Narcissism, bullying, and social dominance in youth a longit

Narcissism, Bullying, and Social Dominance in Youth:
A Longitudinal Analysis
Albert Reijntjes & Marjolijn Vermande &
Sander Thomaes & Frits Goossens & Tjeert Olthof &
Liesbeth Aleva & Matty Van der Meulen
Published online: 3 February 2015
# The Author(s) 2015. This article is published with open access
at Springerlink.com
Abstract A few previous studies have shown that narcissistic
traits in youth are positively associated with bullying. However,
research examining the developmental relationship between
narcissism and bullying is lacking. Moreover, it is unclear
whether narcissists constitute a homogeneous group and
whether the bullying of narcissistic youth results in establishing
social dominance over peers. The present work addresses these
gaps. Children (N=393; Mage=10.3; 51 % girls) were followed
during the last 3 years of primary school. Person-centered anal-
yses were used to examine whether groups with distinct devel -
opmental trajectories for narcissism and two bullying forms
(direct and indirect) can be identified, and how these
trajectories
are related. Multiple groups emerged for all constructs exam-
ined. For girls, higher narcissism was neither related to more
intense bullying, nor to higher social dominance. In contrast,
highly narcissistic boys were more likely than their peers to
show elevated direct bullying, and in particular elevated
indirect bullying. Hence, high narcissism is a risk factor for
bullying in boys, but not in girls. However, narcissism is not

always accompanied by high bullying, given that many boys on
the high bullying trajectories were not high in narcissism.
Results show that among narcissistic youth only those who
engage in high levels of bullying are high in social dominance.
Keywords Narcissism . Bullying . Social dominance . Joint
trajectory analysis . Gender differences
During the past decades, researchers have increasingly ac-
knowledged that bullying is a strategic attempt to acquire a
central, powerful and dominant position in the peer group
(e.g., Olthof et al. 2011; Salmivalli and Peets 2009). For in-
stance, Farrington (1993) observed that when asked Bwhy do
you bully?^, the most frequently reported answers are Bto feel
powerful^ and Bto look cool^. Moreover, in early adolescence
bullies score significantly higher on status, power, and pres-
tige goals than do their peers (Sijtsema et al. 2009).
The motivation of bullies to gain power, dominance, and
prestige over others suggests that elevated narcissism might be
a contributing factor. Narcissism is a dispositional trait that
involves a sense of entitlement of privileged status over
others, the belief that one is unique and more important than
others, and an excessive need for approval and admiration
from others to feed the grandiose - but ultimately vulnerable
- self (Miller et al. 2007; Morf and Rhodewalt 2001). Similar
to adults, youth with narcissistic traits often display rather
aversive interpersonal behavior, such as arrogance, lack of
empathy, exploitativeness and aggression (Morf and
Rhodewalt 2001; Thomaes and Brummelman 2015).
According to the self-regulatory model of Morf and
Rhodewalt (2001), narcissistic individuals use several
Electronic supplementary material The online version of this
article

(doi:10.1007/s10802-015-9974-1) contains supplementary
material,
which is available to authorized users.
A. Reijntjes (*): M. Vermande
Department of Pedagogical and Educational Sciences,
Utrecht University, PO Box 80150, Utrecht, The Netherlands
e-mail: [email protected]
S. Thomaes
University of Southampton, Southampton, UK
F. Goossens: T. Olthof
Free University of Amsterdam, Amsterdam, The Netherlands
L. Aleva
Department of Developmental Psychological,
Utrecht University, Utrecht, The Netherlands
M. Van der Meulen
Groningen University, Groningen, The Netherlands
J Abnorm Child Psychol (2016) 44:63–74
DOI 10.1007/s10802-015-9974-1
http://dx.doi.org/10.1007/s10802-015-9974-1
techniques to maintain or enhance their inflated self-esteem.
For instance, when faced with criticism, they often react aggres-
sively in an attempt to restore their self-esteem. With regard to
bullying, Salmivalli (2001) posited that the exploitativeness
and lack of empathy that figure prominently in narcissists
may lead to aggression being employed instrumentally to foster
their grandiose self-views.
During the past decade, a few studies have examined the link

between narcissistic features and bullying in youth. Taken
togeth-
er, findings support Salmivalli’s (2001) hypothesis. For
instance,
in an inpatient sample of youth aged 10–15, Stellwagen and
Kerig
(2013) found that psychopathy-linked narcissism (i.e., the
grandi-
ose self-perceptions and sense of entitlement characteristic of
youth with psychopathic traits) was concurrently positively
linked
with scores for (ringleader) bullying. Similarly, Ang and col-
leagues (Ang et al. 2010) showed that narcissistic
exploitativeness
in Asian youth was concurrently positively associated, albeit
weakly, with bullying. Moreover, longitudinal work among
Greek-Cypriot adolescents aged 12–14 has demonstrated that
bullying was higher and more stable among those scoring higher
on narcissism at baseline (Fanti and Kimonis 2012). In a recent
prospective study,Fanti and Henrich (2015) found that
narcissistic
children with low general self-esteem are in particular likely to
bully.
Notwithstanding the merits of these studies, important re-
search gaps remain. First, except for the study of Fanti and
colleagues, there is a paucity of longitudinal research examining
the link between narcissism and bullying. Consequently, also
because Fanti and coworkers only assessed narcissism once (at
baseline), the stability of the core constructs over time is
largely
unknown. Moreover, the dynamic, longitudinal relationship be-
tween narcissism and bullying remains to be investigated.
Second, previous work has almost exclusively employed a
variable-centered approach. A significant drawback of this

approach is that participants are treated as one homogeneous
group in terms of how the predictors operate on the outcomes
(Laursen and Hoff 2006). Importantly, in the case of distinct
subgroups of bullies or narcissists (e.g., when a summary sta-
tistic such as a correlation does not equally apply to all partic -
ipants), mean-level parameters may not describe any sub-
group validly (Von Eye and Bogat 2006), and they are often
least applicable to children with the most extreme scores.
Moreover, when the potential heterogeneity of narcissism
and bullying in this respect is taken into account, interventions
can be fine-tuned for specific groups of children. At his point,
it should be noted that several studies have shown that differ -
ent trajectories of bullying behavior exist (e.g., high and me-
dium) that differ in their functioning and development (e.g.,
Pepler et al. 2008; Reijntjes et al. 2013b). Although Fanti and
Henrich (2015) distinguished between bullies and Bunin-
volved^ children, they did not distinguish between potentially
different bullying trajectory subgroups, and narcissism was
treated as a continuous variable.
Third, studies examining the link between narcissism and
bullying have used instruments that do not tap all facets of the
narcissism construct. Specifically, the Antisocial Process
Screening Device (APSD), employed by Stellwagen and Kerig
(2013) as well as Fanti and Kimonis (2012), assesses
narcissistic
behavior that tends to co-occur with psychopathic traits, but not
narcissistic cognitions and feelings (e.g., feelings of
entitlement)
that are at the core of the narcissism construct. For this reas on,
researchers using the APSD typically refer to the measured con-
struct as Bpsychopathy-linked^ narcissism (for differences be-
tween this construct and narcissism, see Thomaes and
Brummelman 2015). In a similar vein, Ang et al. (2010) only
used the BExploitativeness^ subscale of the Narcissistic
Personality Questionnaire for Children-Revised (NPQC-R). In

the present study, the Childhood Narcissism Scale (CNS;
Thomaes et al. 2008) was employed. This widely used, compre-
hensive and psychometrically sound measure indexes narcissism
as a general construct, and is well validated in Dutch samples of
youth.
Fourth, to the extent that narcissistic children high in bul -
lying pursue social dominance and power, no study has exam-
ined whether they are successful in this regard. As in resource
control theory (RCT; Hawley 1999), we construe social dom-
inance as competitive superiority, which is an aspect of rela-
tionship asymmetry. Social dominance is indexed by resource
control; i.e., having access to desirable, but scarce social and
material resources (Hawley 1999).
The present three-wave study that followed children from late
childhood into early adolescence addressed these limitations by
examining the relations between narcissism and bullying as they
unfold over time. In so doing, we distinguished between direct
and indirect bullying. Direct bullying pertains to behaviors in
which the victim is overtly harassed (e.g., physically, verbally),
while indirect bullying refers to behaviors that do not directly
confront the victim (e.g., gossiping). An important reason to
make this distinction is that both forms may be differentially
linked to both gender and narcissism. For instance, boys are
more
inclined to engage in direct forms of aggression than girls,
where-
as both genders are about equally likely to display indirect
forms
of aggression (Card et al. 2008). Moreover, it has been argued
that for narcissistic youth the use of indirect, relational
aggression
may be more effective than direct aggression in terms of acquir-
ing and preserving a dominant position in the peer group
(Golmaryami and Barry 2010). We therefore wanted to examine

whether narcissistic youth differ in the extent to which they
enact
both forms of bullying. We studied children in this developmen-
tal period because during preadolescence the formation of posi -
tive peer relations is an essential developmental task (Hartup
1996), and concerns about social status figure prominently
(Fossati et al. 2010; Salmivalli 2001).
To capture potential subgroup differences in the strength
and form of the association between the constructs examined,
person-centered analyses were employed (Nagin 2005).
64 J Abnorm Child Psychol (2016) 44:63–74
Specifically, joint trajectory analyses were used to examine
whether groups with distinct developmental trajectories of
narcissism and bullying can be identified, and how these tra-
jectories are related. This person-centered approach relates the
longitudinal course of two constructs of interest (Nagin and
Tremblay 2001), thereby comprehensively investigating their
dynamic co-occurrence over time. We also examined the out-
comes of bullying and narcissism in terms of social domi -
nance. Specifically, after identifying joint trajectories of bul -
lying and narcissism, we compared the joint trajectory groups
on their resource control scores. For the reasons outlined
above, as well as research showing that the link between nar -
cissism and aggression may differ between boys and girls
(e.g., Pauletti et al. 2012), we also examined potential gender
differences.
We expected to find at least two different developmental
trajectories for both narcissism and the two types of bullying,
including a high and a low trajectory. We also expected that
only a relatively small group of participants, predominantly

