Motivational surfeits and motivational deficits
The Unique Role of Spirituality in the Process of Growth Following Stress and Trauma
Mary Beth Werdel & Gabriel S. Dy-Liacco & Joseph W. Ciarrocchi & Robert J. Wicks & Gina M. Breslford
Published online: 30 April 2013 # Springer Science+Business Media New York 2013
Abstract This study investigated the unique contributions and moderating effects of pos- itive and negative experiences of spirituality in predicting variance in stress-related growth and positive affect over and above the variance explained by the domains of the Five-Factor Model of Personality, social support, perceived stress, and gender in a sample of 109 male and 320 female volunteers. Responses were analyzed using hierarchical linear regression. Results indicated that faith maturity predicted unique variance in stress-related growth after controlling for the influence of personality and gender; faith maturity and spiritual struggle predicted significant unique additional variance in positive affect over and above the variance predicted by personality, social support, and stress-related growth; and spiritual struggle moderated the relationship between stress-related growth and positive affect. The implications of these results are discussed in light of Park’s (Journal of Social Issues 61:707– 729, 2005) model of religion as a meaning-making framework.
Keywords Spirituality. Stress-relatedgrowth .Faithmaturity. Spiritual struggle . Incremental validity
The idea that positive psychological changes may occur following stress or trauma is a concept that is well documented. There are a number of terms used to describe the phenomenon. The term posttraumatic growth, coined by Calhoun and Tedeschi (2006), conceptualizes growth following stress and trauma as fitting into one of three categories: changes in the perception of self; changes in relating to others; and philosophical changes
Pastoral Psychol (2014) 63:57–71 DOI 10.1007/s11089-013-0538-4
M. B. Werdel (*) Graduate School of Religion and Religious Education, Fordham University, Bronx, NY 10458, USA e-mail: email@example.com
G. S. Dy-Liacco School of Psychology and Counseling, Regent University, Virginia Beach, VA, USA
J. W. Ciarrocchi : R. J. Wicks Pastoral Counseling Department, Loyola University Maryland, Columbia, MD, USA
G. M. Breslford University of Pennsylvania Harrisburg, Harrisburg, PA, USA
of priorities, appreciations, and spirituality (Calhoun and Tedeschi 2006). Posttraumatic growth is the most widely used term in the literature. Other closely related terms include ‘stress-related growth’ (Park et al. 1996), ‘adversarial growth’ (Linley and Joseph 2004) and ‘benefit-finding’ (Tennen and Affleck 1998).
Not everyone who experiences stress or trauma will experience posttraumatic growth. In an attempt to understand why, the empirical research provides evidence that posttraumatic growth has a correlational relationship with a number of environmental and personal factors. For instance, research indicates that a heightened level of distress is a key factor in initiating the process of posttraumatic growth (Park 2005; Tedeschi and Calhoun 2004). If too little distress is experienced, a person post-stress or -trauma may not initiate the cognitive process necessary to experience growth (Levine et al. 2008).
A second factor that is important to posttraumatic growth is social support. Research suggests that when a person discloses information about a trauma in a supportive relation- ship, a person receives emotional support, informational feedback about the trauma, and tangible assistance which over time may lead a person to view others more positively and increase their self-confidence (Swickter and Hittner 2009). Additionally, close relationships may help people discover new ways of perceiving or thinking about the world or new coping methods (Tedeschi and Calhoun 2004).
A third factor relative to growth is personality (Calhoun and Tedeschi 2006). Personality contributes to the way a person’s experiences are perceived (DeNeve and Copper 1998). Evidence suggests that people who score high on levels of the personality factors of extraversion and openness to new experiences may be more likely to experience growth after trauma than people who have lower levels of the same factors (Calhoun and Tedeschi 1999). (For a more detailed discussion of the variables of perceived stress, social support, and personality, see Calhoun and Tedeschi 2006.)
While the aforementioned variables explain a significant amount of the variance in models of growth, there may be other variables that contribute to the process of growth that are important to uncover for both research and clinical purposes. However, knowing that perceived stress, social support, and personality have evidenced relationships with posttraumatic growth, these factors should be considered in future studies of growth so as to increase parsimony in the literature.
The role of spirituality
The conversation that posttraumatic growth researchers have engaged in the psychological literature is one that religious and spiritual texts have long discussed. People often turn to religion with questions that result from the experience of suffering (Bemporad 2005). It is not surprising that a number of religious and spiritual variables have been examined in regards to their relationship with growth. Research suggests that posttraumatic growth is linked to religious and spiritual practices (Cadell et al. 2003), prayer (Levine et al. 2009), positive religious coping (Proffitt et al. 2007), negative religious coping (Pargament et al. 2004), and religious orientation (Calhoun et al. 2000).
Religiosity and spirituality are multidimensional variables that incorporate cognitions, sentiments, and behaviors including both positive and negative aspects of an individual’s relationship with the Divine. There are a number of relationships between religiosity, spirituality, and growth after stress and trauma that have not been explored. Additionally, spirituality has primarily been explored without a model that considers the potential overlap of personality and spiritual variables. Furthermore, the studies heretofore have not consid- ered the potential of spirituality as a moderator variable. Two specific spiritual variables worthy of attention are faith maturity and spiritual struggles.
