and use of the sequencer in a standardized
manner, the composition task (30 min) allows participants to record 30-s musical segments which, once
played simultaneously, overlay to form an integrated piece (up to eight segments may be included in the
final composition).
For the Literary-Verbal and Graphic domains, two tasks from the Evaluation of Potential Creativity
(EPoC; Lubart, Besanc
ßon, & Barbot, 2011) were employed. The Literary-Verbal task (IV2) consisted of pro-
ducing a complete story based on the integration of three imposed characters: An elderly person, a child,
and a bird. Students were prompted to write down their story on a provided 2-sided paper-pencil form to
control for length, within a maximum of 50 min.
4
For the Graphic domain (IG2), using a 12-coloursmarker
set, students were asked to produce an original drawing within 15 min, combining at least four of eight pic-
tures of familiar objects provided to them on a photo sheet. Both tasks, and the EPoC in general, have
demonstrated robust psychometric properties with respect to reliability, criterion, and factorial validity (Bar-
bot, Besanc
ßon, et al., 2016; Lubart et al., 2011).
PROCEDURE
Data sampling
Data used for this work were sampled from two related studies involving partly common measures
(n
studyA
=46, and n
studyB
=124), assembled in a way that resembles classic planned missingness design
(Graham, 2009). This type of design has recently shown great promise for the field of creativity research
with respect to administration time and scoring procedure (F€
urst, 2018). For the present work, all partici-
pants who completed the music task across both studies were included (n=99; 58% of the total sample),
some of whom also completed the drawing task, while others completed both the drawing and the story-
writing task. To ensure sufficient covariance coverage across creativity measures, an additional subsample
(n=71; 42% of the total sample, sampled form study B) was randomly selected from a larger pool of par-
ticipants who completed both the drawing and the story-writing task. This additional subsample size was
determined based on prior power analysis in such a way that the total sample size would allow for detection
4
For the needs of the main study program, the time allotted for this task was extended to 50 min, but the original time set in
the original version of the task is 15 min.
4
Creativity and Self-Esteem
of effect of at least medium size (r=.30) with sufficient power (.80) at risk a=.05. In all, participants
completed either one (27%), two (50%), or the three domain-specific creativity tasks (23%). According to
the study source, 74% of the sample was administered the global self-esteem measure, whereas all partici-
pants (100%) completed the EMES.
Measure administration
Both studies were approved by the Vice Chancellor of education of the metropolitan district used for
data collection, and all appropriate consents were secured. They both followed the same procedure with
respect to the administration of measures: one main data collection session (conducted in classroom set-
tings, within 50 min’ sections), included all self-report measures, as well as the drawing task. Separately, the
story-writing task was administered in up to 50-min collective sessions to volunteer participants within
2 weeks prior to, or following, the main data collection. The music task was administered in one-hour indi-
vidual sessions scheduled within 6 weeks prior to, or following, the main data collection. All assessment ses-
sions typically occurred during class or work-study time.
Creative productions ratings
Four raters (two males and two females) including qualified experts (two faculty with over 10 years of
expertise in psychology of creativity), and two quasi-experts (i.e., graduate students working under their
supervision; see Kaufman & Baer, 2012; Kaufman et al., 2013) independently rated all creative productions
gathered for the present study. Three separate rating sessions (one per creativity domain) were implemented
in an online adaptation of the Consensual Assessment Technique (Consensual Assessment Technique-inter-
face, CAT-i; Barbot, Orriols, & Pouyade, 2008). CAT-i implements the CAT following its main guidelines
(e.g., randomization of production for each rater), to optimize judgments of raters remotely accessing the
creative productions through the platform. Within each rating session, productions were rated for creativity
on 7-point Likert-type scales, and the raters had the possibility to review their scoring (sorting the produc-
tions by rating) to refine their judgments prior to validating their work.
ANALYTIC PLAN
After a set of preliminary missing values analyses, internal consistency and inter-rater reliability checks,
two distinct sets of analyses were conducted. The measurement set focused on examining the relationship
between domains of creativity, using the Multi-Informant, Latent-Consensus modeling approach (Barbot,
Bick, et al., 2016). This approach uses structural equation modeling (SEM) to partition-out scores gathered
by multiple informants on the same objects, in terms of shared variance (i.e., latent consensus), and residual
variance (i.e., unshared raters’ subjectivity and measurement error). Residual covariance within each raters’
variables (but not across raters) is freely estimated to account for raters’ cross-domains biases. This
approach was used to test alternative measurement models in line with domain-general, domain-specific,
and hybrid views of performance-based creativity measurement.