boys, would engage in consistently high levels of bullying.
Similarly, given that the narcissistic traits in youth in the gen-
eral population are normally distributed, although somewhat
positively skewed (Thomaes and Brummelman 2015), we al-
so expected that relatively few children would be consistently
high in narcissism. Moreover, we hypothesized that children
following the highest narcissism trajectory would be more
likely than their peers to simultaneously follow a higher bul -
lying trajectory. Finally, we expected that social dominance
would be highest for children displaying both high bullying
and high narcissism.
Method
Participants
Participants were 393 children (51 % girls) from 12
elementary schools throughout the Netherlands. The
children were followed during their last 3 years at ele-
mentary school. At the start of the study in 2006 (T1),
all participants were in fourth grade (Mage=10.3; SD=
0.5). There was no school transition during this period,
and almost all children remained in the same classroom
with the same peers. Participation rates within classroom
were very similar across years. Although SES was not
formally assessed, the sample included pupils from a
wide range of social backgrounds. Parents received a
letter in which they were informed about the purpose
of the study. They could either provide passive consent
for their child’s participation by not communicating fur-
ther with the researchers (96 %), or refuse by returning
a preprinted objection form (4 %). Parents and children
could withdraw from the study at any time. All children
provided their own assent. We also obtained IRB ap-
proval and permission from the schools. The large ma-

jority (83 %) of the children was Caucasian (native
Dutch). Other groups represented were pupils with at
least one parent originating from Turkey, Morocco,
Surinam, or another European country.
At T2, data were available for 371 participants (94.4 % of
the original sample). At T3, the sample contained 336 partic-
ipants (85.5 %). Attrition was mainly due to participants mov-
ing to other, non-participating schools. Children not par-
ticipating at T2 and/or T3 did not differ from children
with complete data in their scores on relevant variables
at T1 (p’s>0.10).
Procedure
Children’s self-reported narcissism scores were obtained dur-
ing a classroom session run by trained research assistants.
Teachers rated children’s resource control at their own conve-
nience. The peer nominations were collected during an inter -
view by a research assistant in a quiet room at the school
grounds. Children could discontinue their participation at
any time, but no child did. To minimize interviewer effects,
research assistants were extensively trained, written research
protocols were employed, and standardized interviews were
laptop administered.
When providing peer nominations for bullying, chil-
dren used a list containing the names of their class-
mates. The number of nominations was unlimited.
Research on peer sociometric status has shown that,
relative to the limited nominations approach, the unlim-
ited nominations procedure yields a more reliable and
valid assessment (Terry 2000). We consider it likely that
this will also be the case for bullying.
Children could only nominate children from their own

classroom, and not themselves. Mixed-sex nominations were
used and nominations were conducted within classrooms rath-
er than within grade. Although in early adolescence gender
segregation is salient, in their study examining peer sociomet-
ric nominations (i.e., Blike most^ and Blike least^) in sixth
graders, Poulin and Dishion (2008) observed that including
nominations from other-sex classmates improved the predic-
tive validity of the sociometric measure. Moreover, for chil -
dren confined to a stable classroom in which they mainly
interact with their classmates and not much with other
grademates (which is the case in the Netherlands), restricting
the voting population to the classroom peers did not affect the
predictive validity of the measure.
Measures
Narcissism The Childhood Narcissism Scale (CNS; Thomaes
et al. 2008) is a 10-item self-report measure that indexes trait
J Abnorm Child Psychol (2016) 44:63–74 65
narcissism in youth aged 8 and older. The CNS has a one-
factor structure and was developed to measure narcissism as a
general construct, without distinguishing between more nar-
rowly defined dimensions or facets such as adaptive versus
maladaptive narcissism. Previous research (Thomaes et al.
2008) has shown that CNS scores have both adaptive (agentic
interpersonal orientation) and maladaptive correlates
(exploitativeness). Using a large sample (N=1020), a single-
factor model was tested in MPlus using confirmatory factor
analysis (Thomaes et al. 2008). All factor loadings were freely
estimated and no residual correlations between items were
allowed. Several measures of model fit indicated that a
single-factor model provided a good fit to the data.

Specifically, RMSEA = 0.05; CFI = 0.95; NFI =0.94.
Standardized factor loadings ranged from 0.47 to 0.64 and
all were significant. The internal consistency and the test-
retest stability of the instrument are good (see Thomaes et al.
2008).
Sample items are BI am a great example for other kids to
follow,^ and BI love showing all the things I can do.^ Items are
rated on a scale ranging from 0 (not at all true) to 3 (complete-
ly true). In the present study, sum scores were used. Research
shows that childhood narcissism has largely similar correlates
and outcomes as adult narcissism. For instance, CNS scores
are positively associated with self-appraised superiority, but
only weakly with self-esteem (see Thomaes et al. 2008;
Thomaes and Brummelman 2015). Moreover, attesting to
construct validity, scores on the CNS are positively associated
with self-esteem contingency, agentic but not communal so-
cial goals, psychopathic and Machiavellian personality traits,
aggression, and negatively with empathy (Thomaes et al.
2008). Cronbach’s alpha was adequate at all assessment points
(>0.75 for both genders).
Bullying The bullying role nomination procedure (BRNP;
Olthof et al. 2011) was used. The BRNP is a comprehensive
measure that indexes both direct forms of bullying (e.g., hit-
ting, damaging belongings of others, calling names) and indi -
rect forms of bullying (e.g., gossiping, spreading rumors,
stirring others up to exclude someone). Previous work
(Olthof et al. 2011; Reijntjes et al. 2013a, b) has shown
that bullying as indexed by the BRNP is associated in
theoretically meaningful ways with peer-nominated per-
ceived popularity, peer-rated likeability, and self-
perceived social competence.
To avoid potential interpretation differences of the term
bullying, children first received an elaborate description of

the concept, in which its three core features were explained:
intent to harm, repetition over time, and a patent power differ -
ence between perpetrator and victim (Salmivalli and Peets
2009). Children were told that bullying can occur in a number
of ways and differs from a quarrel or fight between two equal-
ly strong peers.
Subsequently, five specific types of bullying were de-
scribed (i.e., physical, possession-related, verbal, direct rela-
tional, and indirect relational) and nominations were obtained
by asking BDo you know which classmates carry out that
particular form of bullying?^, and BIf so, could you give us
their names?^ Continuous scores for both direct and indirect
bullying were computed within classrooms by dividing
the total number of nominations by the number of nom-
inators minus 1 (the participant himself; see Goossens
et al. 2006; Kärnä et al. 2011). Scores were then stan-
dardized within classes to take differences between
nominating groups into account. A total bullying score
was also calculated by summing the scores for the tw o
forms of bullying. Analyses were performed for the two
bullying forms separately and for the total bullying
score.
Resource Control This construct indexes social dominance
and involves having access to scarce, desirable material and
social resources (Hawley 1999). Teachers rated participants
on six items, on a scale ranging from 0 (never or almost never)
to 4 (very often). Sample items are: BTo what extent is this
child usually at the center of attention in a group of
children^?; and BTo what extent does this child usually get
what it wants?^ The ratings were averaged. The scale showed
high internal consistency (Cronbach’s alpha > 0.90 at all three
assessment points). To take differences between teachers into
account, the scores were standardized within each class.

Plan of Analyses
We first present descriptive data and correlational analyses.
Next, the person-centered analyses proceeded in three steps.
First, the group-based trajectory approach (Nagin 2005) was
used to estimate separate models for the developmental trajec-
tories of narcissism, the two forms of bullying, and total bul -
lying. Using MPlus version 6.11 (Muthén and Muthén 2010),
latent class growth analyses models (LCGA’s; Muthén and
Muthén 2000) were employed. Missing data for participants
who did not complete an entire measure (as opposed to indi -
vidual items) and for those who did not complete one or two
complete assessments were handled through full information
maximum likelihood (FIML).
Latent class growth analysis uses an outcome variable
measured at multiple time points to define a latent class
model in which the latent classes correspond to different
growth curves for that variable, thereby yielding clusters
of individuals who follow distinct developmental trajec-
tories. In the case of three assessment points, these tra-
jectories are identified on the basis of two parameters;
i.e., intercepts (starting values) and linear slopes. The
proportion of individuals following each of the trajecto-
ries is estimated.
66 J Abnorm Child Psychol (2016) 44:63–74
For narcissism, the two bullying forms, and total bullying a
series of models was fitted, starting with a one-trajectory mod-
el and moving to a five-trajectory model. To make a well-
founded decision regarding the optimal number of groups,
several statistical indicators were used (as recommended by
Nagin 2005), including the Bayesian Information Criterion

(BIC), the Lo-Mendell-Rubin likelihood ratio test (LMR-
LRT), and entropy.
Following Nagin (2005), we also examined whether
for all groups the average posterior probability (AvePPj)
exceeded 0.70. This signifies that, on average, the
chance that children assigned to a particular trajectory
group actually belong to that group is at least 0.70.
Second, we examined whether the odds of correct clas-
sification (OCCj) were at least 5 for all groups. Finally,
we compared the model estimated proportion of the
population following a particular trajectory group (πj)
with the corresponding proportion of the sample
assigned to that trajectory (Pj), with less discrepancy
indicating better model fit.
After determining the best fitting trajectory models for the
targeted variables separately, in the second step the joint tra -
jectories for (a) narcissism and (b) each of the two bullying
forms, as well as the total bullying score were estimated. Key
outputs of a joint model are joint probabilities and two sets of
conditional probabilities. Joint probabilities pertain to
the proportion of children estimated to belong simulta-
neously to certain trajectory groups of both variable A
and variable B (e.g., children who follow both the high
narcissism trajectory as well as the high physical bully-
ing trajectory). When j and k index the trajectory groups
associated with bullying and narcissism, the joint prob-
abilities are denoted by π jk and are provided as part of
the output. Conditional probabilities pertain to the esti -
mated probability of belonging to a specific trajectory
group for variable A (e.g., high direct bullying) given
membership of a specific trajectory group for variable B
(e.g., high narcissism), and vice versa. These probabili -

ties are denoted by π j|kand πk|j and are calculated as follows:
π j kj ¼ π jk
πk
; with πk ¼ ∑ jπ jk; k ¼ 1; …; K and πk jj ¼ π
jk
π j ;
with π j ¼ ∑ jπ jk; j ¼ 1; …; J:
Importantly, conditional probabilities do not imply a time
order relationship but reflect the probability of simultaneously
following two trajectories during the same period. To evaluate
differences between observed probabilities, we used a
Bayesian model selection approach with (in)equality con-
straints between the parameters of interest (Klugkist et al.
2005). The results of the Bayesian approach are expressed in
terms of posterior model probabilities (PMP’s), representing
the probability that the specific model at hand receives most
support from the data among a set of models (e.g., Model 1:
probability A is larger than probability B, versus Model 2:
probability A is equal to probability B). A model was consid-
ered to outperform another model when its PMP was at least
0.95 (Klugkist et al. 2005).
Finally, we examined how social dominance scores dif-
fered as a function of the joint effects of bullying and narcis -
sism. Specifically, after identifying the joint trajectories of
narcissism and total bullying score, we compared these joint
groups on their mean resource control scores across the three
waves. We also aimed to investigate how the joint trajectories
of narcissism and the two different bullying forms separately
are related to social dominance scores. However, examining

