Content uploaded by Erin Myers
Author content
All content in this area was uploaded by Erin Myers on Nov 15, 2017
Content may be subject to copyright.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Brief Report
How much do narcissists really like themselves? Using the bogus pipeline procedure
to better understand the self-esteem of narcissists
Erin M. Myers
a,
⇑
, Virgil Zeigler-Hill
b
a
Department of Psychology, Western Carolina University, Cullowhee, NC 28723, United States
b
Oakland University, Rochester, MI 48309, United States
article info
Article history:
Available online 10 October 2011
Keywords:
Narcissism
Self-esteem
Bogus pipeline
abstract
Narcissism is considered to be a pathological form of self-love but the extent to which narcissists actually
like themselves is unclear. The present study utilized the bogus pipeline technique in order to examine
how narcissists actually feel about themselves. Participants were 71 women who completed measures of
narcissism and self-esteem before responding to a self-esteem measure under either bogus pipeline or
control conditions. Women with high levels of narcissism reported lower levels of self-esteem in the
bogus pipeline condition than the control condition. This suggests that women with high levels of narcis-
sism may not actually feel as good about themselves as they often claim.
Ó2011 Elsevier Inc. All rights reserved.
1. Introduction
The dynamics underlying the narcissistic personality are very
complex. For example, narcissistic individuals tend to express
grandiose beliefs about themselves but these beliefs are often
highly susceptible to threat (see Morf and Rhodewalt (2001), for
a review). This combination of grandiosity and vulnerability is be-
lieved to be responsible for many of the self-serving and defensive
behaviors that are exhibited by narcissists (e.g., self-aggrandize-
ment). One of the most influential perspectives concerning the
dynamics of narcissism is the psychodynamic mask model of narcis-
sism which is derived from the models of narcissism offered by
Kohut (1966) and Kernberg (1975). Despite differences in their
views of narcissism, Kohut and Kernberg agree that the overt gran-
diosity of narcissists serves as a façade to conceal the underlying
feelings of inferiority and low self-esteem that are believed to stem
from early experiences of inadequate and insensitive parenting.
That is, the grandiosity expressed by narcissists is not believed to
be completely authentic. Variations of this basic idea are found
in many current perspectives of narcissism (e.g., the dynamic
self-regulatory model of narcissism; Morf & Rhodewalt, 2001)
and have been incorporated into the diagnostic criteria for Narcis-
sistic Personality Disorder which specifies that the self-esteem of
narcissists is ‘‘almost invariably very fragile’’ (American Psychiatric
Association, 2000, p. 714). Despite the abundance of past research
examining narcissism, the extent to which narcissists actually like
themselves remains unclear (see Zeigler-Hill and Jordan (2011), for
a review of this issue).
The idea that narcissists harbor hidden feelings of inferiority
and undisclosed negative self-views has been of tremendous
interest to researchers interested in unraveling the dynamics of
narcissism but it has been a daunting task to find a means for
delving behind the grandiose façade that narcissists present to
the world. The development of measures designed to capture
implicit self-esteem provided one possible means for bypassing
the self-proclaimed grandiosity of narcissists by allowing research-
ers to access their underlying feelings of self-worth. Implicit self-
esteem is often assessed using nonreactive measures which cap-
ture feelings of self-worth through reaction times or responses to
ambiguous stimuli rather than simply asking respondents how
much they like themselves (Zeigler-Hill & Jordan, 2010). Initial
studies concerning implicit self-esteem supported the psychody-
namic mask model by showing that the highest levels of narcissism
were reported by individuals with high explicit self-esteem but
low implicit self-esteem (Jordan, Spencer, Zanna, Hoshino-Browne,
& Correll, 2003; Zeigler-Hill, 2006); however, subsequent studies
have failed to replicate this pattern (Bosson et al., 2008; Gregg &
Sedikides, 2010). The inconsistencies between these results have
left researchers with an unclear picture of how narcissists truly feel
about themselves.
The goal of the present study was to improve our understanding
of narcissism by attempting to determine how narcissists really
feel about themselves. That is, we wanted to know if narcissists
actually believe the high levels of self-esteem they report experi-
encing or if those feelings of self-worth are at least somewhat
inflated. This was accomplished by creating a situation in which
individuals would be motivated to provide more honest responses
0092-6566/$ - see front matter Ó2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jrp.2011.09.006
⇑
Corresponding author. Address: Department of Psychology, Western Carolina
University, 91 Killian Building Lane, Room 301, Cullowhee, NC 28723, United States.