In the structural set, the MILC model was extended into a structural regression model, using the three
domain-specific MILC variables as endogenous latent variable (DVs), and the indicators of self-esteem, used
as exogenous manifest variables (IVs). This approach estimates the unique contribution of each predictor to
each dependent variable (controlling for exogenous and endogenous covariance), over and above the contri-
bution of all other predictors. All model parameters were estimated using Full Information Maximum Like-
lihood (FIML), which makes full use of the available data to estimate parameters (including from cases that
are only partially complete). Assessment of model fit relied on established indexes and cutoffs in the litera-
ture, including the non-significance of the Chi-square tests of fit, a ratio Chi-square to degree of freedom
(DF) <2, a comparative fit index (CFI) >.95, and the root mean square error of approximation (RMSEA)
<.08 (e.g., Kline, 2015).
RESULTS
PRELIMINARY ANALYSES
Per the study data sampling design, missing values analysis showed a total of 25.3% missing data, which
was under the amount typically obtained with the classic 3-form planned missingness design (e.g., Graham,
2009). Specifically, missing values ranged from 4.1% for the EMES (unplanned missingness), to 41.8% for
the Music domain (with 26.5%, 27.6%, 37.1% for RSES, Drawing, and Story-writing task, respectively).
However, the pattern of missing data was completely random (Little’s MCAR test v
2[df]
=98.9
[101]
,
5
Journal of Creative Behavior
p=.54), and unrelated to participants’ gender and age, as reflected by Little’s MCAR test incorporating
these background variables (v
2[df]
=153.3
[135]
,p=.14). On this ground, FIML could be validly used to
make inferences on the whole sample (See electronic Appendix S1 for all FIML estimated pairwise correla-
tions between all variables in this study). Reliability analyses of all self-report measures were conducted,
showing overall good internal consistency (mean a=.76), ranging from limited (Future self-esteem a=.62,
Social self-esteem a=.65) to excellent (Global self-esteem a=.87, Physical self-esteem a=.84). Although
the expert-based ratings of creative production were handled through the aforementioned MILC approach
(resulting in estimated domain-specific scores of creativity free of raters’ unshared subjectivity and random
error), inter-rater reliability was computed for reference, showing acceptable degree of agreement for the
music productions (a=.65), and good agreement for story (a=.81), and drawing productions (a=.83).
DOMAIN GENERALITY-SPECIFICITY MODELS
The unidimensional MILC model aligned with the “c-factor” view of creativity (one factor loaded into
all experts’ ratings variables across domains) returned a poor fit to the data (v
2[df]
=259.3 [42],p<.001,
v
2
/
[df]
=6.174, CFI =.664, RMSEA [CI-90%] =.175 [.155–.196]) and was deemed unacceptable for further
analyses. In contrast, the domain-specific correlated model showed a nearly perfect fit to the data
(v
2[df]
=28.4 [39],p=.90, v
2
/
[df]
=.728, CFI =1.00, RMSEA [CI-90%] =.000 [.000–.025]), so was its
hierarchical extension
5
(v
2[df]
=29.4 [41],p=.91, v
2
/
[df]
=.717, CFI =1.00, RMSEA [CI-90%] =.000
[.000–.021]), although not significantly better than the domain-specific, correlated model (Dv
2[df]
=1.01
[2]
,
p=.60). Using the domain-specific correlated MILC model as baseline (Figure 1), factor loadings were siz-
able across domains with an average standardized k=.67 (ranging from k=.43, p=.002, to k=.85, all
other ps<.001), suggesting substantial amount of shared variance across informants within each domain.
Estimated relationships between domains showed only a moderate latent correlation between story and
drawing MILC variables (latent r=.44, p<.001), and no relationships between music and drawing MILC
variables (latent r=.10, p=.57), nor music and story MILC variables (latent r=.05, p=.81). Corre-
spondingly, loadings obtained with the hierarchical model showed that the domain-general factor loaded
substantially into both the drawing and the story MILC factors (standardized k=.67 and .63, p<.001,
respectively), but was unrelated to the music MILC factor (standardized k=.10, p=.68). In other words,
the domain-general factor did not explain the music MILC factor (R
2
=.01), whereas it explained about
40% of the variance in the drawing and the story MILC factors (R
2
=.45 and .39, respectively).