how narcissism and each of the two bullying forms separately
contribute to social dominance proved problematic.
Specifically, when examining social dominance as a function
of narcissism and direct (indirect) bullying, one should control
for the effect of indirect (direct) bullying. Given our approach,
which yields distinct trajectory groups (latent classes) for both
types of bullying, this requires controlling for class member -
ship of indirect bullying when examining the effects of direct
bullying (and vice versa). However, whereas controlling for a
continuous covariate is possible, current software does not
allow for controlling for (the probability of) latent class
membership.
At this point, it should be noted that an additional joint
trajectory analysis showed that almost all participants who
were assigned to the high (medium, low) direct bullying tra-
jectory group, simultaneously belonged to the high (medium,
low) indirect bullying trajectory group. This finding indicates
that an observation of high (low) indirect bullying is almost
synonymous with an observation of high (low) direct bully-
ing, and vice versa. Given that both bullying forms were
strongly interwoven, we decided to examine how the joint
trajectories of narcissism and both bullying behaviors com-
bined (i.e., total bullying) relate to social dominance (resource
control) scores.
Results
Preliminary Analyses
Table 1 displays the means and standard deviations for all
variables at the three assessment points. Across waves, boys
scored higher on narcissism than did girls (p’s<0.05), al -
though the differences were small (Cohen d’s<0.25). No gen-
der differences were observed for resource control (p’s>0.20).
For both forms of bullying, boys consistently scored higher

than girls, except for indirect bullying at T3. Gender differ -
ences were largest for direct bullying. Across gender, the two
forms of bullying were substantially correlated at all three
assessment points (r’s>0.65; p’s<0.001; see correlation
Tables in the Electronic supplementary material).
J Abnorm Child Psychol (2016) 44:63–74 67
Repeated measures ANOVA analyses revealed substantial
stability for all constructs over time. For narcissism and re -
source control, the mean score did not change over time.
Scores for bullying were also quite stable over time. Only a
decrease in direct and total bullying was observed from T1 to
T2, and this change only applied to boys.
At all three time points, the linkage between narcissism and
resource control was low (r’s<0.13), although significant at
T1 and T3 (see Electronic supplementary material). Similarly,
at all three time points scores for narcissism and each of the
two bullying forms and total bullying were only weakly relat-
ed (r’s<0.18), albeit significantly in several instances. In con-
trast, across time substantial linkages were found between
resource control and both bullying types, as well as total bul -
lying (r’s ranging from 0.39 to 0.49; p’s<0.01).
LCGA Analyses
Separate trajectory analyses were performed for narcissism,
direct bullying, indirect bullying, and the total bullying score.
Participants were assigned to the trajectory group for which
they showed the highest posterior probability.
Narcissism The statistical indicators provided most support
for a three-group model. Specifically, when moving from a

two-group to a three-group model, entropy increased from
0.68 to 0.70, the LMR-statistic was significant, and the BIC
value decreased from 6230.8 to 6189.3. However, when mov-
ing to a four-group model, entropy decreased to 0.61, the
LMR- statistic was not significant, and the BIC value in-
creased to 6200.3. Importantly, the fit indices for the three-
group model were good (AvePP j’s>0.83; OCC j’s>5; differ -
ences between Pj and πj less than 2 %, entropy = 0.70).
As depicted in Figure 1 (see Electronic supplementary
material), the largest group (n=184; 46.8 % of the sample)
displayed stable medium narcissism scores (intercept (I)=
9.60, p<0.001; slope (S)=0.27, p>0.20). Children in the sec-
ond largest group (n=171; 43.5 %) showed consistently low
scores (I=5.46, p<0.001; S=−0.36, p>0.20). Children in the
third and smallest group (n=37; 9.4 %) displayed the highest
scores that were stable over time (I=15.56, p<0.001; S=−0.21;
p>0.20). Boys were overrepresented in the high and medium
trajectory groups (56.8 % and 55.4 %, …
Raising Children With High Self-Esteem (But Not
Narcissism)
Eddie Brummelman,
1
and Constantine Sedikides
2
1University of Amsterdam and 2University of Southampton
ABSTRACT—With the rise of individualism since the 1960s,
Western parents have become increasingly concerned with

raising their children’s self-esteem. This is understandable,
given the benefits of self-esteem for children’s psychologi-
cal health. However, parents’ well-intentioned attempts to
raise self-esteem, such as through inflated praise, may
inadvertently breed narcissism. How can parents raise
self-esteem without breeding narcissism? In this article,
we propose a tripartite model of self-regard, which holds
that the development of self-esteem without narcissism
can be cultivated through realistic feedback (rather than
inflated praise), a focus on growth (rather than on outper-
forming others), and unconditional regard (rather than
regard that is conditional). We review evidence in support
of these practices and outline promising directions for
research. Our model integrates existing research, stimu-
lates the development of theory, and identifies leverage
points for intervention to raise self-esteem and curtail nar-
cissism from a young age.
KEYWORDS—narcissism; self-esteem; development; social-

ization
In the late 20th century, with the rise of individualism, self-es-
teem became a touchstone of Western parenting. Most Western
parents believe that children need self-esteem to achieve suc-
cess and happiness in life, and that parents play a crucial role
in building their children’s self-esteem (Miller & Cho, 2018).
Although parents are right that self-esteem is important (Orth &
Robins, 2014), their ideas about how to instill it may be mis -
guided. In particular, parents’ well-intentioned attempts to raise
self-esteem, such as lavishing children with praise, may
inadver-
tently cultivate narcissism (Brummelman, Thomaes, & Sedi-
kides, 2016). Narcissism is a subclinical personality trait that
predicts considerable maladjustment in children, ranging from
anxiety and depression to rage and aggression (Thomaes &
Brummelman, 2016). In 4%–15% of children, narcissism devel-
ops into Narcissistic Personality Disorder (Bernstein et al.,
1993).
How can parents raise children’s self-esteem without breeding

narcissism? Building on a burgeoning literature, we propose a
tripartite model, which holds that self-esteem without
narcissism
is cultivated through realistic feedback, a focus on growth, and
unconditional regard. We review evidence in support of this
model and discuss implications.
PILLARS OF NARCISSISM AND SELF-ESTEEM
Self-esteem is defined as a sense of one’s worth as a person
(Orth & Robins, 2014), whereas narcissism is defined as an
inflated sense of one’s importance and deservingness (Krizan &
Herlache, 2018). From childhood, narcissism can manifest as
grandiose (characterized by boldness, extraversion, and
boastful-
ness) or vulnerable (characterized by neuroticism, shyness, and
withdrawal; Derry, Ohan, & Bayliss, 2019). Here, we focus on
grandiose narcissism.
Eddie Brummelman, Research Institute of Child Development
and Education, University of Amsterdam, The Netherlands;
Constan-
tine Sedikides, Center for Research on Self and Identity,
Psychology

Department, University of Southampton, Southampton, United
Kingdom.
We thank Peggy Miller for encouraging us to situate narcissism
and self-esteem in a sociocultural context.
Correspondence concerning this article should be addressed to
Eddie Brummelman, Research Institute of Child Development
and
Education, University of Amsterdam, P.O. Box 15780, 1001 NG
Amsterdam, The Netherlands; e-mail: [email protected]
© 2020 The Authors
Child Development Perspectives published by Wiley Periodicals
LLC on behalf of Society
for Research in Child Development
This is an open access article under the terms of the Creative
Commons Attribution
License, which permits use, distribution and reproduction in any
medium, provided the
original work is properly cited.
DOI: 10.1111/cdep.12362
Volume 14, Number 2, 2020, Pages 83–89
CHILD DEVELOPMENT PERSPECTIVES
https://orcid.org/0000-0001-7695-5135
https://orcid.org/0000-0001-7695-5135
https://orcid.org/0000-0001-7695-5135
https://orcid.org/0000-0003-3681-4332

https://orcid.org/0000-0003-3681-4332
https://orcid.org/0000-0003-3681-4332
mailto:
http://creativecommons.org/licenses/by/4.0/
http://crossmark.crossref.org/dialog/?doi=10.1 111%2Fcdep.123
62&domain=pdf&date_stamp=2020-04-26
A common belief is that narcissism is simply an extreme form
of self-esteem. Psychologists have characterized narcissism as
inflated, exaggerated, or excessive self-esteem, or even as “the
dark side of high self-esteem” (Baumeister, Smart, & Boden,
1996, p. 5). These labels suggest that self-esteem represents a
continuum, with narcissism at its end. If this is so, narcissism
and self-esteem should correlate highly and there should be no
narcissists with low self-esteem. However, narcissism and self-
esteem are only weakly positively correlated, and about as many
narcissists have high self-esteem as low self-esteem (Brummel-
man et al., 2016). Thus, narcissism and self-esteem are indepen-
dent dimensions of the self.
How do narcissism and self-esteem differ? Researchers have
begun to identify differences in terms of underlying components

and the socialization experiences that give rise to them (Brum-
melman et al., 2016; Donnellan, Trzesniewski, Robins, Moffitt,
& Caspi, 2005; Hyatt et al., 2018; Tracy, Cheng, Robins, &
Trzesniewski, 2009). Here, we propose the first developmental
model that integrates these findings. The model describes the
distinct components, or pillars, that underlie narcissism and
self-esteem, and identifies the socialization practices that culti -
vate the development of these pillars. Our model seeks to
explain how narcissism and self-esteem differ in their underly-
ing components and socialization so it can address how parents
can raise self-esteem without breeding narcissism.
We theorize that narcissism and self-esteem are each based
on three distinct pillars (see Figure 1). In particular, we theorize
that narcissistic children have unrealistically positive views of
themselves (illusion), strive for superiority (superiority), and
oscillate between hubris and shame (fragility). By contrast,
chil-
dren with high self-esteem have positive but realistic views of
themselves (realism), strive for self-improvement (growth), and

feel intrinsically worthy, even in the face of setbacks (robust-
ness). Our model describes general patterns rather than univer -
sal laws. For example, most children with high self-esteem
strive for self-improvement, but some do not (Waschull & Ker-
nis, 1996).
Realism
Narcissistic children hold exalted views of themselves. An 11-
year-old narcissistic boy “unhesitatingly shared his certainty of
becoming president of the United States as soon as he graduated
from college with degrees in nuclear physics and brain surgery”
(Bleiberg, 1984, p. 508). Narcissistic children uphold such
views, even in the face of disconfirming evidence. For example,
after failing to complete challenging puzzles, narcissistic chil -
dren still believe that they performed extraordinarily well
(Derry
et al., 2019). Such grandiose self-views persist into adulthood.
Adult narcissists see themselves as geniuses, even if their IQ
scores are average; they think they are superb leaders, even if