Fax: +1 828 227 7005.
E-mail address: emmyers@email.wcu.edu (E.M. Myers).
Journal of Research in Personality 46 (2012) 102–105
Contents lists available at SciVerse ScienceDirect
Journal of Research in Personality
journal homepage: www.elsevier.com/locate/jrp
Author's personal copy
than may be typical for them. We used the bogus-pipeline tech-
nique to foster this sort of situation (e.g., Jones & Sigall, 1971).
The bogus pipeline technique promotes greater honesty on the part
the participant through the use of physiological equipment (e.g., a
lie detector) that ostensibly allows the researcher to detect if indi-
viduals are attempting to lie. This technique has been used suc-
cessfully in past research (e.g., Boysen, Vogel, & Madon, 2006)
and was employed in the present study in order to gain a better
understanding of the way narcissists actually feel about them-
selves. Our prediction was that individuals with high levels of nar-
cissism would report significantly lower levels of self-esteem in
the bogus pipeline condition than in the control condition. The
rationale for this prediction was that the high levels of self-esteem
reported by narcissists may not be entirely authentic as suggested
by the psychodynamic mask model.
2. Method
2.1. Participants and procedure
Participants were 71 undergraduate women at the University of
Southern Mississippi who received partial fulfillment of a research
participation requirement in exchange for their participation.
1
The
mean age of participants was 21.41 years (SD = 3.83) and the racial/
ethnic composition was 38% White, 48% Black, and 14% Other. It
should be noted that the inclusion of race in the preliminary analy-
ses did not qualify the present results. Consequently, race will not be
discussed further.
During the first phase of the study, participants completed the
Narcissistic Personality Inventory (NPI; Raskin & Hall, 1979) and
the State Self-Esteem Scale (SSES; Heatherton & Polivy, 1991)–
along with other measures that are not relevant for the present
study – via a secure website. The NPI is a true–false measure that
captures narcissistic personality features (e.g., ‘‘I am more capable
than other people’’). The present research utilized the 37-item
shortened version of the NPI that Morf and Rhodewalt (1993)
adapted from Emmons’s (1987) factor analysis of the original
NPI. This abridged instrument only includes items with factor load-
ings greater than .35 (Emmons, 1987) and eliminates duplicate
items. The construct validity and internal consistency of the NPI
has been previously demonstrated (Emmons, 1987; Raskin & Hall,
1981; Raskin & Terry, 1988). For the present sample, the internal
consistency of the NPI was .86.
The SSES is a 20-item measure of self-esteem designed to cap-
ture feelings of self-worth across three domains: performance
(e.g., ‘‘I feel frustrated or rattled about my performance’’), social
(e.g., ‘‘I feel that others respect and admire me’’), and appearance
(e.g., ‘‘I feel unattractive’’). Responses were made on scales ranging
from 1 (not at all)to5(extremely). The SSES has been shown to be a
valid and reliable measure of state self-esteem (Heatherton &
Polivy, 1991). For the present sample, the internal consistency of
the SSES composite score was high (
a
= .91).
During the second phase of the study, participants attended a
laboratory session and were randomly assigned to either the bogus
pipeline or the control condition. In both conditions, participants
were asked to complete the Rosenberg Self-Esteem Scale (RSES;
Rosenberg, 1965) along with other measures that were intended
to disguise the true purpose of the study. The RSES is a widely used
10-item measure of global self-esteem (e.g., ‘‘I take a positive
attitude toward myself’’) for which participants were asked to
provide ratings of agreement based on how they generally feel
about themselves using scales that range from 1 (strongly disagree)
to 5 (strongly agree). Past research has demonstrated the construct
validity and reliability of the RSES (Blascovich & Tomaka, 1991;
Rosenberg, 1986). The internal consistency of the RSES was .84.
2.1.1. Bogus-pipeline condition
Participants in the bogus pipeline condition were told that they
would be monitored by lie-detecting physiological equipment and
videotaped for the duration of the session. This was done in order
to create the impression that the experimenter would be able to de-
tect dishonest responses. Participants were then escorted into a
small room containing physiological testing equipment (i.e., gal-
vanic skin response, automatic blood pressure monitor, Grass Mod-
el 78D polygraph) where they were seated in a recliner, connected
to the physiological instruments, and restrained with arm straps. In
order to enhance the believability of the situation, participants
were asked to respond to several questions that would ostensibly
allow the experimenter to confirm that the physiological equip-
ment was functioning properly (e.g., ‘‘Tell me your name’’ and
‘‘Lie to me about the color of my lab coat’’). Participants were again
reminded that their physiological measurements were being col-
lected for the express purpose of assessing their potential misrepre-
sentation of attitudes (e.g., ‘‘We will be able to tell if you are lying’’).