58 Pastoral Psychol (2014) 63:57–71
Faith maturity Faith maturity captures the “degree to which a person embodies the priori- ties, commitments, and perspectives of vibrant life-transforming faith” (Benson et al. 1993, p. 3). To date there have been no studies that have examined the relationship of faith maturity to posttraumatic growth. However, previous research suggests that faith maturity is a robust indicator of positive psychological adjustment (Salsman and Carlson 2005). Additionally, research suggests that faith maturity correlates with variables that also correlate with posttraumatic growth. For example, in a study of 251 male and female undergraduate students, faith maturity was correlated with intrinsic religious orientation, a variable with known relationships to growth. Therefore, it is reasonable to propose that faith maturity will also correlate with posttraumatic growth and may predict unique variance in posttraumatic growth as well. The results of research on posttraumatic growth and faith maturity may distinguish differences that assessing religious and spiritual practices or religious coping style do not. An individual may attend church services, pray, or attempt to use religion to cope, but any relationship found between such activities and growth does not explain the relationship if the use of a mature faith versus an immature faith matters when engaging in such religious and spiritual practices and coping. It seems reasonable to suppose that a person with a higher maturity of faith would be able to hold the paradox of a loving God and tragic and painful life experiences in a way that allows new positive meaning about their self, about others, or about their God to be discovered.
Spiritual struggle A second aspect of spirituality that warrants more attention is the rela- tionship that exists between growth and spiritual struggle. Spiritual struggle is an aspect of negative religious coping concerned with whether or not a person feels abandoned by, punished by, or angry with God. Research suggests that there are profound negative mental and physical health implications associated with spiritual struggle. For instance, viewing God as punishing, as well as expressions of spiritual discontent, have been linked to poorer physical health, decreased quality of life, and greater depression (Koenig et al. 2004; Koenig et al. 1998). Negative religious coping has even been associated with increased likelihood of mortality in samples of medically ill patients (Koenig et al. 1998; Oxman et al. 1995). Whereas secure relationships with God are thought to aid in the coping process following stress and trauma, the presence of a spiritual struggle is thought to exacerbate distress (Ano and Vasconcelles 2005).
In a two-year longitudinal study consisting of 236 medically ill elderly patients, Pargament et al. (2004) examined the relationship between negative religious coping and posttraumatic growth. Results indicated that participants who evidenced high degrees of negative religious coping at both the start of the study and two years later also evidenced significant declines in quality of life, declines in functional status, and increased levels of depressed moods.
Given the profoundly negative relationship that exists between negative religious coping and posttraumatic growth, research that includes spiritual struggle may be useful. One way that the construct of spiritual struggle may be helpful is in understanding the conflicting data that exist in the literature as to whether an association exists between posttraumatic growth and positive affect. While time has been indicated as a potential condition, there may be other intrapersonal moderators of the relationship such as the presence of a spiritual struggle. The psychological literature on emotions provides evidence to support the idea that negative emotional experiences have a stronger, more lasting effect on subjective measures of well- being than positive experiences (Baumeister et al. 2001). With this evidence, it is hypoth- esized that spiritual struggle is negatively related to growth following stress and trauma. Furthermore, the presence of a spiritual struggle may be one condition that would affect the relationship between growth and positive affect.
Pastoral Psychol (2014) 63:57–71 59
No prior studies have examined how negative religious coping may interact with the process of posttraumatic growth. An answer would lead the field further along from the principal question of whether or not spirituality relates to growth and closer to answering the question of how spirituality relates to growth. If spiritual struggles are an obstacle to the experience of growth, then a clinical implication of understanding the role that spiritual struggle plays in relationship to growth is that it may lead therapists and pastoral counselors to assess for spiritual struggles when working with people experiencing stressful life events that would otherwise go unexplored and untreated.
The incremental validity paradigm in psychological research
It is imperative that personality variables be included in psychological research of spiritual variables. Piedmont (2005) suggested the use of the Incremental Validity Paradigm in the psychological research of religious and spiritual variables. This paradigm considers the role of personality as a potential mediator of relationships between spirituality and outcome variables (Piedmont 2005). There is evidence of a significant relationship be- tween personality and religious and spirituality variables (Saroglou 2002). Including personality variables in research studies is a way to demonstrate the distinctiveness of spiritual variables’ contributions by providing evidence that spiritual variables capture something unique about outcome variables not already understood by models of person- ality (Piedmont 2005). The use of the incremental validity paradigm provides evidence for religious and spiritual variables as more than a “religification” (Van Wicklin 1990) of psychological constructs.
Incremental validity is tested through mediational analysis. Baron and Kenny (1986) explain that a variable functions as a mediator “to the extent that it accounts for the relationship between the predictor and the criterion” (p. 1176). A predictor variable may correlate with both an outcome variable and a proposed mediator variable. When the relationship between the predictor and the outcome is reduced to zero or a nonsignificant relationship, after controlling for their common correlation with the proposed mediator, there is evidence for a mediator variable. However, if the correlation is not reduced to zero and remains statistically significant, then there is evidence of the unique contributions of the predictor to the outcome variable.