they disrupt group performance; and they believe they are
attractive, even if others disagree (Grijalva & Zhang, 2016). By
contrast, children with high self-esteem have positive self-
views,
but those views tend to be more grounded in reality. For exam-
ple, they do not overestimate their performance as much as nar-
cissistic children do (Derry et al., 2019). Thus, narcissism is
marked by illusion, whereas self-esteem is marked by realism.
Growth
Narcissistic children strive for superiority. Narcissism is rooted
in the desire to stand out from and get ahead of others (Grapsas,
Brummelman, Back, & Denissen, 2020). In the service of supe-
riority, narcissistic children may look down on others and com-
pare themselves favorably to others (Thomaes & Brummelman,
2016). As narcissistic children look down on others, they may
feel little care, concern, or empathy for them (Thomaes, Stegge,
Bushman, Olthof, & Denissen, 2008). By contrast, children with
high self-esteem are more interested in improving themselves
than in outperforming others. For example, they believe they

can hone their abilities through effort and education (Robins &
Pals, 2002). Consequently, they are curious, interested, and
ready to assume challenging tasks to better themselves
(Waschull & Kernis, 1996). Children with high self-esteem may
not habitually compare themselves to others, but instead tend to
reflect on how they have improved over time (cf. G€urel, Brum-
melman, Sedikides, & Overbeek, 2020). Thus, narcissism is
marked by a striving for superiority, whereas self-esteem is
marked by a striving for growth.
Robustness
Narcissistic children have fragile feelings about themselves.
According to attribution theory, narcissistic children are
inclined to make stable and global self-attributions of both suc-
cesses and failures, causing them to oscillate between hubris
and shame (Lewis, 1992; Tracy et al., 2009). For example, when
narcissistic children receive negative feedback, they feel disap -
pointed in themselves and may even blush—a hallmark of
shame (Brummelman, Nikoli�c, & B€ogels, 2018). In response
to
shame, narcissistic children may lash out angrily or

aggressively
(Donnellan et al., 2005). Over time, shame may spiral into anxi -
ety and depression (Barry & Malkin, 2010). By contrast, chil-
dren with high self-esteem have relatively robust feelings about
themselves. They feel worthy, even in the face of failure
(Kernis,
Brown, & Brody, 2000; Tracy et al., 2009). Consequently, they
are unlikely to feel ashamed, and hence to become angry or
aggressive (Donnellan et al., 2005). These children are at
reduced risk of developing anxiety and depression (Orth &
Robins, 2014). Thus, narcissism is marked by fragility, whereas
self-esteem is marked by robustness.
Discussion
Our model holds that narcissism and self-esteem are built on
distinct pillars. Why then are narcissism and self-esteem weakly
but positively correlated? First, they share an agentic core—a
tendency to focus on oneself and the pursuit of one’s goals
(Hyatt et al., 2018). Like their counterparts with high self-
Child Development Perspectives, Volume 14, Number 2, 2020,
Pages 83–89

84 Eddie Brummelman and Constantine Sedikides
esteem, narcissists value competence and achievement (Hyatt
et al., 2018). Second, the pillars of narcissism and self-esteem
are not mutually exclusive. For example, some children might
strive for both growth and superiority, and others might strive
for neither. Thus, narcissism and self-esteem are not opposites
and can fluctuate independently of one another. Our thesis is
that pillars are foundational, that is, they precede the develop-
ment of narcissism and self-esteem. Of course, given the recur-
sive character of developmental processes, pillars may also
appear to simply co-occur with narcissism and self-esteem.
SOCIALIZATION OF NARCISSISM AND SELF-ESTEEM
By delineating the distinct pillars underlying narcissism and
self-esteem, our model opens the possibility of identifying
socialization experiences that cultivate self-esteem without
breeding narcissism. Psychologists have focused mostly on the
reverse—the socialization experiences that breed narcissism.

According to some psychoanalytic theories, narcissism develops
in response to lack of parental warmth (Kernberg, 1975). Chil-
dren raised this way are thought to develop deep-seated, uncon-
scious shame and self-loathing, and to engage in narcissistic
ideation to ward off these discomforting states (Lewis, 1987).
However, there is no evidence that narcissism is cultivated by
lack of parental warmth (Brummelman, Thomaes, Nelemans,
Orobio de Castro, Overbeek, et al., 2015; Wetzel & Robins,
2016) or that narcissists harbor unconscious shame or self-
loath-
ing (Bosson et al., 2008).
Recent studies indicate that narcissism is cultivated, in part,
by parental overvaluation. In longitudinal research, 7- to 11-
year
olds and their parents were studied prospectively over four mea -
surement waves (Brummelman, Thomaes, Nelemans, Orobio de
Castro, Overbeek, et al., 2015). Narcissism was predicted by
parental overvaluation—how much parents saw their child as a
special individual entitled to privileges. Cross-sectional
research

has replicated these findings (Derry, 2018; Nguyen & Shaw,
2020).
How exactly do overvaluing parents cultivate narcissism in
children? Conversely, how can parents raise children’s self-es-
teem without breeding narcissism? Our model suggests that nar-
cissism and self-esteem are cultivated by three classes of
socialization practices, corresponding to the three pillars (see
Figure 1). The model applies to children ages 7 and older, who
are aware that others, such as parents, evaluate them from an
external perspective (Harter, 2012). These evaluations can be
internalized and develop into stable self-evaluations. Indeed,
stable individual differences in narcissism and self-esteem can
be assessed reliably from age 7 (Thomaes & Brummelman,
2016).
Realistic Feedback
Overvaluing parents may foster narcissism by cultivating illu-
sion. These parents overestimate, overclaim, and overpraise
their
children’s qualities (Brummelman, Thomaes, Nelemans, Orobio

de Castro, & Bushman, 2015). For example, they believe that
their children are smarter than others, even when their chil -
dren’s IQ scores are average. They claim that their children
know about a wide range of topics, even ones that do not exist
(e.g., the fictional book, The Tale of Benson Bunny). Also, they
praise their children more often than other parents do, even
when the children do not perform well.
Overvaluing parents may express their overestimation through
inflated praise. Praise is inflated when it contains an adverb
(e.g., incredibly) or adjective (e.g., amazing) signaling an extre-
mely positive evaluation, such as “You did incredibly well!” In
a longitudinal observational study (Brummelman, Nelemans,
Figure 1. The hypothesized pillars of self-esteem (left) and
narcissism (right), and the socialization experiences
hypothesized to cultivate them. The circles’
overlap reflects the weak but positive correlation between self-
esteem and narcissism.
Pages 83–89
Tripartite Model 85

Thomaes, & Orobio de Castro, 2017), parents’ inflated praise
was coded from parent–child interactions; in a subgroup of chil-
dren, parents’ inflated praise predicted higher narcissism 6, 12,
and 18 months after the observations.
By contrast, parents may foster self-esteem by cultivating real-
ism. They can do so by providing children with realistic feed-
back (i.e., feedback that is relatively close to objective
benchmarks), which can help children understand themselves
more accurately. Because children prefer positive over negative
feedback, they may be inclined to dismiss negative feedback,
even if is diagnostic, because such feedback hurts (Sedikides,
2018). Yet, children may benefit more from moderately positive
feedback than from inflated praise. Although researchers have
not examined the causal link between realistic feedback and the
development of self-esteem, correlational evidence supports this
link. For example, when parents gave children realistic praise
(rather than overpraised or underpraised them), children earned
higher grade point averages and had fewer sympto ms of depres-

sion (Lee, Kim, Kesebir, & Han, 2017). When children felt that
their parents’ praise was slightly but not greatly overstated,
they
benefited as much as they did from realistic praise. Such find-
ings tentatively suggest that positive feedback helps children if
it closely matches reality.
Focus on Growth
Overvaluing parents may foster narcissism by cultivating chil -
dren’s striving for superiority. Such parents may pressure chil -
dren to stand out from others. For example, overvaluing parents
are likely to give their children uncommon first names (Brum-
melman, Thomaes, Nelemans, Orobio de Castro, & Bushman,
2015). Also, these parents are emotionally invested in their
chil-
dren’s social status (Grapsas, Brummelman, et al., 2020). For
example, while monitoring children’s status on social media,
overvaluing parents smiled when their child gained status, but
frowned when their child lost status, as revealed by their facial
muscle activity (Grapsas, Denissen, Lee, Bos, & Brummelman,

2020). Encouraging children to think they are better than others
triggers their narcissistic desire to be superior to others at the
expense of their desire to grow and learn (G€urel et al., 2020).
By contrast, parents may foster self-esteem by cultivating chil-
dren’s striving for growth. For example, when children succeed
at a task, parents may praise children’s effort and strategies
(e.g., “You found a good way to do it!”) to acknowledge the
suc-
cess but highlight that it was the result of hard work and effi-
cient strategies. When children receive such praise, they are
more likely to embrace challenges and persist in the face of set-
backs (Mueller & Dweck, 1998). Similarly, when children fail,
parents may talk to them about what they can learn from the
experience, how they can study their mistakes to improve, and
how they might ask for help. Over time, these practices can help
children embrace learning and growth rather than superiority
(Haimovitz & Dweck, 2016). Encouraging children to reflect on
their growth (e.g., how their skills have improved) raises their
self-esteem and sparks their desire for self-improvement
without

triggering narcissistic strivings for superiority (G€urel et al.,
2020).
Unconditional Regard
Overvaluing parents may foster narcissism by cultivating fragi-
lity. They can do so by making their regard conditional on the
child living up to their narcissistic standards. Overvaluing par -
ents state, “I would find it disappointing if my child was just a
‘regular’ child” (Brummelman, Thomaes, Nelemans, Orobio de
Castro, & Bushman, 2015, p. 678). Indeed, when the child
stands out from others, overvaluing parents may feel proud,
basking in the child’s reflected glory. However, when the child
is just “regular,” overvaluing parents may become disappointed
or even hostile (see Wetzel & Robins, 2016). Consequently, nar -
cissistic children may infer that their worth hinges on them
meeting their parents’ standards (Tracy et al., 2009), leading
them to attribute successes and failures to their whole selves
(Lewis, 1992). Although causal evidence on the link between
conditional regard and the development of narcissism is lacking,

correlational research shows that children who experienced con-
ditional regard from their parents displayed more narcissistic
traits, such as self-aggrandizement after success and self-
devalu-
ation after failure (Assor & Tal, 2012).
By contrast, parents may foster self-esteem by cultivating ro-
bustness. They can do so by giving children unconditional
regard. This does not mean that parents lavish their children
with praise no matter what they do; rather, it means that parents
accept children for who they are, even when the children fail
(Harter, Marold, Whitesell, & Cobbs, 1996). For example, when
children misbehave, parents may express unconditional regard
by correcting children’s behavior while continuing to be warm
and accepting toward them as a person (Kernis et al., 2000).
Similarly, when children work toward an achievement, parents
may express unconditional regard by valuing children regardless
of the outcome of their efforts. When parents express such
unconditional regard, children feel more connected to their true

selves (Harter et al., 1996) and have higher as well as more
stable self-esteem (Kernis et al., 2000). Extending these find-
ings, in a randomized intervention, children were invited to
reflect on times when they were accepted and valued by others
unconditionally (Brummelman et al., 2014). Three weeks later,
children received their first report card of the school year.
With-
out the intervention, children who received low grades felt
ashamed; with it, these painful feelings faded. Thus, uncondi -
tional regard made children’s feelings about themselves more
robust.
Discussion
The tripartite model posits that parents can raise self-esteem
without breeding narcissism by providing children with realistic
feedback (rather than inflated praise), focusing on growth
(rather
than on outperforming others), and giving unconditional regard
Pages 83–89
86 Eddie Brummelman and Constantine Sedikides