Participants were instructed to read each item aloud from the RSES
as it was projected on the screen in front of the participant and pro-
vide a verbal response that was recorded by the researcher.
2.1.2. Control condition
In accordance with past research (Boysen et al., 2006), the control
condition was designed to be nearly identical to the bogus pipeline
condition and differed only by the nature of the cover story. Partic-
ipants were informed that they would be connected to the physio-
logical equipment during the session for the purpose of allowing
the experimenter to gain practice with the equipment. Before begin-
ning the task, the experimenter clearly deactivated the physiological
equipment (i.e., ‘‘I am turning the equipment off now’’) and ex-
plained that the session would be videotaped for the sole purpose
of allowing the experimenter to review laboratory procedures.
3. Results
Table 1 presents descriptive statistics for the measures in the
present study. Preliminary analyses utilizing pre-manipulation
narcissism and self-esteem scores confirmed that participants in
the bogus pipeline condition did not significantly differ from par-
ticipants in the control condition with regard to pre-manipulation
(Phase 1) measures of narcissism (t<1, ns) or self-esteem (t<1,
ns). In addition, the zero-order correlations between narcissism
and post-manipulation (Phase 2) self-esteem emerged in the
expected directions for the control (r= .43, p< .05) and bogus
pipeline conditions (r=.16, ns).
A hierarchical multiple regression analysis was performed to
examine whether narcissism would be associated with lower
levels of Phase 2 self-esteem under bogus pipeline conditions. Prior
to conducting the analysis, the continuous variables were mean-
centered (Aiken & West, 1991). The main effect terms of experi-
mental condition (0 = control,1=bogus pipeline), narcissism, and
self-esteem (Phase 1) were entered on Step 1 and the two-way
interaction between experimental condition and narcissism was
entered on Step 2.
2
The results for this analysis are presented in
1
Data were collected for 78 undergraduates (7 men, 71 women) and preliminary
analyses were conducted using this sample. However, due to the small number of
male participants, final analyses were conducted using women only (N= 71). The
results were highly consistent between the full and women-only samples.
2
Preliminary analyses included all of the two-way interaction terms as well as the
three-way interaction term. These additional interaction terms did not approach
conventional levels of statistical significance and their inclusion did not influence the
two-way interaction of condition narcissism. As a result, these additional interac-
tion terms were trimmed from the final analysis in the interest of parsimony.
E.M. Myers, V. Zeigler-Hill / Journal of Research in Personality 46 (2012) 102–105 103
Author's personal copy
Table 2. This analysis was followed by the simple slopes tests recom-
mended by Aiken and West (1991) to probe interactions involving a
continuous variable. Simple slopes tests are used to determine the
degree of association between a predictor variable (experimental
condition) and a criterion variable (self-esteem level during Phase
2) at specific values of the moderator variable (one standard
deviation above and below the mean for narcissism). We used these
simple slopes tests to determine whether the association between
experimental condition and self-esteem was significantly different
from zero for those with high and low levels of narcissism.
The total regression model was significant (R
2
= .47, F[4, 66] =
14.53, p< .001) and the expected interaction between experimen-
tal condition and narcissism reached conventional levels of signif-
icance (b=.26, t[66] = 2.09, p< .05, d= .51). The predicted
values for this interaction are presented in Fig. 1. Simple slopes
tests revealed that the slope of the line representing the associa-
tion between experimental condition and self-esteem was negative
for those with high levels of narcissism (b=.39, t[66] = 3.03,
p< .01, d= .74) but that the slope of the line representing this
association for those with low levels of narcissism did not ap-
proach conventional levels of significance (b=.01, t[66] < 1, ns).
These results show that individuals with high levels of narcissism
reported lower levels of self-esteem under bogus-pipeline condi-
tions than control conditions, whereas the self-esteem levels re-
ported by those with low levels of narcissism did not differ
between the conditions. In other words, individuals with high lev-
els of narcissism reported lower levels of self-esteem when they
believed that the experimenter would know if they were lying. This
pattern suggests that narcissistic individuals are only willing to ad-
mit their relatively negative attitudes about themselves when they
believe that others will be able to tell that they are lying.