This study proposes that growth following stress and trauma will be significantly positively correlated with faith maturity, positive affect, and social support with a medium effect size. Additionally, stress-related growth will be significantly negatively correlated with perceived stress with a small effect size. As well, this study proposes that the domains of the Five- Factor Model of Personality, social support, and perceived stress will predict stress-related growth. Faith maturity and spiritual struggle will explain significant unique variance in stress-related growth over and above the variance explained by the domains of the Five- Factor Model of Personality, social support, and perceived stress.
Also, this study proposes that the domains of the Five-Factor Model of Personality, stress- related growth, social support, and perceived stress will predict positive affect. Faith maturity and spiritual struggle will explain significant unique variance in positive affect over and above the variance explained by the domains of the Five-Factor Model of Personality, social support, and perceived events. Furthermore, spiritual struggle will mod- erate the effects of stress-related growth on positive affect.
60 Pastoral Psychol (2014) 63:57–71
This study is an analysis of unanalyzed relationships in a data set that was partially collected for a Positive Psychology and Spirituality Research Group at a mid-Atlantic university. A total of 255 participants in the data set completed all measures of interest. An additional 174 participants were recruited. The sample is predominantly female (74.6 %), Caucasian (80.1 %), and Christian (56.8 %). Participants’ ages range from 17 to 80 years (mean = 42.18). A total of 49.6 % are married, 35.9 % are single, and 12.9 % are divorced. The sample is highly educated; 80.5 % have at least a college degree, and 50.8 % have a graduate degree.
Demographic questionnaire The Demographic Questionnaire developed by the second and fourth author captures age, gender, marital status, ethnic group, and educational level.
Faith maturity scale (FMS) short form Developed by Benson et al. (1993), the FMS scale captures values and behavioral manifestations of faith (Tisdale, 1999). The FMS measures two domains: Faith Maturity Vertical (FMS-V, the emphasis that an individual places on his or her relationship with the transcendent) and Faith Maturity Horizontal (FMS-H, the emphasis that a person places on serving others) (Piedmont and Nelson 2001). The scale is an 11-item, 7-point Likert-type scale with responses ranging from 1 = Never true to 7 = Always true. Eight items make up the FMS-Vand three items make up the FMS-H.
Brief religious and spiritual coping scale (RCOPE)/spiritual struggle The RCOPE is a 63- item measure with 21 subscales that “assess the degree to which [people make] use of various religious methods of coping” (Pargament et al. 2004, p. 716). The Brief RCOPE (Pargament et al. 1998) is a shortened version of the instrument that consists of 14 items and 2 subscales, each subscale containing 7 items. The Negative Religious Coping subscale is “an expression of a less secure relationship with God” (Pargament et al. 1998, p. 712), indicated by punishing God appraisals, interpersonal religious discontent, demonic appraisals, spiritual discontent, and questioning God’s powers. Items are answered on a 4-point frequency scale, and responses range from 0 = Not at all to 3 = A great deal. For the purpose of this study, a composite variable, Spiritual Struggle, was created from two items of the Negative Religious Coping Subscale of the RCOPE. The two items used to assess spiritual struggle are: “I feel that God is punishing me for my sins or lack of spirituality” and “I wonder whether God has abandoned me.” The items are answered on a 4-point frequency scale with responses of: 1 = A great deal, 2 = Quite a bit, 3 = Somewhat, and 4 = Not at all. A total score is derived by summing the two items, with high scores indicating high levels of struggle. The composite variable of spiritual struggle used in this study is supported by the work of Piedmont et al. (2006). See article for more information.
Perceived stress scale 4-item version (PSS) Developed by Cohen et al. (1983), the PSS is a measure of the degree to which a person appraises situations in life as stressful. A 4-item short-form measure was developed from the original scale. Two of the items have a positive valence and two have a negative valence. The items are answered on a 5-point Likert-type scale from 0 = Never to 5 = Very often. The positive valence items are reverse coded, and the total score is then derived by summing all 4 items.
Pastoral Psychol (2014) 63:57–71 61
Perceived social support- friends (PSS-Fr) Developed by Procidano and Heller (1983), the Perceived Social Support—Friends Scale is a 20-item scale that was designed to measure “the extent to which an individual perceives that his/her needs for support, information, and feedback are fulfilled by friends” (p. 2). Participants read declarative statements and respond by indicating, Yes, No, or I don’t know. A total score is derived by summing all the items, with a high score indicating high levels of perceived social support.
50-item IPIP-NEO inventory Goldberg et al. (2006) developed the short form of the questionnaire to capture the domains of the Five-Factor Model of Personality (FFM): Neuroticism, Extraversion, Openness, Agreeableness, and Consciousness (the degree to which one has personal organization). The 50-item IPIP-NEO is a self-report measure to which individuals respond with Likert-type item responses ranging from 1 = Very inaccurate to 5 = Very accurate.
Stress-related growth scale 15-item short form (SRGS) Park et al. (1996) developed the SRGS, a 50-item self-report measure, to capture positive changes following an identified stressful event in a number of areas: personal resources, social relationships, life philosophy, and coping skills. Items are answered on a 3-point frequency scale: 0 = Not at all, 1 = Somewhat, and 2 = A great deal. The 15-item short form of the SRGS, developed by Cohen et al. (1998), consists of the 15 highest-loading items of the 50-item measure.