(rather than regard that is conditional). The model focuses on
the independent effects of these practices, but joint effects are
certainly possible. For example, parents’ realistic feedback may
be most effective in raising self-esteem when paired with
uncon-
ditional regard, teaching children that critical feedback on their
behavior does not signal lack of regard for them as a person.
Although evidence for these three classes of socialization
experiences has been accumulating, the field faces three major
challenges. First, the bulk of the literature has relied on subjec -
tive reports of socialization experiences. We call for observa-
tional and experience sampling methods to track socialization
experiences in children’s everyday lives. Second, most research
is cross-sectional or longitudinal. We call for experiments that
examine causal effects of socialization experiences. Third, no
interventions have sought to change socialization practices to
raise children’s self-esteem without breeding narcissism. We

call for research into the effectiveness of parenting
interventions
that teach realistic feedback, a focus on growth, and uncondi-
tional regard. By addressing these challenges, the field can
build a more precise understanding of what does—and does not
—contribute to children’s development of healthy self-esteem.
LOOKING AHEAD
The tripartite model generates new directions for research on
self-esteem interventions, the heterogeneity of narcissism and
self-esteem, and the sociocultural foundations of narcissism and
self-esteem.
Interventions
Our model can be used to rethink self-esteem interventions and
better understand their effectiveness. Our model suggests that
interventions can raise self-esteem by targeting its pillars—real-
ism, growth, and robustness. A challenge is that parents of nar -
cissistic children, who are most in need of such interventions,
are often narcissistic themselves (Brummelman, Thomaes, Nele-
mans, Orobio de Castro, & Bushman, 2015; Miller & Campbell,
2008). Given their inflated views of themselves and their chil -

dren, narcissistic parents may believe they do not need interven-
tion. Interventions can circumvent these concerns by changing
parenting through nudges, such as text messages suggesting
short, simple, specific activities for parents to do with their
chil-
dren (York, Loeb, & Doss, 2019). These nudges, even as they
change how parents behave, may not be seen as interventions
and therefore may engage even narcissistic parents.
Heterogeneity
Our model can unravel the heterogeneity of narcissism and self-
esteem. For example, narcissism can be manifested in both
grandiose and vulnerable ways (Derry et al., 2019). From the
perspective of the tripartite model, both manifestations are
marked by striving for superiority, but may differ in terms of
illusion and fragility. Would vulnerable narcissism be
characterized less by illusion and more by fragility than its
grandiose counterpart, as initial evidence indicates (Derry et al.,
2019)? If so, would inflated praise be more likely to predict

grandiose narcissism and conditional regard be more l ikely to
predict vulnerable narcissism? Addressing these questions will
help us understand why grandiose and vulnerable narcissism,
despite their shared foundation, manifest differently.
Sociocultural Foundations
Our model can examine the sociocultural foundations of narcis-
sism and self-esteem. Narcissism is more common in Western
cultures than in other cultures because these cultures embrace
individualism (Thomaes & Brummelman, 2016). Parents’ under-
standing of individualism reflects the specific needs, values,
beliefs, and concerns of their local worlds (Kusserow, 1999).
Working-class and low-income parents often adopt hard individ-
ualism, teaching their children to be tough and resilient in a
world of scarcity. Middle- and upper-class parents often adopt
soft individualism, helping children cultivate their unique
talents
and abilities in a world of opportunity—a process known as
con-
certed cultivation (Lareau, 2011). Unsurprisingly, middle- and
upper-class parents are more likely to cultivate narcissism in

their children than are working-class or low-income parents
(Martin, Côt�e, & Woodruff, 2016). Do these parents do so by
lavishing children with praise, comparing them favorably to
others who are less well-off, or making approval conditional on
worldly successes? Also, are these class-based practices more
common in unequal societies, where parents are more concerned
about their children’s relative standing? To address these ques -
tions, researchers need to expand their methodological reper-
toire because most studies are monocultural and include
predominantly Western middle-class participants.
CONCLUSION
Raising children’s self-esteem has become an important goal for
parents throughout Western society. Scholars and policymakers
have long feared that, in raising children’s self-esteem exces-
sively, parents may have risked turning them into narcissists
(Baumeister, Campbell, Krueger, & Vohs, 2003). However, evi -
dence indicates that narcissism and self-esteem are more dis-
tinct than previously thought. Our model shows that narcissism

and self-esteem are built on distinct pillars and that it is possi -
ble to raise children’s self-esteem without breeding narcissism.
We hope our model not only provides researchers with a frame-
work to study the origins and nature of narcissism and self-es-
teem but also helps parents build a strong foundation for their
children’s self-esteem.
REFERENCES
Assor, A., & Tal, K. (2012). When parents’ affection depends
on child’s
achievement: Parental conditional positive regard, self-
Pages 83–89
Tripartite Model 87
aggrandizement, shame and coping in adolescents. Journal of
Ado-
lescence, 35, 249–260.
https://doi.org/10.1016/j.adolescence.2011.
10.004
Barry, C. T., & Malkin, M. L. (2010). The relation between
adolescent
narcissism and internalizing problems depends on the
conceptual-

ization of narcissism. Journal of Research in Personality, 44,
684–
690. https://doi.org/10.1016/j.jrp.2010.09.001
Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K.
D. (2003).
Does high self-esteem cause better performance, interpersonal
suc-
cess, happiness, or healthier lifestyles? Psychological Science
in the
Public Interest, 4, 1–44. https://doi.org/10.1111/1529-
1006.01431
Baumeister, R. F., Smart, L., & Boden, J. M. (1996). Relation
of threat-
ened egotism to violence and aggression: The dark side of high
self-esteem. Psychological Review, 103, 5–33.
https://doi.org/10.
1037/0033-295x.103.1.5
Bernstein, D. P., Cohen, P., Velez, C. N., Schwab-Stone, M.,
Siever, L.
J., & Shinsato, L. (1993). Prevalence and stability of the DSM-
III-
R personality disorders in a community-based survey of adoles-
cents. The American Journal of Psychiatry, 150, 1237–1243.
https://doi.org/10.1176/ajp.150.8.1237
Bleiberg, E. (1984). Narcissistic disorders in children. A
developmental
approach to diagnosis. Bulletin of the Menninger Clinic, 48,
501–
517.
Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler-Hill, V.,
Jordan,

C. H., & Kernis, M. H. (2008). Untangling the links between
nar-
cissism and self-esteem: A theoretical and empirical review.
Social
and Personality Psychology Compass, 2, 1415–1439. https://doi.
org/10.1111/j.1751-9004.2008.00089.x
Brummelman, E., Nelemans, S. A., Thomaes, S., & Orobio de
Castro, B.
(2017). When parents’ praise inflates, children’s self-esteem
defla-
tes. Child Development, 88, 1799–1809.
https://doi.org/10.1111/
cdev.12936
Brummelman, E., Nikoli�c, M., & B€ogels, S. M. (2018).
What’s in a
blush? …
Ahmet Hamdi Imamoglu1 , Aysegul Durak Batigun2
DOI: 10.14744/DAJPNS.2020.00107
Dusunen Adam The Journal of Psychiatry and
Neurological Sciences 2020;33:388-401
How to cite this article: Imamoglu AH, Durak Batigun A. The
assessment of the relationship between narcissism, perceived
parental rearing
styles, and defense mechanisms. Dusunen Adam The Journal of
Psychiatry and Neurological Sciences 2020;33:388-401.
The assessment of the relationship between
narcissism, perceived parental rearing styles, and
defense mechanisms

1University of Health Sciences, Institute of Health Sciences,
Department of Psychology, Istanbul - Turkey
2Ankara University, Faculty of Languages History and
Geography, Department of Psychology, Ankara - Turkey
Correspondence: Aysegul Durak Batigun, Ankara University,
Faculty of Languages History and Geography, Department of
Psychology,
Ankara - Turkey
E-mail: [email protected]
Received: June 09, 2020; Revised: July 12, 2020; Accepted:
September 26, 2020
ABSTRACT
Objective: The aim of the study was to determine the
relationships between grandiose and vulnerable narcissism,
perceived
parental rearing styles and defense mechanisms. Besides, it was
investigated how grandiose narcissism and vulnerable
narcissism scores differ in terms of demographic variables such
as gender and age.
Method: The study was carried out with 508 participants
between the ages of 18-65 determined by the appropriate
sampling
method. 271 of the participants were female (53.3%), 237 of
them were male (46.7%). The data were collected through a
battery
including Demographic Information Form, Pathological
Narcissism Inventory, Narcissistic Personality Inventory-16,
Abbreviated
Perceived Parental Attitudes Scale-Child Form, Defense Style
Questionnaire-40, and Splitting Scale.