4. Discussion
The present study utilized the bogus pipeline technique in an
effort to elicit responses from participants that more accurately re-
flected their true feelings of self-worth. The results of this study
found that women with high levels of narcissism reported higher
levels of self-esteem in the control condition than the bogus pipe-
line condition. Further, narcissistic individuals reported feelings of
self-worth in the bogus pipeline condition that were lower than
those of individuals with low levels of narcissism. These results
provide compelling evidence for the idea that the grandiose feel-
ings of self-worth reported by narcissists may not be the entire
story concerning how they feel about themselves. That is, women
with narcissistic tendencies appear to report feelings of self-worth
that are more positive than their actual attitudes about them-
selves. Our results provide support for the psychodynamic mask
model of narcissism by showing that individuals with high levels
of narcissism reported lower levels of self-esteem in the bogus
pipeline condition than the control condition. Thus, we found sup-
port for the idea that narcissists inflate their feelings of self-worth
on self-report measures and that these overly positive self-evalua-
tions may disguise underlying feelings of self-loathing and inferi-
ority as suggested by Kohut (1966) and Kernberg (1975). These
findings are important because they suggest that the bogus pipe-
line procedure has the potential to allow researchers to briefly peel
back the grandiose facades that are worn by narcissistic individuals
and catch a glimpse of how these individuals actually feel about
themselves.
These results suggest that the positive feelings of self-worth
reported by individuals with high levels of narcissism may be
partially due to their attempts to convince themselves of their
own worth. The fact that narcissistic individuals claim to like
themselves may be at least somewhat helpful for them as they
strive to bolster and maintain their tenuous feelings of self-worth.
Another possibility is that narcissistic individuals may report posi-
tive feelings of self-worth in an attempt to present themselves to
others in a favorable manner. To put it another way, individuals
with narcissistic tendencies may intentionally portray themselves
as feeling more positively about themselves than they actually
do in order to influence how others perceive them. Narcissists
Table 2
Analysis regressing self-esteem (Phase 2) onto experimental condition, narcissism,
and self-Esteem (Phase 1).
Self-esteem (Phase 2)
Cumulative R
2
Increase in R
2
b
Step 1 .43
***
.43
***
Condition .20
*
Narcissism .06
Self-esteem (Phase 1) .62
***
Step 2 .47
***
.04
*
Condition narcissism .26
*
*
p< .05.
***
p< .001.
Fig. 1. Predicted values for Phase 2 self-esteem illustrating the interaction of
experimental condition and narcissism at one standard deviation above and below
the mean.
Table 1
Descriptive statistics for narcissism, self-esteem (Phase 1), and self-esteem (Phase 2).
Control condition Bogus pipeline condition t(df)p
N=35 N=36
MSDM SD
Phase 1
Narcissism 20.66 7.51 19.39 6.96 .74 (69) .46
Self-esteem (SSES) 3.68 .68 3.53 .71 .86 (69) .39
Phase 2
Self-esteem (RSES) 4.49 .43 4.25 .51 2.22 (69) .03
104 E.M. Myers, V. Zeigler-Hill / Journal of Research in Personality 46 (2012) 102–105
Author's personal copy
may engage in this form of impression management because of the
social benefits that are associated with being viewed as possessing
high self-esteem such as political competence (Zeigler-Hill &
Myers, 2009) or romantic desirability (Zeigler-Hill & Myers,
2011). These results also suggest the intriguing possibility that
individuals with high levels of narcissism may claim to possess
high levels of self-esteem in order to gain the sorts of social re-
wards (e.g., romantic desirability) that will actually bolster their
feelings of self-worth. It is important to note that these results di-
verge from those of earlier studies which suggested that the gran-
diosity of narcissists was largely a result of narcissists honestly
believing they are superior to others rather than trying to convince
themselves or impress others (e.g., Paulhus, 1998).
5. Limitations
Despite the fact that the present work suggests that revisiting
the bogus pipeline procedure may be useful for revealing ‘‘hidden’’
attitudes, there are several limitations associated with the present
study. First, the sample was comprised entirely of female college
students. Future research should explore whether the present re-
sults generalize to samples where men and members of the popu-
lation at large are well-represented. Second, it is impossible to rule
out alternative explanations for the present findings. Future stud-
ies should examine outcome variables other than self-esteem to
make certain that narcissistic individuals do not simply adopt
more moderate response tendencies under bogus pipeline condi-
tions. A third limitation is that narcissism scores were obtained
using self-report measures which leaves open the possibility of re-
sponse bias. Finally, future research could examine whether self-
reported levels of narcissism actually remain stable under bogus
pipeline conditions.