Midlife development inventory—affect scales (MIDI) The Midlife Development Inventory was developed by Brim and Featherman (as cited in Mroczek and Kolarz 1998) for a national survey that studied midlife in the United States in regard to health and well-being. The survey contains a number of measures related to aspects of midlife and aging. Participants respond with a 5-point Likert-type scale ranging from 1 = All of the time to 5 = None of the time.
Participants completed self ratings on all measures through Psychdata, the online survey software. First, correlations were used to determine a relationship between the variables. Second, a series of hierarchical multiple regressions were used to determine if positive and negative experiences of spirituality explain unique variance in stress-related growth, without mediation by the domains of the Five-Factor Model of Personality. Third, a series of hierarchical multiple regressions were used to determine (a) if positive and negative expe- riences of spirituality explain variance in positive affect without mediation by the domains of the Five-Factor Model of Personality and (b) if the relationship between stress-related growth and positive affect is moderated by levels of negative spirituality.
The data were reviewed for possible gender differences. Independent sample t-tests for gender differences for scores for all the scales indicated a significant difference between males and females on four scales. As can be seen in Table 1, females scored significantly higher on levels of stress-related growth, agreeableness, conscientiousness, and social support. The result of the t-test indicating gender differences in the Stress-Related Growth Scale shown in Table 1 is congruent with previous research that suggests women tend to report more posttraumatic growth than men (Tedeschi and Calhoun 1996). Table 2 displays
62 Pastoral Psychol (2014) 63:57–71
the means, standard deviations, and acceptable alpha reliability coefficients for all predictor and criterion variables. All of the alpha reliability coefficients are acceptable.
As predicted, faith maturity and positive affect correlate positively with stress-related growth. Social support correlates positively with stress-related growth with an effect size that approaches medium. There is a significant negative correlation between perceived stress and stress-related growth with a small effect size. As can be seen in Table 3, the data support that positive affect and spiritual struggle have a significant negative correlation with a moderate effect size.
Table 1 Independent Sample t-tests for Gender Differences on Scale Scores
Variable Gender N M SD t value
1. Growth 3.87***
Females 317 36.42 6.51
Males 111 33.52 7.54
2. Agreeableness 2.15**
Females 317 39.61 5.67
Males 111 38.21 6.45
3. Conscientiousness 3.07**
Females 317 38.52 6.52
Males 111 36.24 7.30
4. Social Support 3.73***
Females 317 14.43 3.94
Males 111 12.73 4.71
N =429 for all scales
*p <. 05, **p<.01, ***p <. 001
Table 2 Descriptive Statistics
Variable M SD Alpha
Stress-related growth 35.70 6.90 .93
Positive affect 20.50 4.17 .90
Neuroticism 25.10 8.23 .88
Extraversion 36.20 7.00 .82
Openness 40.80 5.81 .71
Agreeableness 39.30 5.90 .78
Conscientiousness 37.90 6.79 .84
Faith maturity 54.50 15.00 .92
Spiritual struggle 2.47 .97 .74
Perceived stress 7.65 3.56 .78
Social support 14.00 4.21 .89
N for all scales = 429
Pastoral Psychol (2014) 63:57–71 63
C or re la tio
on g P re di ct or
an d O ut co m e V ar ia bl es
. 11 .
1. S tr es s- re la te d gr ow
2. P os iti ve
af fe ct
3. H op
e .4 1*
** .5 8*
4. N eu ro tic is m
.5 0* **
5. E xt ra ve rs io n
** .3 4*
** .4 6* **
6. O pe nn es s
* .1 4*
** .2 6* **
.2 4* **
7. A gr ee ab le ne ss
** .3 3*
** .3 0* **
.2 3* **
.2 1* **
8. C on sc ie nt io us ne ss
** .4 3*
** .5 2* **
.1 9* **
.3 9* **
9. F ai th
m at ur ity
** .3 5*
** .3 3* **
.2 1* **
.2 0* **
.4 0* **
.2 3* **
10 . S pi ri tu al
st ru gg le
.3 7* **
−. 11 **
11 . P er ce iv ed
st re ss
.4 6* **
12 . S oc ia l su pp or t
** .3 7*
** .4 7* **
.3 5* **
.2 1* **
.3 3* **
.2 3* **
** −. 21
N = 42
if ic an t at
th e p < . 00
1 le ve l
** p < .0 1,
< . 00
64 Pastoral Psychol (2014) 63:57–71
Table 4 presents the first hierarchical regression of stress-related growth on the Five-Factor Model of Personality, perceived stress, social support, faith maturity, and spiritual struggle. As can be seen in Table 4, personality is a significant predictor of growth on the first step. Perceived stress and social support are entered in a forward fashion on the second step of the model after the variances of personality and gender are removed from the equation. These predictors also significantly predict growth. However, an inspection of the beta weights shows that social support accounts for the entire additional 2 % (ΔR2=.02) of significant unique variance in the model, β=.14, p<.01 (social support) and β=−.04 p>.05 (perceived stress).
When the numinous variables are entered in a forward fashion on step 3 of Table 4, after the effects of personality, social support, and perceived stress are removed from the equation, the numinous variables are a significant predictor of stress-related growth. The numinous variables accounted for an additional 8 % (ΔR2=.08) of the variance in stress-related growth. However, as can be seen in Table 4, an inspection of the beta weights shows that faith maturity explains significant unique variance β=.31, p<.001 but spiritual struggles does not β=.05, p>.05. Contrary to the hypothesis, the correlation analysis revealed that spiritual struggle did not correlate with stress-related growth and so it is expected that spiritual struggle would not explain any variance in stress-related growth in a regression model, as one of the assumptions of regression is that the predictor and outcome variables are correlated. Table 4, Model 4, step 3 shows that the beta weight for faith maturity was medium sized, β=.31, p<.001.