MULTIDIMENSIONAL SELF-ESTEEM AND CREATIVITY MODELS
The domain-specific MILC model was extended into a structural regression model, whereby all the
self-esteem indicators were used as predictors of the domain-specific creativity MILC variables (Fig-
ure 2). This model, returned an excellent fit to the data (v
2[df]
=119.5 [102],p=.11, v
2
/
[df]
=1.171,
CFI =.974, RMSEA [CI-90%] =.032 [.000–.053]) allowing for a valid interpretation of model parame-
ter estimates. As shown in Table 1, the contribution of domain-specific self-esteem on domain-specific
creativity greatly varies according to both the domain of creativity and the domain of self-esteem under
consideration. Six out of the 21 tested pathways of associations showed significant effects. Musical cre-
ativity was strongly predicted by global self-esteem (standardized c=.54, p<.05) and creative self-
esteem (standardized c=.35, p<.01); Story-writing was predicted by a combination of three domain-
specific self-esteem with effects of medium size, including future self-esteem (standardized c=.23,
p<.05), academic self-esteem (standardized c=.27, p<.05), and creative self-esteem (standardized
c=.23, p<.01). Finally, graphic creativity was only significantly predicted by creative self-esteem (stan-
dardized c=.26, p<.01). Together, the combined effect of all self-esteem variables on the MILC cre-
ativity variables was sizable, explaining 30.7% of the variance in musical creativity, 21.3% of story-
writing creativity, and 10.1% of creativity in drawing.
Noteworthy in this analytical set was the rather equivalent contribution of creative self-esteem across all
three creativity domains which suggested that this domain of self-esteem might have a somewhat more “do-
main-general” contribution on creative performance. To evaluate this hypothesis, the hierarchical MILC cre-
ativity model was used in a structural regression extension analog to the previously described model, in
which all self-esteem variables predicted the general creativity factor. This model showed an acceptable fit to
5
In order to identify the Hierarchical MILC model, a constraint of equality of variance of the first order latent variables (i.e.,
domain-specific MILC)’s residual terms was set, and the variance of the higher order variable (i.e., “C”) was set to 1.
6
Creativity and Self-Esteem
the data (v
2[df]
=146.4 [118],p=.04, v
2
/
[df]
=1.241, CFI =.958, RMSEA [CI-90%] =.038 [.009–.056]),
despite a substantial degradation compared to the previous model (Dv
2[df]
=29.9
[16]
,p=.04). Estimates
obtained in this last analytical set revealed that creative self-esteem was the only significant contributor to
the “general creativity” factor (standardized c=.43, p<.01), with the combined effect of all self-esteem
variables explaining 31.8% of its variance.
FIGURE 1. Domain-specific multi-informant latent consensus (MILC) model and its standardized
estimates. Note. All estimates presented are standardized; Factor loadings are all significant at
p<.001; For clarity, within-rater correlated residuals are only represented for Judge 1, but the
pattern extends to all four judges.
7
Journal of Creative Behavior
DISCUSSION
Despite the extensive acknowledgement of the dimensionality and domain-specificity of both creativity
(Baer & Kaufman, 2005; Barbot & Tinio, 2015) and self-esteem (Harter, 1990), the present study was the
first to investigate their relationship in a multivariate, domain-specific fashion, including three domains of
creative activity and seven domains of self-evaluation. Results confirmed that (a) creativity remains a mainly
domain-specific phenomenon, although the contribution of common factors is not to be excluded for some
subdomains of creative activity (here literary-verbal, and graphic), and (b) that the contribution of domain-
specific self-esteem on domain-specific creativity greatly varies according to both the domain of creativity
and the domain of self-esteem under consideration.
In spite of several limitations, this study has also contributed to (c) several important methodological
directions for this line of research, and for the product-based assessment of creativity in general.
DOMAIN GENERALITY-SPECIFICITY OF CREATIVITY IN THREE DOMAINS
This study is one of the rare attempts to integrate more than two domains of creative production in the
same study (Kaufman & Baer, 2012; Kaufman et al., 2013), including a musical composition task, a story-
FIGURE 2. Structural regression model of domain-specific self-esteem to domain-specific creativity. Note.
All estimates presented are standardized. For clarity, within-rater correlated residuals are not
represented; non-significant structural paths are dashed; the size of the significant structural
paths is proportional to their corresponding standardized estimate.
8
Creativity and Self-Esteem
writing task, and a drawing composition task. The highest association observed between domains was for
the story-writing and the graphic task (latent correlation r=.44, p<.001), whereas performance in music
composition was independent from performance in the other domains (latent r<|.10|). Correspondingly, a
“c-factor” of creativity was only meaningful for the story-writing and graphic domains, explaining a sizable
share of the variance in these domains’ latent performance, whereas this factor appeared unrelated to the
performance in music composition.
First, it is important to note that the Multi-Informant Latent-Consensus (MILC) approach used to cap-
ture latent performance in each creativity domain, yields disattenuated relationships between performance
domains (i.e., controls for unreliability). As a result, the correlation between the story-writing and the gra-
phic tasks used here is larger than what is usually reported in the literature with these tasks (Barbot,
Besanc
ßon, et al., 2016; Lau, Cheung, Lubart, Tong, & Chu, 2013; Lubart et al., 2011), typically in the .30
range. Second, although limited to three domains, this pattern of associations resembles the distinction
between different subtypes of creative activities as outlined in the Kaufman Domains of Creativity Scale (K-
DOC; Kaufman, 2012). In this model, common perceptions of fifty creative activities were empirically cate-
gorized in five broad domains in which both fiction writing and musical composition loaded in a common
“performance” factor, whereas drawing-related activities loaded into an “artistic” common factor (Kaufman,
2012). Although the present study suggests that actual performance in story-writing relates more to the
drawing composition activity than the musical composition activity, the findings are in line with a possible
contribution of distinct common factors of creativity according to broader content domains.