Results: Statistical analyses revealed that while there was no
significant difference in vulnerable narcissism scores between
male and female participants, male participants had
significantly higher scores in grandiose narcissism. The
findings indicate a
decrease in both vulnerable and grandiose narcissism as the age
of the participants increase. The regression analyses showed
that grandiose narcissism scores were predicted by the paternal
rejection and the maternal emotional warmth as perceived
parental attitudes, and by the immature defense style, splitting
defense mechanism, and neurotic defense style as defense
styles; vulnerable narcissism scores were predicted by the
paternal rejection, maternal overprotection, splitting defense,
and all
forms of defense.
Conclusion: Findings that narcissistic personality may be
correlated with some inadequate parental attitudes and more
frequent use of defense mechanisms were discussed within the
framework of this topic. In addition, the results were elaborated
regarding the theoretical framework of narcissism, and how it
can be used in clinical practice with narcissistic individuals.
Keywords: Defense mechanisms, grandiose narcissism,
perceived parental rearing styles, splitting defense mechanism,
vulnerable narcissism
RESEARCH ARTICLE
INTRODUCTION
The concept of narcissism expresses the exaggerated
love that one directs at oneself and his indifference
towards others. Narcissism, which is referred to in the
literature with its unique forms of relating and

defending, has been frequently examined by theorists
particularly from the psychoanalytic tradition since the
Imamoglu et al. The assessment of the relationship between
narcissism, perceived parental rearing styles, and defense
mechanisms 389
beginning of the last century and has become one of the
popular terms of our time (1). While some individuals
with narcissistic personalities exhibit typical narcissistic
traits such as arrogance, dominance and grandiosity; it
is stated that some of them have an implicit narcissistic
nature concealed by characteristics such as shyness and
humility (2). Due to its complex structure, narcissism
has been classified in different ways by many theorists
and evaluated as a multidimensional structure (3). Cain
et al. (4) determined that these dimensions generally
reflect two themes: grandiose and vulnerable. This
distinction has also been supported by various studies
(5,6) and has been widely accepted in the narcissism
literature (7). Grandiose narcissism is basically
characterized by exploiting, low empathy, jealousy,
aggression and pretentiousness (3). It has been reported
t h at i n d i v i d u a l s w i t h g r a n d i o s e n a r c i s s i s
t i c
characteristics have an intense desire to maintain their
positive self-perception and feel the need to gain the
admiration of others (8). According to Gabbard (9),
individuals with such a personality structure have a low
awareness of what kind of effect they have on others and
are insensitive to the needs of others. These individuals
also tend to have rude and arrogant attitudes in their
interpersonal relationships (10).

Vulnerable narcissism, on the other hand, includes
personality patterns that are often regarded as the
opposites of grandiose narcissistic traits. However, it is
thought that these two sub-dimensions of narcissism do
not completely exclude each other and that many
narcissistic individuals can exhibit the characteristics of
both types together (11). Individuals with vulnerable
narcissistic traits are hyper-sensitive to the reactions of
others, avoiding being the center of attention, and are
shy (9). However, it has been stated that they have
grandiose fantasies that are not clearly displayed under
their humble and shy images (12). Also, over-idealizing
others (13); embarrassment for grandiose desires (9);
excessive critical attitudes towards the self (14),
dysphoric affection and pessimism have also been
associated with vulnerable narcissism (12).
Studies show that the attitudes adopted by parents in
their interactions with their children play an important
role in the development of the child’s personality and
psychopathologies (15). Regarding the effects of
parental attitudes on the development of narcissistic
personality structure, views emphasizing excessively
tolerant, extremely intrusive or cold/strict parenting
styles come to the fore. (16). For example, according to
Kernberg (17), one of the important factors underlying
pathological narcissism is that it superficially displays
functional parental behaviors; but more fundamentally
it is the parent (usually mother) figure with an
indifferent, cold, or implicitly aggressive attitudes.
Young et al. (18) listed childhood experiences
accompanying narcissistic personality development
within a schematic model as loneliness, inadequate
boundaries, being used or directed, and conditional
approval. According to this approach, narcissistic

individuals did not acquire any true love, empathy and
closeness in their childhood. In a study conducted by
Cramer (19), vulnerable narcissism was positive with
the authoritarian parenting style of the mother; it was
found to be negatively correlated with maternal attitudes
perceived as sensitive and permissive. However,
grandiose narcissism is positive with the father’s
authoritarian parenting style and it was concluded that
there was a negative relationship with the father’s
perceived sensitive and tolerant attitudes.
Defense mechanisms are also one of the variables
whose relationship with narcissism is often discussed.
These mechanisms generally serve to keep affections
within the limits of which the individual can cope with,
to restore the psychological balance disturbed by the
increase in impulses, and to deal with life events that
create sudden and drastic changes in self-design, and
conflicts with other important people (20). However, it
has been indicated that strict, inappropriate and
excessive use of defense mechanisms are associated with
various psychopathologies and interpersonal problems
(21,22). Studies examining the relationship between
narcissism and defense mechanisms highlighted the
relationship between narcissism and immature defenses.
It has been suggested that these individuals mostly use
defense mechanisms of splitting, avoidance, denial,
outpacing, commitment, projection, and projective
identification (23). One of them, the splitting defense
mechanism has a distinct feature in its relationship to
narcissism. It is stated that narcissistic individuals often
use the defense of splitting (17,23-25). The splitting
defense mechanism refers to the separation of opposing
affections and positive and negative representations of
the self and others, and it is seen as the basic defense
mechanism of infancy when the ego still lacks the

capacity to integrate good and bad (26). In this period,
the baby wants to separate the good self and object
designs that are formed as a result of satisfying
experiences from the bad self and object designs
determined by frustration and aggressive impulses.
Thus, the splitting functions as a defense against the
anxiety created by ambivalent effects (27). It is accepted
that as a result of normal functioning developmental
Dusunen Adam The Journal of Psychiatry and Neurological
Sciences 2020;33:388-401390
processes, the splitting defense mechanism is replaced by
the defense of suppression, and the effect of splitting
decreases in adulthood. However, as a result of a
development process in which self and object
representations cannot be integrated, contradictory
representations continue to be actively separated from
each other (17). This leads to sudden transitions from
emotional situations in which the outside world and the
self are perceived completely well to emotional situations
in which they are perceived as completely bad (26).
Researchers draw attention to the function of some
defense mechanisms closely related to personality
disorders such as splitting, in coping with negative
affections that occur as a result of inappropriate parental
attitudes in childhood (17,25). Research findings on the
subject are generally based on a limited number of
longitudinal studies (28) and studies measuring
perceived parental attitudes within an adult sample (29).
For example, in a study with children and adolescents,
perceived maternal acceptance was positive with mature
defense; perceived maternal and paternal acceptance

was negatively associated with the immature form of
defense (30).
Considering all these, it is noteworthy that the
studies examining the relationship between narcissism
and perceived parental attitudes in our country are
limited to some thesis and do not focus on the
relationship of narcissism and splitting defense and
forms of defenses. In most of the studies conducted
abroad, it was observed that the perceived parental
attitudes were not assessed separately for the mother
and father, and the dimension of grandiosity was
emphasized in analyzing the relationship between
defense mechanisms and narcissism. Therefore, the
objective of this study is to determine the relationship
between grandiose and vulnerable narcissistic traits and
perceived parental attitudes and defense mechanisms.
In addition, determining how grandiose narcissism and
vulnerable narcissism scores differ in terms of
demographic variables such as gender and age is the
secondary aim of the study.
METHOD
Sample
The sample of the study was reached using the
convenience sampling method, who reside in Ankara and
Istanbul provinces. Considering the features and
conditions such as the purpose of the study, research
opportunities, the number of independent variables, and
sample selection technique, it was concluded that a
sample of approximately 500 people would be sufficient.
As a result, 508 participants between the ages of 18-65
(Mean=31.17, st andard de viat ion [SD]=11.37)
constituted the sample group. 271 of the participants

were female (53.3%) and 237 were male (46.7%). 3.4%
were primary school graduates, 9.4% were high school
graduates and 29.3% were university students whereas
57.7% were university and above graduates. 62% of the
sample was single and 35.6% were married. Information
on the monthly income levels of the participants was also
received. Accordingly, 23% was below 1500 TL, 17.1%
between 1500 TL-2499 TL, 17.1% between 2500 TL- 3499
TL, 24% between 3500 TL-5000 TL, while 17.1% of them
reported that they had an income of over 5000 TL.
Measures
Demographic Information Form: It is a form prepared
by the researcher to get responses from the participants
regarding their basic demographic information (gender,
age, education level, income level, family structure,
whether the mother and father are alive, marital status,
where they live most, and whether there has been a
psychiatric diagnosis in the last six months).
Pathological Narcissism Inventory (PNI): It is a
6-point Likert-type self-report scale scored (0=not like
me, 5=very similar to me) developed by Pincus et al.
(31). It was adapted to Turkish was conducted by
Buyukgungor (32). In this study, the number of items
was reduced to 40 by removing 12 items in the original
scale due to item correlations and factor loadings. As a
result of the analysis, a seven-factor structure was
obtained: Contingent Self-Esteem, Denial of the
Dependency, Grandiose Fantasy, Exploitativeness,
Entitlement Rage, Self-Sacrificing, Self-enhancement.
In the Turkish version of the PNI, it was observed that 6
subscales, excluding Exploitativeness, were clustered in
the Narcissistic Vulnerability dimension and this
dimension explained 45.27% of the variance. It was
reported that the Narcissistic Grandiosity dimension,

which consists only of the Exploitativeness subscale,
explained 15.21% of the variance. The Cronbach alpha
reliability coefficient obtained for the total score of the
scale was calculated as 0.91; in the subscales, this value
was determined to range from 0.23 to 0.63. Since all
subscales except Exploitativeness are included in the
vulnerable narcissism dimension, the Turkish version of
the PNI was evaluated mainly as a tool to measure the
vulnerable appearance of narcissism (32). In the present
study, the Narcissistic Vulnerability related dimensions
of the scale were used and the Cronbach alpha reliability
coefficient was determined as 0.94.
mechanisms 391
Narcissistic Personality Inventory-16 (NPI-16): It
is a self-report scale developed by Raskin and Hall (33)
according to the narcissistic personality disorder criteria
in DSM-III. Ames et al. (34) formed the 16-item form of
the NPI and each of these forms has two statements.
One of them indicates a narcissistic trait. Participants
are asked to read these pairs of items and mark the
statement they think reflects them. The adaptation to
Turkish study was carried out by Atay (35), and Gungor
and Selcuk (36) revised and rearranged some of its
statements. The Cronbach’s alpha internal consistency
coefficient of the scale was calculated as 0.75 and 0.74. It
is accepted that the grandeur narcissistic traits of the
participants increase as the scores obtained from the
scale whose total score can range from 0 and 16 increase.
In the present study, the Cronbach alpha reliability
coefficient for NPI-16 was calculated as 0.71.