Table 5 presents the second hierarchical regression of positive affect on the Five-Factor Model of Personality, stress-related growth, perceived stress, social support, faith maturity, and spiritual struggle. As can be seen in Table 5, as predicted, personality is a significant predictor of positive affect. Also, as predicted, when stress-related growth, perceived stress, and social support are entered in a forward fashion on the second step of the model after the
Table 4 Final Hierarchical Regression Model of Stress-related Growth on the Five-actor Model of Person- ality, Perceived Stress, Social Support, Faith Maturity and Spiritual Struggle
Predictor R2 ΔR2 F Change df B T β
Step 1 .13 .13 10.55 *** 6,421
Neuroticism −.03 −.70 −.04 Extraversion .08 1.55 .08
Openness .01 .06 .01
Agreeableness .01 .07 .01
Conscientiousness .07 1.31 .07
Gender −1.89 −2.70 −.12* Step 2 .15 .02 4.44** 2,419
Perceived stress −.09 −.85 −.04 Social support .24 2.89 .14**
Step 3 .23 .08 21.38***
Faith maturity .14 6.39 .31***
Spiritual struggle .37 −1.09 .05
*p<.05, **p<.01, ***p<.001
Pastoral Psychol (2014) 63:57–71 65
variance of personality is removed from the equation, the predictors also significantly predict positive affect and explain an additional 9 % of the variance. As predicted again, when the numinous variables are entered in a forward fashion on step 3, after the effects of personality, stress-related growth, social support, and perceived stress were removed from the equation, the numinous variables were a significant predictor of positive affect. The numinous vari- ables explain an additional 1 % of the variance (ΔR2=.01) in positive affect. An inspection of the beta weights shows that faith maturity and spiritual struggle explain significant unique variance in positive affect.
As indicated in Table 5, there is a significant interaction effect between levels of spiritual struggle and levels of growth on positive affect indicating that, as hypothesized, spiritual struggle moderates the effects of stress-related growth on positive affect. Graphing the interaction (see Fig. 1) reveals that individuals who score high on spiritual struggle and low on stress-related growth score low on positive affect, and individuals who score low on spiritual struggle and high on stress-related growth score high on positive affect. Individuals who score high on struggle and medium on growth experience a decrease in positive affect. Yet, people who score low on struggle and medium on growth experience an increase in positive affect. This may imply that those who have found some positive meaning in a stressful life event, but who continue to struggle spiritually, will most likely experience a dip in positive emotion. However, those who have found some positive meaning in a stressful life event but do not experience a spiritual struggle will most likely experience an increase in positive emotions.
Table 5 Final Hierarchical Regression Model of Positive Affect on the Five Factor Model of Personality, Stress-related Growth, Perceived Stress, Social Support, Faith Maturity, and Spiritual Struggle
R2 ΔR2 F Change df B T β
Step1 .49 .49 82.55*** 5,423
Neuroticism −.20 −9.40 −.40*** Extraversion .06 2.65 .09**
Openness .02 .69 .02
Agreeableness −.04 −1.59 −.06 Conscientiousness .05 2.36 .09*
Step 2 .58 .09 31.90*** 3,420
Stress-related growth −.01 −1.50 −.01 Social support .06 1.65 .06
Perceived stress −.29 −6.53 −.24*** Step 3 .59 .01 5.80** 2,418
Faith maturity .03 2.60 .10**
Spiritual struggle −2.12 −2.81 −.49** Step4 .60 .01 5.77** 1,417
Spiritual struggle X growth .05 2.40 .45**
*p<.05 **p<.01, ***p<.001
66 Pastoral Psychol (2014) 63:57–71
The results suggest that the spirituality variables in this study have both positive and negative associations with positive psychological constructs, similar to previous research suggesting that positive forms of spirituality relate in positive ways to aspects of psycho- logical flourishing and negative forms of spirituality relate in negative ways to aspects of psychological flourishing (Pargament et al. 1998). Experiences of growth after stressful life events have both positive and negative associations with aspects of a person’s spirituality. The positive emotions that can be experienced as a result of climbing life’s wreckage are dependent upon individual personality factors, the presence of social support, and, in this sample, to a greater degree, the presence of a mature faith and the absence of a belief in a punishing and/or abandoning God image. This sample suggests that contrary to previous suggestions, people are not merely sadder and wiser (Janoff-Bulman 1992) as a result of stressful life events but that increased levels of stress-related growth are predictive of increased levels of positive affect. The presence of a mature faith has an additive effect on a person’s experience of positive affect. The relationship between stress-related growth and positive affect, however, is complicated by the level of spiritual struggle a person experi- ences. While focusing on increasing a person’s faith maturity may contribute to the process of stress-related growth, the factor that contributes the most to the experience of positive affect is the degree to which a person feels abandoned or punished by God.