However, it is important to keep in mind that the estimated contribution of such “c-factors” of creativity
would dramatically decrease if tasks of a different nature (e.g., engaging mainly divergent thinking) were also
included across the domains investigated in this study. Indeed, the relationship between distinct domains of
creative activity may not only be explained by domain-specificity, but also by the nature of the thinking
processes mainly involved in each task (Barbot, Besanc
ßon, et al., 2016). By construction, both the drawing
and the story-writing tasks primarily engage convergent-integrative thinking processes (Lubart et al., 2011)
TABLE 1. Parameters Estimates for the Multivariate Structural Regression Model
Creativity domain
(Endogenous)
Self-esteem
domain (Exogenous) cEstimate (SE)
a
cSt. estimate
b
p*
Music Emotional 0.111 (0.221) 0.100 0.614
Physical 0.169 (0.147) 0.169 0.249
Social 0.157 (0.206) 0.102 0.445
Future 0.164 (0.225) 0.089 0.467
Academic 0.177 (0.144) 0.154 0.220
Creative 0.362 (0.133) 0.350 0.007*
Global 0.587 (0.281) 0.543 0.037*
Literary-Verbal Emotional 0.202 (0.209) 0.155 0.333
Physical 0.202 (0.147) 0.172 0.168
Social 0.286 (0.212) 0.158 0.179
Future 0.502 (0.235) 0.234 0.033*
Academic 0.362 (0.147) 0.270 0.014*
Creative 0.284 (0.124) 0.234 0.022*
Global 0.032 (0.223) 0.025 0.885
Graphic Emotional 0.050 (0.213) 0.036 0.815
Physical 0.098 (0.149) 0.078 0.514
Social 0.175 (0.217) 0.092 0.419
Future 0.291 (0.238) 0.129 0.221
Academic 0.013 (0.148) 0.009 0.931
Creative 0.336 (0.128) 0.262 0.008*
Global 0.057 (0.227) 0.043 0.802
Note.
a
Unstandardized structural regression estimate (Standard Error of estimate);
b
Standardized structural
regression estimate; *p<.05 (N=170).
9
Journal of Creative Behavior
which accounts for some of their commonality (Barbot, Besanc
ßon, et al., 2016). While the music composi-
tion task (MET) was also designed to mainly capture this convergent-integrative thinking process cluster
(Barbot & Lubart, 2012), it is possible that its more “open” format—(students are free to compose each
musical segments) representing a less restrictive and structured task than the two other tasks (stronger
constraints on the basic material to compose a production),—increased the observed domain-specificity of
performance in that task.
SELF-ESTEEM AND CREATIVITY
Results regarding the relationship between creativity and self-esteem stand apart from most studies
within this line of investigation (e.g., Deng & Zhang, 2011; Wang & Wang, 2016). One of the main reasons
for such a difference is the unique design of the present study which incorporated multiple domains of cre-
ativity and self-esteem. Specifically, this study revealed a rather large contribution of domains of self-esteem
for musical creativity as measured with the MET (over 30% of the variance explained). The largest contribu-
tor was the global self-esteem dimension, which was uniquely related to musical creativity with sizable effect
(standardized c=.54, p<.05). Consistent with theoretical expectations on the relationship between musical
creativity and self-esteem (VanderArk, 1989) and based on the general hypothesis guiding this work, global
self-esteem would thus be considered more “domain-relevant” for musical composition than it is for the
other domains accounted for. Considering the various demands of the tasks, the musical composition task
surely stands apart from the others not only by its more open format outlined above, but also by the condi-
tion of task administration. The MET was indeed the only one that was administered on a one-on-one basis,
while involving video-recording, a context which may heighten the self-confidence needed to perform a cre-
ative production without self-inhibition (Barbot & Lubart, 2012). In other words, the contribution of gen-
eral self-esteem in creative performance in music may have been somewhat overemphasized by the
situational context of production.
While pointing to important directions for future research, such as the need of more standardized assess-
ment contexts across domains, this result confirms that domain-specific “demands” in a creativity task,
including the situational context of production, may account for the differential contribution of self-esteem
to creative performance. As suggested by Wang and Wang (2016), self-esteem could indeed be even more
engaged in tasks associated with high evaluation stress, which has shown to inhibit creative performance
(e.g., Silvia & Phillips, 2004). In the present study, it is very possible that the musical composition task
induced high evaluation stress, which could only be overcome with high self-esteem, while impeding perfor-
mance in the case of low self-esteem.