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. London: Sage.
American Psychiatric Association (2000). Diagnostic and statistical manual of mental
disorders (4th ed., Text revision). Washington, DC: Author.
Blascovich, J., & Tomaka, J. (1991). Measures of self-esteem. In J. P. Robinson, P. R.
Shaver, & L. S. Wrightsman (Eds.). Measures of personality and social
psychological attitudes (Vol. 1, pp. 115–160). New York: Academic.
Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler-Hill, V., Jordan, C. H., & Kernis, M.
H. (2008). Untangling the links between narcissism and self-esteem: A
theoretical and empirical review. Social and Personality Psychology Compass, 2,
1415–1439.
Boysen, G. A., Vogel, D. L., & Madon, S. (2006). A public versus private administration
of the implicit association test. European Journal of Social Psychology, 36,
845–856.
Emmons, R. A. (1987). Narcissism: Theory and measurement. Journal of Personality
and Social Psychology, 52, 11–17.
Gregg, A. P., & Sedikides, C. (2010). Narcissistic fragility: Rethinking its links to
explicit and implicit self-esteem. Self and Identity, 9, 142–161.
Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for
measuring state self-esteem. Journal of Personality and Social Psychology, 60,
895–910.
Jones, E. E., & Sigall, H. (1971). The bogus pipeline: A new paradigm for measuring
affect and attitude. Psychological Bulletin, 76, 349–364.
Jordan, C. H., Spencer, S. J., Zanna, M. P., Hoshino-Browne, E., & Correll, J. (2003).
Secure and defensive high self-esteem. Journal of Personality and Social
Psychology, 85, 969–978.
Kernberg, O. F. (1975). Borderline conditions and pathological narcissism. New York:
Jason Aronson.
Kohut, H. (1966). Forms and transformations of narcissism. Journal of the American
Psychoanalytic Association, 14, 243–272.
Morf, C. C., & Rhodewalt, F. (1993). Narcissism and self-evaluation maintenance:
Explorations in object relations. Personality and Social Psychology Bulletin, 19,
668–676.
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A
dynamic self-regulatory processing model. Psychological Inquiry, 12, 177–196.
Paulhus, D. L. (1998). Interpersonal and intrapsychic adaptiveness of trait self-
enhancement: A mixed blessing? Journal of Personality and Social Psychology, 74,
1197–1208.
Raskin, R. N., & Hall, C. S. (1979). A narcissistic personality inventory. Psychological
Reports, 45, 590.
Raskin, R. N., & Hall, C. S. (1981). The narcissistic personality inventory: Alternative
form reliability and further evidence of construct validity. Journal of Personality
Assessment, 45, 159–162.
Raskin, R., & Terry, H. (1988). A principal-components analysis of the Narcissistic
Personality Inventory and further evidence of its construct validity. Journal of
Personality and Social Psychology, 54, 890–902.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.
Rosenberg, M. (1986). Self-concept from middle childhood through adolescence. In
J. Suls & A. G. Greenwald (Eds.), Psychological perspectives on the self
(pp. 107–136). Hillsdale, NJ: Lawrence Erlbaum.
Zeigler-Hill, V. (2006). Discrepancies between implicit and explicit self-esteem:
Implications for narcissism and self-esteem instability. Journal of Personality, 74,
119–143.
Zeigler-Hill, V., & Jordan, C. (2010). Two faces of self-esteem: Implicit and explicit
forms of self-esteem. In B. Gawronski & B. K. Payne (Eds.), Handbook of implicit
social cognition: Measurement, theory, and applications (pp. 392–407). New York:
Guilford Press.
Zeigler-Hill, V., & Jordan, C. H. (2011). Behind the mask: Narcissism and implicit
self-esteem. In W. K. Campbell & J. Miller (Eds.), Handbook of narcissism and
narcissistic personality disorder: Theoretical approaches, empirical findings, and
treatment (pp. 101–115). Hoboken, NJ: Wiley.
Zeigler-Hill, V., & Myers, E. M. (2009). Is high self-esteem a path to the White
House? The implicit theory of self-esteem and the willingness to vote for
presidential candidates. Personality and Individual Differences, 46, 14–19.
Zeigler-Hill, V., & Myers, E. M. (2011). An implicit theory of self-esteem: The
consequences of perceived self-esteem for romantic desirability. Evolutionary
Psychology, 9, 147–180.
E.M. Myers, V. Zeigler-Hill / Journal of Research in Personality 46 (2012) 102–105 105