There are a number of clinical implications of this study. First, the data suggest that in this sample spiritual and religious variables are the factors that contribute most to the predictive
Fig. 1 Moderation Effects of Spiritual Struggle on Stress-Related Growth and Positive Affect
Pastoral Psychol (2014) 63:57–71 67
outcome of stress-related growth. Therefore, there is clinical relevance to assessing for spirituality in the initial intake process and throughout the clinical experience. Spiritual assessments that specifically assess a person’s faith maturity and the degree to which he or she is experiencing a spiritual struggle may be particularly useful when working with a person who has experienced a stressful life event. Beyond formal assessments of spirituality, it may be helpful for clinicians to understand not only the specific religious or spiritual faith tradition of their client, but how the client has personally integrated their faith tradition into their meaning system and the ways in which a person incorporates such meaning into the behaviors of their daily lives.
Second, the additive effect of faith maturity on a positive affect implies that the maturity of a person’s faith aids in the process of rebuilding life assumptions shattered by stressful and traumatic life events. Therefore, clinicians who are working with a person who has experienced a stressful or traumatic life event, and who has been identified as spiritual or religious through various forms of spiritual assessments, may find benefits in exploring the ways in which a person engages positively with their religious and spiritual traditions. It may also be beneficial to help the client increase their awareness of the connection between their spiritual framework and their search for meaning.
Lastly, the strong negative relationship between spiritual struggle and positive affect implies that feelings of abandonment and punishment by God have strong negative effects on psychological indicators of mental health and well-being. Clinicians should be aware that the cognitive restructuring of negative images of God, such as those articulated by the definition of a spiritual struggle, may alleviate the emotional pain from religious sources and thus have positive effects on a person’s experience of positive emotions. Therefore, address- ing a spiritual struggle, while it certainly has religious themes, may have strong implications for psychological well-being and so may be clinically necessary to address within the clinical hour. The absence or presence of a spiritual struggle for people with medium levels of growth makes the difference between a person experiencing an increase or a decrease in positive affect. Therapists need to be qualified to work specifically with negative spiritual themes with clients who indicate a spiritual identity.
There are a number of limitations to the study. First, only self-report data were used. Also, participants were a volunteer sample of convenience and not a random sample. From the demographic information it is known that the participants in the sample were highly educated and highly spiritual and that the majority were Caucasian (82.1 %). Generalization of results to a less well-educated, less spiritual, and more racially diverse sample is not possible from this data. Additionally, the study is a correlational design; causality cannot be assumed or implied. Finally, the study had a cross-sectional design. The research design cannot provide information on the process of the variables over time.
Implications for future research
The results of the study have implications for future research in the area of spirituality and stress-related growth. First, this study was one of the first to examine spirituality as a potential moderator of stress-related growth on positive mental health outcomes. Results suggest that negative spirituality does serve to moderate growth in positive affect. There may be a need for the examination of other possible spirituality variables as moderator variables of the relationship between stress-related growth and mental health outcomes. This study did
68 Pastoral Psychol (2014) 63:57–71
not ask participants to disclose details of their stressful life experience. It is uncertain if the type of trauma or the level of exposure a person had to a traumatic event had an impact on the results of this study. Future research needs to examine the possibility of influences that type of trauma and level of exposure to trauma have on the relationship between spirituality and stress-related growth. A third question that is uncertain from this study is the influence of time. Longitudinal studies on stress-related growth and spirituality variables would allow researchers to understand better the trajectory of stress-related growth.
How one comes to find the face of God in suffering may be a theological question, but the results of this study suggest that there are significant psychological ramifications. While psychology cannot and should not seek to answer the question of God’s existence in suffering, it can andmust understand that the theodicy a person holds can alter a person’s perception of life after stress and trauma and, consequently, their emotional experience of life. At best, experi- ences of spirituality provide meaning and purpose in a world filled with existential anxiety produced by life’s unanswerable questions (Crews 1986). At worst, experiences of spirituality fuel loneliness, guilt, and shame. While positive aspects of spirituality may be a beneficial resource for transcending life stressors (Kim and Seidlitz 2002), negative aspects of spirituality may interact with stress-related growth and thus a person’s experience of positive affect. Both aspects of spirituality have implications on mental health, and as such researchers, clinicians, and pastoral counselors need be aware of this.
Spirituality and the paradox of stress-related growth appear not to be unrelated constructs but rather ones that intersect in important ways. While perhaps once it could be argued that psychology and spirituality were uniquely separate ways of examining the world, research such as that presented in this study continues to suggest a more inclusive frame of reference. Stressful life events encourage and perhaps even demand the need to initiate the meaning- making process in order to decrease painful feelings of psychological distress (Park 2005). For individuals who have a spiritual identity, every aspect of life, including those uniquely positive, those uniquely negative, and all the many in between, have the possibility of being construed as a spiritual experience; “religion, whatever it is, is a man’s total reaction upon his life” (James 1997, p. 45). Therefore, as supported by the results of this study, for a person who claims a religious or spiritual identity, the religious and spiritual influences in the research of psychological phenomena and in the practice of clinical work cannot and should not be ignored.
Ano, A. G., & Vasconcelles, E. B. (2005). Religious coping and psychological adjustment to stress: a meta- analysis. Journal of Clinical Psychology, 61, 461–480. doi:10.1002/jclp.20049.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychol- ogy, 51, 1173–1182. doi:10.1037/0022-35126.96.36.1993.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5, 323–370.