Defense Styles Questionnaire-40 (DSQ-40): It is a
self-report scale consisting of 40 items organized by
Andrews et al. (37). The items are scored in Likert type
between 1 (not suitable for me at all) to 9 (very suitable
for me). The adaptation study of the scale was carried
out by Yılmaz et al. (38). As a result of the study, three
dimensions; mature, neurotic, and immature defense
were obtained and the Cronbach alpha internal
consistency coefficients for these dimensions were
calculated as 0.70, 0.61 and 0.83, respectively. The
increase in the scores obtained from the scale indicates
the increase in the use of the defense style to which the
relevant defense mechanism belongs. In this study,
Cronbach’s alpha reliability coefficients for mature,
neurotic and immature defense style subscales were
calculated as 0.63, 0.59 and 0.79, respectively.
Perceived Parenting Attitudes in Childhood- Short
EMBU-Children Form (S-EMBU-C): It is a 23-item
scale developed by Arrindell et al. (39) to assess the
perceived parental attitudes of adult individuals regarding
their childhood. On the scale, participants evaluate the
parental attitudes they perceive during childhood
separately for both their mothers and fathers. This
assessment is scored 1-4 Likert-type items in three
dimensions: overprotection, rejection, and emotional
warmth. The Turkish adaptation study of the scale was
carried out by Dirik et al. (40). In this study, the
Cronbach’s alpha internal consistency values in the
maternal subscales were 0.71, 0.68 and 0.65 for
overprotection, rejection and emotional warmth,
respectively while it was 0.50, 0.72 and 0.73 for paternal
overprotection, rejection and emotional warmth,
respectively. High scores from subscales indicate an
increase in perceived parenting attitudes for that subscale.

In the present study, Cronbach’s alpha reliability
coefficients were calculated as 0.80, 0.83 and 0.80,
respectively for the dimensions of the S-EMBU-C of
p e r c e i v e d e m o t i o n a l w a r mt h , r e j e c t i o n a n
d
overprotection regarding the mother. Cronbach’s alpha
reliability coefficients for the dimensions of perceived
p a t e r n a l e m o t i o n a l w a r m t h , r e j e c t i o n a n
d
overprotection were found to be 0.82, 0.84 and 0.80,
respectively.
Splitting Scale (SS): It is a 7-point Likert-type scale
with 14 items developed by Gerson (41) to assess the
splitting defense mechanism in individuals. The scores
obtained from the scale range from 14 to 98, and higher
scores indicate more frequent use of splitting the defense
mechanism. The Turkish adaptation study of the scale
was carried out by Alkan (42), and the Cronbach alpha
interior consistency coefficient was calculated as 0.70.
The test-retest and Guttman two-half test reliability
coefficients were reported as 0.85 and 0.78, respectively.
Cronbach alpha reliability coefficient in this study was
calculated as 0.73.
Procedure
First, approval was obtained from the Ankara University
Ethics Committee (Approval no: 17/280 73921) and the
above-mentioned measurement tools were converted
into a battery. Before the study, participants were asked
to sign the informed consent form by providing written
and oral information about the scope and the ethical
framework of the study. The application was made on
an individual basis and lasted about 20-25 min.
Participants with end values and the missing data above

acceptable levels (more than 10% of the number of
items in the scale) were excluded from the data set to
make the data obtained ready for the analysis. Statistical
analyses were performed with the SPSS-21 program in
line with the research questions, after assigning new
values with an acceptable level to replace the missing
data with the mean assignment method. In order to
minimize the sequence effect, other scales were included
in the battery in a different order, with the informed
consent form and demographic information form
appeared at the beginning.
RESULTS
Analyzes on the Gender Variable
T-test analysis was conducted for independent groups
to determine whether the dependent variable scores of
the study, grandiose narcissism and vulnerable
narcissism, differed according to gender. As a result of
Sciences 2020;33:388-401392
the analysis, no significant difference was
observed in terms of vulnerable narcissism
scores (Male: Mean=2.39, SD=0.81) (Female:
Mean=2.35, SD=0.90), (t=0.40, p>0.05] while
men’s grandiose narcissism scores (Mean=5.46,
SD=3.14) was found to be significantly higher
(t=3.11, p<0.001) than women (Mean=4.62,
SD=2.99).
Association Between Variables (Correlation
Analysis)

Pe a r s on P ro du c t - Mom e nt C or re l at i on
Coefficients were calculated to determine the
association between all variables considered in
the study. Statistically, values of 0.05 and below
(p<0.05) were considered significant. In this
and the following statistical analyzes, two
items related to splitting defense under the
immature defense style subscale of the Defense
Styles Questionnaire were excluded, and the
splitting defense mechanism was included in
the analysis as a variable measured only by the
Splitting Scale.
As a result of the correlation analysis, it
was observed that there were negative and
significant associations between the age
variable and grandiose narcissism scores and
vulnerable narcissism scores. In addition,
grandiose narcissism scores show a significant
relationship with the perceived maternal
ove r prote c t i on and p e rc e ive d p ate r na l
r e j e c t i o n . H o w e v e r, t h e c o r r e l a t i o n
coefficients between grandiose narcissism
scores and other perceived parental attitudes
were found to be not significant. However,
vulnerable narcissism showed significant
associations with all perceived parental
attitudes, except for the perceived maternal
emotional warmth.
The grandiose narcissism variable was
found to be significantly associated with
immature defense style, mature defense style
and splitting defense mechanism; while the
vulnerable narcissism variable was found to be
associated with immature defense style,

neurotic defense style, and splitting defense
m e c h a n i s m . Ho w e v e r, n o s i g n i f i c a nt
correlation was found between grandiose
narcissism and neurotic defense style, and
between vulnerable narcissism and mature
defense style. The results are shown in Table 1. T
ab
le
1
: C
o
rr
el
at
io
n
c
o
ef
fi
ci
en
ts
b
et
w

ee
n
v
ar
ia
b
le
s
1
2
3
4
5
6
7
8
9
10
11
12
13
1.
A

g
e
-
-0
.2
3*
**
-0
.1
5*
**
-0
.0
5
-0
.0
4
-0
.0
9*
-0
.0
8
-0

.0
5
-0
.0
6
-0
.1
6*
**
0.
03
-0
.0
5
-0
.1
7*
**
2.
V
u
ln
er
ab

le
N
ar
ci
ss
is
m
-
0.
34
**
*
-0
.0
7
-0
.1
2*
*
0.
30
**
*
0.

29
**
*
0.
26
**
*
0.
29
**
*
0.
54
**
*
0.
34
**
*
-0
.0
2
0.
67

**
*
3.
G
ra
n
d
io
se
N
ar
ci
ss
is
m
-
0.
07
-0
.0
1
0.
09
*

0.
06
0.
08
0.
14
**
0.
28
**
*
0.
03
0.
10
*
0.
23
**
*
4.
P
er
ce

iv
ed
M
at
er
n
al
E
m
o
ti
o
n
al
W
ar
m
th
-
0.
70
**
*
-0

.1
2*
*
-0
.1
2*
*
-0
.4
9*
**
-0
.3
3*
**
-0
.1
5*
**
0.
07
0.
21
**

*
-0
.1
5*
**
5.
P
er
ce
iv
ed
P
at
er
n
al
E
m
o
ti
o
n
al
W

ar
m
th
-
-0
.1
7*
**
-0
.0
8
-0
.3
8*
**
-0
.4
4*
**
-0
.1
8*
**
0.

06
0.
17
**
*
-0
.1
8*
**
6.
P
er
ce
iv
ed
M
at
er
n
al
O
ve
rp
ro

te
ct
io
n
-
0.
74
**
*
0.
50
**
*
0.
42
**
*
0.
20
**
*
0.
05
-0

.1
0*
0.
27
**
*
7.
P
er
ce
iv
ed
P
at
er
n
al
O
ve
rp
ro
te
ct
io

n
-
0.
38
**
*
0.
53
**
*
0.
24
**
*
0.
14
**
-0
.1
0*
0.
29
**
*

8.
P
er
ce
iv
ed
M
at
er
n
al
R
ej
ec
ti
o
n
-
0.
68
**
*
0.
25

**
*
0.
05
-0
.0
6
0.
25
**
*
9.
P
er
ce
iv
ed
P
at
er
n
al
R

ej
ec
ti
o
n
-
0.
31
**
*
0.
12
**
-0
.0
6
0.
27
**
*
10
. I
m
m

at
u
re
D
ef
en
se
S
ty
le
-
0.
41
**
*
0.
25
**
*
0.
56
**
*
11

. N
eu
ro
ti
c
D
ef
en
se
S
ty
le
-
0.
25
**
*
0.
33
**
*
12
. M
at

u
re
D
ef
en
se
S
ty
le
-
-0
.0
1
13
. S
p
lit
ti
n
g
D
ef
en
se

M
ec
h
an
is
m
-
A
ve
ra
g
e
31
.1
7
2.
37
5.
01
20
.1
8
18
.7

6
20
.7
9
19
.0
7
10
.2
0
9.
98
91
.2
9
38
.3
9
44
.2
1
53
.7
3

St
an
d
ar
d
d
ev
ia
ti
o
n
11
.3
7
0.
86
3.
08
4.
42
4.
76
5.
59

5.
43
3.
84
3.
84
23
.0
1
9.
81
9.
71
12
.2
7
C
ro
n
b
ac
h
a
lf

a
-
0.
94
0.
71
0.
80
0.
82
0.
80
0.
80
0.
83
0.
84
0.
79
0.
59
0.
63

0.
73
*p
<
0.
05
, *
*p
<
0.
01
, *
**
p
<
0.
00
1
mechanisms 393
Regression Analysis
In this part of the study, two separate-stages of linear

regression analysis were conducted to determine the
variables that predicted the grandiose narcissism and
vulnerable narcissism scores of the participants. In both
regression analyzes; demographic variables (age, gender,
income level) in the first stage, perceived parental
attitudes (emotional warmth, rejection, overprotection)
in the second stage; defense styles (immature, neurotic,
mature) and splitting defense mechanism in the final
stage were included in the equation.
In the first regression analysis in which grandiose
narcissism was considered as the dependent variable
(Table 2), the first predictor variable was age and
explained 2% of the variance (F=11.05, p<0.001). It was
observed that the gender variable included in the
equation in the same step increased the explained
variance to 4% (F=11.11, p<0.001). Among the
perceived parental attitudes included in the second
stage in the analysis, the perceived paternal rejection
increased the explained variance to 5% (F=10.60,
p<0.001), while the perceived maternal emotional
warmth to 6% (F=9.56, p<0.001). In the final step,
defense styles and splitting defense were included in the
analysis. Among these variables, the variance explained
by the immature defense style reached 11% (F=13.98,
p<0.001). While 12% of the variance (F=12.69, p<0.001)
was explained by the addition of the splitting defense
mechanism, it was observed that the total variance
explained by the neurotic defense style increased to 13%
(F=12.04, p<0.001). Examining the beta values, it was
observed that the strongest predictor was the immature
defense style (β=0.22) followed by the perceived
maternal emotional warmth (β=0.16).
In the second regression analysis in which vulnerable
narcissism was considered as the dependent variable