In the story-writing task, two domains of self-esteem were distinctly and uniquely related to creative per-
formance: academic and future self-esteem. Here too, the “domain-relevance” hypothesis provides solid
ground to interpret these observed associations. Indeed, the story-writing task was perhaps the most
“scholastic-like” task of the three creative tasks investigated because the domain of expression and format
resemble typical writing production tasks that adolescents routinely experience in school settings. As such,
the (modest) contribution of academic self-esteem appears predictably domain-relevant. Perhaps less
expected, was its association with future self-esteem. It is possible that the fictional component of creative
story-writing is somewhat domain-relevant with future self-esteem, reflecting one’s (positive) projection of
self in the future, that adolescent may build-upon in creating narrative fictions (e.g., Dollinger & Clancy,
1993).
Finally, none of the domain-specific self-esteem was uniquely related to creativity in the drawing task,
which remains in line with the domain-relevance idea (no self-esteem scale appeared conceptually aligned
with the graphic domain). However, graphic creativity was predicted by creative self-esteem with a magni-
tude comparable to what was observed for the other creativity domains (i.e., on the .30 range). This result
suggests a possible “domain-general” contribution of creative self-esteem across domains, confirmed in fur-
ther analysis focusing on the “c-factor” of creativity. This later analysis evidenced a moderate contribution
(standardized c=.43, p<.01) of creative self-esteem to the domain-general aspects of creativity (no other
domain-specific facet of self-esteem was significantly related to the c-factor). To fully understand this result,
it is important to note that the creative self-esteem scale incorporated in this study was somewhat “domain-
general” (covering multiple artistic outlets), and that an investigation of more specific facets of this con-
struct (i.e., domain-specific creative self-esteem) may have produced “domain-relevant” patterns of associa-
tions with creative performance across domains (e.g., musical creative self-esteem may be more related to
musical creativity than is creative drawing self-esteem).
10
Creativity and Self-Esteem
LIMITATIONS AND FUTURE DIRECTIONS
Despite a limited sample size inherent to this type of investigation (yet leading to a rather large number
of 329 creative productions generated by the same participants across three domains), the design of this
study inspired from the classic 3-forms planned missingness design (Graham, 2009) has proved efficient to
obtain unbiased inter-variable relationship estimates, while limiting participants and judges burden (with
only a fraction of the total participant sample completing all three creativity tasks). Further, the Multi-Infor-
mant latent-Consensus (MILC) approach (Barbot, Bick, et al., 2016) to CAT data employed here, was par-
ticularly well suited to accommodate missing data while controlling for raters’ subjectivity and unreliability.
Hence, future studies involving more than two domains of activity, which are much needed in the field,
could be facilitated by such planned missingness designs of both tasks administered and production rated,
coupled with the MILC approach (see also F€
urst, 2018). However, more research on the use of the MILC
approach is warranted: the decision to partition-out the variance related to raters “subjectivity” to isolate
the latent consensus only has important conceptual bearing with respect to CAT-based assessment of cre-
ativity (e.g., raters’ unshared subjectivity might be an important component of the “creative value” of a
given creative product).
Further, the differences in inter-rater agreement observed according to the domain of creativity activity
(reflecting the differential contribution of unshared subjectivity and unreliability in creativity ratings), with a
lower consensus observed for the music productions, point to the more classic debate on “who” is an
appropriate judge to evaluate creative production in distinct content domains (Kaufman & Baer, 2012;
Kaufman et al., 2013). Decisions in that respect have important bearings on the kind of relationships
observed between creative productions across domains (Barbot, Tan, Randi, Santa-Donato, & Grigorenko,
2012), and naturally, on the relationships observed between creativity and self-esteem reported here. The
structure of this study has great potential for further explorations of the contribution of raters’ domain-spe-
cific expertise in relation to the domain generality-specificity of creativity issue. Extensions of this work
could, for example, evaluate the reproducibility of these results, when involving groups of domain-relevant
experts, such as music composers, story-writers, and visual artists.
Finally, although the larger contribution of general self-esteem for musical creativity seems to be consis-
tent with theoretical expectations (VanderArk, 1989) and supports the importance of domain-specific “de-
mands” (Wang & Wang, 2016), future studies should attempt to measure creativity in multiple domains
using administration conditions and task structure as parallel as possible. Alternatively, this line of investiga-
tion could benefit from studies of domain-specific self-esteem (including a more detailed assessment of cre-
ative self-esteem subdomains) in relation to domain-specific achievements in creative endeavors (as opposed
to potential). Such investigation would eliminate any uncontrolled effects of creativity tasks-induced evalua-
tion stress that may account for some of the effects reported here.