Bemporad, J. (2005). Suffering. In L. Jones, M. Eliade, & C. J. Adams (Eds.), Encyclopedia of religion (pp. 8804–8809). Detroit: Macmillan.
Benson, P. L., Donahue, M. J., & Erikson, J. A. (1993). The Faith Maturity Scale: conceptualization, measurement, and empirical validation. Research in the Social Scientific Study of Religion, 5, 1–26.
Pastoral Psychol (2014) 63:57–71 69
Cadell, S., Regehr, C., & Hemsworth, D. (2003). Factors contributing to posttraumatic growth: a proposed structural equation model. The American Journal of Orthopsychiatry, 73, 279–287.
Calhoun, L. G., & Tedeschi, R. G. (1999).Facilitating posttraumatic growth: A clinician’s guide. Mahwah: Erlbaum. Calhoun, L. G., & Tedeschi, R. G. (2006). The foundations of posttraumatic growth: An expanded framework.
In L. G. Calhoun & R. G. Tedeschi (Eds.), Handbook of posttraumatic growth: Research and practice (pp. 1–23). Mahwah: Erlbaum.
Calhoun, L. G., Cann, A. G., Tedeschi, R. G., & McMillan, J. (2000). A correlational test of the relationship between posttraumatic growth, religion, and cognitive processing. Journal of Traumatic Stress, 13, 521– 527. doi:10.10231/A:1007745627077.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396.
Cohen, L. H., Hettler, T. R., & Pane, N. (1998). Assessment of posttraumatic growth. In R. G. Tedeschi, C. L. Park, & L. G. Calhoun (Eds.), Posttraumatic growth: Positive changes in the aftermath of crisis (pp. 22– 42). Mahwah: Erlbaum.
Crews, C. F. (1986). Ultimate questions: A theological primer. Mahwah: Paulist Press. DeNeve, K. M., & Copper, H. (1998). The happy personality: a meta-analysis of 137 personality traits and
subjective well-being. Psychological Bulletin, 124, 197–229. doi:10.1037/0033-2909.124.2.197. Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C. R., et al. (2006). The
International Personality Item Pool and the future of public domain personality measures. Journal of Research in Personality, 40, 84–96.
James, W. (1997). The variety of religious experience. New York: Touchstone. Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. New York: Free Press. Kim, Y., & Seidlitz, L. (2002). Spirituality moderates the effect of stress on emotional and physical adjustment.
Personality and Individual Differences, 32, 1377–1390. doi:10.1016/S0191-8869(01)00128-3. Koenig, H. G., Pargament, K. I., & Nielsen, J. (1998). Religious coping and health status in medically ill
hospitalized older adults. The Journal of Nervous and Mental Disease, 186, 513–521. Koenig, H. G., George, L. K., & Titus, P. (2004). Religion, spirituality, and health in medically ill hospitalized
older patients. Journal of American Geriatrics Society, 52, 554–562. doi:10.1111/j.1532- 545.2004.52161.x.
Levine, S. Z., Laufer, A., Hamama-Raz, Y., Stein, E., & Solomon, Z. (2008). Posttraumatic growth in adolescence: examining the components and relationships with PTSD. Journal of Traumatic Stress, 21, 492–496. doi:10.1002/jts.20361.
Levine, E. G., Aviv, C., Yoo, G., Ewing, C., & Au, A. (2009). The benefits of prayer on mood and well-being of breast cancer survivors. Supportive Care in Cancer, 17, 295–306. doi:10.1007/s00520-008-0482-5.
Linley, P. A., & Joseph, S. (2004). Positive change following trauma and adversity: a review. Journal of Traumatic Stress, 17, 11–21.
Mroczek, D. K., & Kolarz, C. M. (1998). The effect of age on positive and negative affect: a developmental perspective on happiness. Journal of Personality and Social Psychology, 75, 1333–1349.
Oxman, T. E., Freeman, D. H., Jr., & Manheimer, E. D. (1995). Lack of social participation or religious strength and comfort as risk factors for death after cardiac surgery in the elderly. Psychosomatic Medicine, 57, 5–15.
Pargament, K. I., Smith, B., Koenig, H. G., & Perez, L. (1998). Patterns of positive and negative religious coping with major life stressors. Journal for the Scientific Study of Religion, 37, 710–724.
Pargament, K. I., Koenig, H. G., Tarakeshwar, N., & Hahn, J. (2004). Religious coping methods as predictors of psychological, physical, and spiritual outcomes among medically ill elderly patients: a two-year longitudinal study. Journal of Health Psychology, 9, 713–730. doi:10.1177/1359105304045366.
Park, C. L. (2005). Religion as a meaning-making framework in coping with life stress. Journal of Social Issues, 61, 707–729. doi:10.1111/j.1540-4560.2005.00428.x.
Park, C. L., Cohen, L. H., & Murch, R. (1996). Assessment and prediction of stress-related growth. Journal of Personality, 64, 71–105. doi:10.1111/j.1467-6494.1996.tb00815.x.
Piedmont, R. L. (2005). The role of personality in understanding religious and spiritual constructs. In R. F. Paloutzian & C. L. Park (Eds.), Handbook of the psychology of religion and spirituality (pp. 253–273). New York: Guildford Press.