(Table 3), only age was found to have a significant
predictive effect among the demographic variables
included in the first step and explained 5% of the
variance (F=27.68, p<0.001). In the second step, the
explained variance increased to 13% (F=37.10, p<0.001)
by the inclusion of the perceived paternal rejection from
the parental attitudes in the analysis. The perceived
maternal overprotection increased the explained
variance to 16% (F=31.97, p<0.001). In the final step of
the regression analysis, the splitting defense mechanism
and defense styles were involved. At this step, the
splitting defense mechanism increased the explained
variance to 48% (F=115.11, p<0.001). It was then
obser ved that the explained variance was 50 %
(F=102.78, p<0.001) by the inclusion of the immature
defense style to the model. With the contribution of the
neurotic defense style, this rate reached 51% (F=88.68,
p<0.001). Finally, it was determined that the predictive
effect of mature defense style was significant and the
total explained variance increased to 52% (F=78.07,
p<0.001) with this variable. When the beta values were
examined, it was observed that the strongest predictor
was the splitting defense mechanism (β=0.46) followed
by the immature defense style (β=0.16).
DISCUSSION
In the study, it was first examined whether the grandiose
narcissism and vulnerable narcissism scores differ by
gender and age. According to the findings, while the
vulnerable narcissism scores of the participants showed
no significant difference in terms of gender variable, the
grandiose narcissism scores of the men were observed
to be significantly higher than those of women.
Reviewing the literature, some studies show no gender

difference (43,44) while in some studies, men’s scores of
grandiose narcissism are significantly higher than those
of women, similar to the findings in this study (45-48).
There are several explanations as to why grandiose
narcissism is more common among men. For example,
Wardetzki (49) asserted that although men and women
basically have the same narcissistic disorder; the gender -
related social norms are shaping how this narcissistic
nature is expressed. According to him, while the
narcissistic personality of women tends to be concealed
more harmoniously within the cultural structure, the
narcissistic personality of the men is formed to be
expressed more grandiose and omnipotent way by the
cultural structure. According to Morf and Rhodewalt
(50), since some characteristics such as exploitativeness,
dominance and grandiosity are regarded as socially
more acceptable to men, women meet their narcissistic
needs more implicitly and indirectly per their gender
roles. In addition to these approaches; it has been
suggested that men might have acquired these
characteristics more than women in the evolutionary
process due to the advantages of grandiose narcissistic
traits such as leadership, aggression and competitiveness
in terms of continuity of species and biology (51). For
all these possible reasons, more stereotypic narcissistic
traits, such as grandiose narcissistic personality features,
may be more common in men.
In this study, age was considered as a variable besides
gender. As a result of the correlation and regression
analyzes, it was obser ved that as the age of the
individuals increased, their grandiose and vulnerable

Sciences …
The University of Southern Mississippi The University of
Southern Mississippi
The Aquila Digital Community The Aquila Digital Community
Dissertations
Summer 8-2019
Overparenting and Young Adult Narcissism: Psychological
Control Overparenting and Young Adult Narcissism:
Psychological Control
and Interpersonal Dependency as Mediators and Interpersonal
Dependency as Mediators
Nathan Alexander Winner
University of Southern Mississippi
Follow this and additional works at:
https://aquila.usm.edu/dissertations
Part of the Child Psychology Commons, Clinical Psychology
Commons, Developmental Psychology
Commons, Family, Life Course, and Society Commons, Other
Psychology Commons, Other Social and
Behavioral Sciences Commons, and the Personality and Social
Contexts Commons
Recommended Citation Recommended Citation

Winner, Nathan Alexander, "Overparenting and Young Adult
Narcissism: Psychological Control and
Interpersonal Dependency as Mediators" (2019). Dissertations.
1483.
https://aquila.usm.edu/dissertations/1483
This Dissertation is brought to you for free and open access by
The Aquila Digital Community. It has been accepted
for inclusion in Dissertations by an authorized administrator of
The Aquila Digital Community. For more
information, please contact [email protected]
https://aquila.usm.edu/
https://aquila.usm.edu/dissertations
https://aquila.usm.edu/dissertations?utm_source=aquila.usm.edu
%2Fdissertations%2F1483&utm_medium=PDF&utm_campaign=
PDFCoverPages
http://network.bepress.com/hgg/discipline/1023?utm_source=aq
uila.usm.edu%2Fdissertations%2F1483&utm_medium=PDF&ut
m_campaign=PDFCoverPages
http://network.bepress.com/hgg/discipline/406?utm_source=aqui
la.usm.edu%2Fdissertations%2F1483&utm_medium=PDF&utm_
campaign=PDFCoverPages
la.usm.edu%2Fdissertations%2F1483&utm_medium=P DF&utm_

https://aquila.usm.edu/dissertations/1483?utm_source=aquila.us
m.edu%2Fdissertations%2F1483&utm_medium=PDF&utm_cam
paign=PDFCoverPages
mailto:[email protected]
OVERPARENTING AND YOUNG ADULT NARCISSISM:
PSYCHOLOGICAL CONTROL AND INTERPERSONAL
DEPENDENCY AS MEDIATORS
by
A Dissertation
Submitted to the Graduate School,
the College of Education and Psychology,
and the Department of Psychology
at The University of Southern Mississippi
in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy
August 2019
PSYCHOLOGICAL CONTROL AND INTERPERSONAL
DEPENDENCY AS MEDIATORS
by Nathan Alexander Winner
August 2019
Approved by:
________________________________________________
Dr. Bonnie C. Nicholson, Committee Chair
Associate Professor, Psychology
________________________________________________
Dr. Eric R. Dahlen, Committee Member

Associate Professor, Psychology
________________________________________________
Dr. Ashley B. Batastini, Committee Member
Assistant Professor, Psychology
________________________________________________
Dr. Richard S. Mohn, Committee Member
Associate Professor, Educational Research and Administration
________________________________________________
Dr. D. Joe Olmi
Chair, Department of Psychology
________________________________________________
Dr. Karen S. Coats
Dean of the Graduate School
COPYRIGHT BY

2019
Published by the Graduate School
ii
ABSTRACT
PSYCHOLOGICAL
CONTROL AND INTERPERSONAL DEPENDENCY AS
MEDIATORS
by Nathan Alexander Winner
August 2019
Overparenting, or “helicopter parenting,” is a unique style of
parenting
characterized by parents’ well-intentioned but age-inappropriate
over-involvement and
intrusiveness in their children’s lives. Recent research has
linked overparenting to the
development of narcissistic traits in young adults, although the

mechanisms of this
relationship remain unclear. Two plausible mechanisms include
the parenting behavior of
psychological control and the increased interpersonal
dependency of the child.
Psychological control is a construct that overlaps with
overparenting and has been linked
to both dependent and narcissistic traits. Similarly,
interpersonal dependency is a key
predictor of narcissistic traits. Therefore, the present study
sought to examine
psychological control and interpersonal dependency as
sequential mediators in the
relationship between overparenting and young adult narcissistic
traits. It was
hypothesized that greater levels of overparenting would be
mediated by both greater
levels of parental psychological control and greater levels of
interpersonal dependency
among young adult children in predicting narcissistic traits.
Additionally, it was predicted
that these mediating relationships would be more pronounced
when examining vulnerable
narcissistic traits compared to grandiose narcissistic traits.

Results supported these
hypotheses. These findings highlight the mechanisms by which
overparenting predicts
narcissistic traits, as well as shed light on the multifaceted
nature of narcissism.
iii
ACKNOWLEDGMENTS
This project could not have been accomplished without the
support and
encouragement of my major professor, Dr. Nicholson, as well as
the members of my
committee, including Dr. Dahlen, Dr. Batastini, and Dr. Mohn.
Finally, I wish to thank all
of my peers and colleagues for their support, and especially
those on the Positive
Parenting Research Team, who have encouraged me throughout
this process.

iv
DEDICATION
This dissertation is dedicated to my parents, for their
unwavering love,
encouragement, and support.
v
TABLE OF CONTENTS
ABSTRACT
...............................................................................................
......................... ii
ACKNOWLEDGMENTS
...............................................................................................
.. iii
DEDICATION
...............................................................................................
.................... iv

LIST OF TABLES
...............................................................................................
............. vii
LIST OF ILLUSTRATIONS
...........................................................................................
viii
CHAPTER I - INTRODUCTION
...................................................................................... 1
CHAPTER II - METHODOLOGY
.................................................................................. 22
Measures
...............................................................................................
........................ 23
Demographic Questionnaire
..................................................................................... 23
Helicopter Parenting Instrument (HPI)
..................................................................... 24
Pathological Narcissism Inventory (PNI)
................................................................. 24
Psychological Control Scale- Youth Self-Report (PCS-YSR)
................................. 26
Interpersonal Dependency Inventory (IDI)
............................................................... 27
CHAPTER III - RESULTS
...............................................................................................
29

CHAPTER IV –
DISCUSSION..........................................................................
.............. 37
Limitations
...............................................................................................
..................... 39
Areas for Future Research
............................................................................................
40
Conclusion
...............................................................................................
..................... 41
APPENDIX A – IRB Approval Letter
.............................................................................. 43
vi
APPENDIX B – Electronic Informed Consent
................................................................. 44
REFERENCES
...............................................................................................
.................. 46

vii
LIST OF TABLES
Table 1 Means, Standard Deviations, and Bivariate Correlations
for Study Measures .... 29
viii
LIST OF ILLUSTRATIONS
Figure 1. Differences in Effect of Overparenting on Narcissistic
Phenotypes. ................ 30
Figure 2. Mediation of Parental Psychological Control between
Overparenting and
Narcissistic Phenotypes.
...............................................................................................
.... 32
Figure 3. Mediation of Young Adult Interpersonal Dependency
between Overparenting
and Narcissistic Traits.
...............................................................................................
....... 33

Figure 4. Mediation of Young Adult Interpersonal Dependency
between Overparenting
and Narcissistic Phenotypes.
.............................................................................................
34
Figure 5. Parallel Mediation of Parental Psychological Control
and Young Adult
Interpersonal Dependency between Overparenting and
Narcissistic Traits. .................... 35
Figure 6. Parallel Mediation of Parental Psychological Control
and Young Adult
Interpersonal Dependency between Overparenting and
Narcissistic Phenotypes. ........... 36
1
CHAPTER I - INTRODUCTION
Narcissism, broadly defined as an interpersonal pattern
characterized by a sense
of entitlement, an unhealthy need for admiration, and a general
lack of empathy (Miller &

Narcissism, bullying, and social dominance in youth a longit

Narcissism, bullying, and social dominance in youth a longit

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to Narcissism, bullying, and social dominance in youth a longit

Similar to Narcissism, bullying, and social dominance in youth a longit (20)

More from amit657720

More from amit657720 (20)

Recently uploaded

Recently uploaded (20)

Narcissism, bullying, and social dominance in youth a longit