CONCLUSION
This article was introduced by outlining the common beliefs surrounding the relationship between cre-
ativity and the self in both popular culture and academia. Both lines have, to date, relied on thin and incon-
sistent empirical support, notably because they have greatly ignored the domain-specificity of both creativity
and self-esteem. The study presented here suggests that while (general) creative self-esteem is mostly related
to domain-general aspects of creativity, the relationship between other domains of self-esteem (including
global self-worth) and creativity is largely “domain-relevant”. That is, relationships exist according to the
unique demands of a given creative outlet, including both situational demands (some domains may require
more self-confidence than others), and the nature of the creative work itself (e.g., the ability to positively
project oneself in the future, may be related to writing creative fiction, but not to compose a drawing). Sev-
eral directions have been suggested to pursue this important line of work for both theory and practice. Con-
sidering practical implications, an additional important line of work for future investigations is to decipher
the directionality of the relationship between domain-specific self-esteem and creativity. While model speci-
fication and general results interpretation in this study have emphasized the extent to which self-esteem “ac-
counts for” individual differences in domain-specific creativity, it is important to keep in mind that
reciprocal relationships exist. Indeed, understanding the extent to which creativity contributes to self-esteem
has important implications for the development of (domain-specific) creativity-based interventions designed
to bolster adolescent’s self-esteem in targeted domains.
11
Journal of Creative Behavior
REFERENCES
Baer, J., & Kaufman, J.C. (2005). Bridging generality and specificity: The amusement park theoretical (APT) model of creativity.
Roeper Review,27, 158–163.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman and Co.
Barbot, B., Besanc
ßon, M., & Lubart, T. (2015). Creative potential in educational settings: Its nature, measure, and nurture. Educa-
tion 3–13,43, 371–381.
Barbot, B., Besanc
ßon, M., & Lubart, T. (2016). The generality-specificity of creativity: Exploring the structure of creative potential
with EPoC. Learning and Individual Differences,52, 178–187.
Barbot, B., Bick, J., Bentley, M.J., Balestracci, K., Woolston, J.L., Adnopoz, J.A., & Grigorenko, E.L. (2016). Changes in mental
health outcomes with the intensive in-home child and adolescent psychiatric service: A multi-informant, latent consensus
approach. International Journal of Methods in Psychiatric Research,25,33–43.
Barbot, B., & Heuser, B. (2017). Creativity and identity formation in adolescence: A developmental perspective. In M. Karwowski
& J.C. Kaufman (Eds.), The creative self: Effect of beliefs, self-efficacy, mindset, and identity (pp. 87–98). London: Academic
Press.
Barbot, B., & Lubart, T. (2012). Creative thinking in music: Its nature and assessment through musical exploratory behaviors. Psy-
chology of Aesthetics, Creativity, and the Arts,6, 231–242. https://doi.org/10.1037/a0027307.
Barbot, B., Lubart, T.I., & Besanc
ßon, M. (2016). “Peaks, Slumps, and Bumps”: Individual differences in the development of creativ-
ity in children and adolescents. New Directions for Child and Adolescent Development,2016,33–45.
Barbot, B., Orriols, E., & Pouyade, H. (2008). Consensual assessment technique-interface (CAT-i). Copyrights Cat-i.Org.
Barbot, B., Safont-Mottay, C., & Oubrayrie-Roussel, N. (2018). Echelle Multidimensionnelle de l’Estime de Soi (EMES). In J. Col-
lange, E. Loarer & T.I. Lubart (Eds.), La psychologie diff
erentielle au 21
eme si
ecle : Nouvelles mod
elisations et applications - (pp.
77–80). Paris: Laboratoire Lati.
Barbot, B., Tan, M., Randi, J., Santa-Donato, G., & Grigorenko, E.L. (2012). Essential skills for creative writing: Integrating multiple
domain-specific perspectives. Thinking Skills and Creativity,7, 209–223. https://doi.org/10.1016/j.tsc.2012.04.006.
Barbot, B., & Tinio, P.P. (2015). Where is the “g” in creativity? A specialization–differentiation hypothesis. Frontiers in Human
Neuroscience,8, 1041.
Barbot, B., & Webster, P.R. (2018). Creative thinking in music. In T.I. Lubart (Ed.), The creative process: Perspectives from multiple
domains. Basingstoke, UK: Palgrave Macmillan.
Beghetto, R.A. (2006). Creative self-efficacy: Correlates in middle and secondary students. Creativity Research Journal,18, 447–457.
https://doi.org/10.1207/s15326934crj1804_4.
Beghetto, R.A., & Dilley, A.E. (2016). Creative aspirations or pipe dreams? Toward understanding creative mortificatio n in children
and adolescents. New Directions for Child and Adolescent Development,2016,85–95.
Cantero, M.-J., Alfonso-Benlliure, V., & Melero, R. (2016). Creativity in middle childhood: Influence of perceived maternal sensitiv-
ity, self-esteem, and shyness. Creativity Research Journal,28,105–113. https://doi.org/10.1080/10400419.2016.1125246.
Coopersmith, S. (1967). The antecedents of self-esteem. San Fransisco: Freeman.
Deng, X.-P., & Zhang, X.-K. (2011). Understanding the relationship between self-esteem and creativity: A meta-analysis. Advances
in Psychological Science,19, 645–651.
Dollinger, S.J., & Clancy, S.M. (1993). Identity, self, and personality: II. Glimpses through the autophotographic eye. Journal of Per-
sonality and Social Psychology,64, 1064–1071.
Dollinger, S.J., Clancy Dollinger, S.M., & Centeno, L. (2005). Identity and creativity. Identity,5,315–339.
F€
urst, G. (2018). Measuring creativity with planned missing data. The Journal of Creative Behavior. https://doi.org/10.1002/jocb.352.
Gnambs, T., Scharl, A., & Schroeders, U. (2018). The structure of the Rosenberg self-esteem scale: A cross-cultural meta-analysis.
Zeitschrift F€
ur Psychologie,226,14–29.
Goldsmith, R.E., & Matherly, T.A. (1988). Creativity and self-esteem: A multiple operationalization validity study. The Journal of
Psychology,122,47–56.
Graham, J.W. (2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology,60, 549–576.
Harter, S. (1990). Causes, correlates, and the functional role of global self-worth: A life-span perspective. In R.J. Sternberg, & J.
Kolligan Jr (Eds.), Competence considered (pp. 67–97). New Haven, CT: Yale University Press.
Hughes, R. (1996). Modernism’s patriarch- the cezanne exhibition in phildalephia is and epic humbling event, fully worthy of its
great subject. Time Magazine,147,72–74.
Jacobs, J.E., Lanza, S., Osgood, D.W., Eccles, J.S., & Wigfield, A. (2002). Changes in children’s self-competence and values: Gender
and domain differences across grades one through twelve. Child Development,73, 509–527.
Jaquish, G.A., & Ripple, R.E. (1981). Cognitive creative abilities and self-esteem across the adult life-span. Human Development,24,
110–119.
Judge, T.A., Erez, A., Bono, J.E., & Thoresen, C.J. (2002). Are measures of self-esteem, neuroticism, locus of control, and general-
ized self-efficacy indicators of a common core construct? Journal of Personality and Social Psychology,83, 693–710.
Karwowski, M. (2016). The dynamics of creative self-concept: Changes and reciprocal relations between creative self-efficacy and
creative personal identity. Creativity Research Journal,28,99–104.
12
Creativity and Self-Esteem
Karwowski, M., & Barbot, B. (2016). Creative self-beliefs: Their nature, development, and correlates. In J.C. Kaufman & J. Baer
(Eds.), The Cambridge companion to creativity and reason in cognitive development (pp. 302–326). Cambridge, UK: Cambridge
University Press.
Karwowski, M., & Lebuda, I. (2015). The big five, the huge two, and creative self-beliefs: A meta-analysis. Psychology of Aesthetics,
Creativity, and the Arts,10,214–232.
Kaufman, J.C. (2012). Counting the muses: Development of the Kaufman Domains of Creativity Scale (K-DOCS). Psychology of
Aesthetics, Creativity, and the Arts,6, 298–308. https://doi.org/10.1037/a0029751.
Kaufman, J.C., & Baer, J. (2012). Beyond new and appropriate: Who decides what is creative? Creativity Research Journal,24,83–
91.
Kaufman, J.C., Baer, J., Cropley, D.H., Reiter-Palmon, R., & Sinnett, S. (2013). Furious activity vs. understanding: How much
expertise is needed to evaluate creative work? Psychology of Aesthetics, Creativity, and the Arts,7, 332–340.
Kaufman, J.C., & Beghetto, R.A. (2009). Beyond big and little: The four c model of creativity. Review of General Psychology,13,1–
12. https://doi.org/10.1037/a0013688.
Kemple, K.M., David, G.M., & Wang, Y. (1996). Preschoolers’ creativity, shyness, and self-esteem. Creativity Research Journal,9,
317–326. https://doi.org/10.1207/s15326934crj0904_3.
Kline, R.B. (2015). Principles and practice of structural equation modeling. New York, NY: Guilford Publications.
Lau, S., Cheung, P.C., Lubart, T., Tong, T.M., & Chu, D.H. (2013). Bicultural effects on the creative potential of Chinese and
French children. Creativity Research Journal,25, 109–118.
Lau, S., Li, C.S., & Chu, D.H. (2004). Perceived creativity: Its relationship to social status and self-concept among Chinese high
ability children. Creativity Research Journal,16,59–67.
Lebuda, I., & Csikszentmihalyi, M. (2017). Me, myself, I, and creativity: Self-concepts of eminent creators. In M. Karwowsk i &
J.C. Kaufman (Eds.), The creative self: Effect of beliefs, self-efficacy, mindset, and identity (pp. 137–152). London: Academic
Press.
Lubart, T.I., Besanc
ßon, M., & Barbot, B. (2011). EPoC:
Evaluation du potentiel cr
eatif (EPoC). Paris, France: Editions Hogrefe
France.
Marsh, H.W., Byrne, B.M., & Shavelson, R.J. (1988). A multifaceted academic self-concept: Its hierarc hical structure and its relation
to academic achievement. Journal of Educational Psychology,80, 366.
Marsh, H.W., Craven, R., & Debus, R. (1998). Structure, stability, and development of young children’s self-concepts: A multico-
hort–multioccasion study. Child Development,69, 1030–1053.
Petkus, E. (1996). The creative identity: Creative behavior from the symbolic interactionist perspective. The Journal of Creative
Behavior,30, 188–196.
Plucker, J.A., & Beghetto, R.A. (2004). Why creativity is domain general, why it looks domain specific, and why the distinction does
not matter. In R.J. Sternberg, E.L. Grigorenko & J.L. Singer (Eds.), Creativity: From potential to realization (pp. 153–167).
Washington, DC: American Psychological Association.
Pretz, J.E., & Nelson, D.N. (2017). Creativity is influenced by domain, creative self-efficacy, mindset, self-efficacy, and self-esteem.
In M. Karwowski & J.C. Kaufman (Eds.), The creative self: Effect of beliefs, self-efficacy, mindset, and identity (pp. 155–170).
London: Academic Press.
Puryear, J.S., Kettler, T., & Rinn, A.N. (2017). Relationships of personality to differential conceptions of creativity: A systematic
review. Psychology of Aesthetics, Creativity, and the Arts,11,59–68.
Rank, J., Nelson, N.E., Allen, T.D., & Xu, X. (2009). Leadership predictors of innovation and task performance: Subordinates’ self-
esteem and self-presentation as moderators. Journal of Occupational and Organizational Psychology,82, 465–489. https://doi.
org/10.1348/096317908X371547.
Robins, R.W., Trzesniewski, K.H., Tracy, J.L., Gosling, S.D., & Potter, J. (2002). Global self-esteem across the life span. Psychology
and Aging,17, 423–434.
Rothenberg, A. (1990). Creativity in adolescence. Psychiatric Clinics of North America,13, 415–434.
Royston, R., & Reiter-Palmon, R. (2017). Creative self-efficacy as mediator between creative mindsets and creative problem-solving.
The Journal of Creative Behavior. https://doi.org/10.1002/jocb.226.
Shavelson, R.J., Hubner, J.J., & Stanton, G.C. (1976). Self-concept: Validation of construct interpretations. Review of Educational
Research,46, 407–441.
Silvia, P.J., & Phillips, A.G. (2004). Self-awareness, self-evaluation, and creativity. Personality and Social Psychology Bulletin,30,
1009–1017.
Vallieres, E.F., & Vallerand, R.J. (1990). Traduction et validation Canadienne-Franc
ßaise de l’Echelle de l’Estime de Soi de Rosen-
berg. International Journal of Psychology,25, 305–316. https://doi.org/10.1080/00207599008247865.
VanderArk, S. (1989). Self-esteem, creativity, and music: Implications and directions for research. Suncoast Music Education Forum
on Creativity,1989,1–19.
Vispoel, W.P. (1995). Self-concept in artistic domains: An extension of the Shavelson, Hubner, and Stanton (1976) model. Journal
of Educational Psychology,87, 134–153.
Wang, Y., & Wang, L. (2016). Self-construal and creativity: The moderator effect of self-esteem. Personality and Individual Differ-
ences,99, 184–189. https://doi.org/10.1016/j.paid.2016.04.086.
13
Journal of Creative Behavior
Wild, L.G., Flisher, A.J., & Lombard, C. (2004). Suicidal ideation and attempts in adolescents: Associations with depression and six
domains of self-esteem. Journal of Adolescence,27, 611–624.
Baptiste Barbot, Pace University, Yale University
Correspondence concerning this article should be addressed to Baptiste Barbot, Department of Psychology, Pace University, 41
Park Row, New York, NY 10038. E-mail: BBarbot@pace.edu.
ACKNOWLEDGEMENT
I am thankful to Alana Miller, Kristen Piering, Henry Eff, Nils Myszkoswki, and James Kaufman for help with vari-
ous aspects of this project and manuscript preparation, and to all participating school staffs and students.
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the
article.
Appendix S1. Barbot—ESM—FIML estimated pairwise correlations between all variables in the study
14
Creativity and Self-Esteem