Piedmont, R. L., & Nelson, R. (2001). A psychometric evaluation of the short form of the Faith Maturity Scale. Research in the Social Scientific Study of Religion, 12, 165–184.
Piedmont, R. L., Mapa, T. A., & Williams, J. E. G. (2006). A factor analysis of the Fetzer/NIA Brief Multidimensional Measure of Religiousness/Spirituality (MMRS). Research in the Social Scientific Study of Religion, 17, 177–196.
Procidano, M., & Heller, K. (1983). Measures of perceived social support from friends and from family: three validation studies. American Journal of Community Psychology, 11, 1–24. doi:10.1007/BF00898416.
70 Pastoral Psychol (2014) 63:57–71
Proffitt, D., Cann, A., Calhoun, L. G., & Tedeschi, R. G. (2007). Judeo-Christian clergy and personal crisis: religion, posttraumatic growth, and well being. Journal of Religion and Health, 46, 219–231. doi:10.1007/s10943-006-9074-1.
Salsman, J. M., & Carlson, C. (2005). Religious orientation, mature faith, and psychological distress: elements of positive and negative associations. Journal for the Scientific Study of Religion, 44, 201–209. doi:10.1111/j.1468-5906.2005.00276.x.
Saroglou, V. (2002). Religion and the five factors of personality: a meta-analytic review. Personality and Individual Differences, 32, 15–25. doi:10.1016/S0191-8869(00)00233-6.
Swickter, R., & Hittner, J. (2009). Social support coping mediates the relationship between gender and posttraumatic growth. Journal of Health Psychology, 14, 387–393.
Tedeschi, R. G., & Calhoun, L. G. (1996). The posttraumatic growth inventory: measuring the legacy of trauma. Journal of Traumatic Stress, 9, 455–472. doi:10.1007/BF02103658.
Tedeschi, R. G., & Calhoun, L. G. (2004). Posttraumatic growth: conceptual foundations and empirical evidence. Psychological Inquiry, 15, 1–15. doi:10.1207/s15327965pli1501_01.
Tennen, H., & Affleck, G. (1998). Personality and transformation in the face of adversity. In R. G. Tedeschi, C. L. Park, & L. G. Calhoun (Eds.), Posttraumatic growth: Positive changes in the aftermath of crisis (pp. 65–98). Mahwah: Erlbaum.
Tisadale, T. C. (1999). Faith Maturity Scale. In P. C. Hill & R. W. Hood (Eds.), Measures of religiosity (pp. 171–174). Birmingham, AL: Religious Education Press.
Van Wicklin, J. F. (1990). Conceiving and measuring ways of being religious. Journal of Psychology and Christianity, 9, 208–219.
Pastoral Psychol (2014) 63:57–71 71
Copyright of Pastoral Psychology is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.
PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN Campbell et al. / NARCISSISM AND SELF-ESTEEM
Narcissism, Self-Esteem, and the Positivity of Self-Views: Two Portraits of Self-Love
W. Keith Campbell University of Georgia
Eric A. Rudich University of North Carolina, Chapel Hill
Constantine Sedikides University of Southampton
The authors hypothesized that both narcissism and high self- esteem are associated with positive self-views but each is associ- ated with positivity in different domains of the self. Narcissists perceive themselves as better than average on traits reflecting an agentic orientation (e.g., intellectual skills, extraversion) but not on those reflecting a communal orientation (e.g., agreeable- ness, morality). In contrast, high-self-esteem individuals per- ceive themselves as better than average both on agentic and com- munal traits. Three studies confirmed the hypothesis. In Study 1, narcissists rated themselves as extraverted and open to experi- ence but not as more agreeable or emotionally stable. High-self- esteem individuals rated themselves highly on all of these traits except openness. In Study 2, narcissists (but not high-self-esteem individuals) rated themselves as better than their romantic part- ners. In Study 3, narcissists rated themselves as more intelligent, but not more moral, than the average person. In contrast, high- self-esteem individuals viewed themselves as more moral and more intelligent.
Two constructs that continue to command the atten- tion of social and personality psychologists are narcis- sism and self-esteem. These two constructs are partially overlapping. First and foremost, both narcissists and high-self-esteem individuals have a high self-opinion: They are said to like—and even love—themselves. Indeed, this similarity may explain why the two variables corre- late positively, as a recent meta-analysis indicated (r = .29, k = 11, n = 2,963, p < .001) (Campbell, 2001). However, narcissism and high self-esteem also have critical differ- ences. Of particular note are the interpersonal implica- tions of these traits. Narcissism is rather detrimental to interpersonal relationships, whereas self-esteem may be beneficial. Perhaps this is why in our culture narcissism is
considered to be a curse, whereas high self-esteem is regarded as a boon.
Our objective in the present research is to explore the bases of the positive self-views that narcissists and high- self-esteem (HSE) individuals have. In particular, we wish to uncover those aspects of the self in which narcis- sists and HSE individuals hold themselves in the highest (and lowest) regard. To presage our hypotheses: We pre- dict that even though both narcissists and HSE individu- als have positive self-views, these groups hold self-views that are distinct in theoretically meaningful ways. Spe- cifically, narcissists’ self-conceptions reflect agentic (but not communal) concerns and HSE individuals’ self-con- ceptions reflect both agentic and communal concerns. That is, narcissists manifest an egoistic bias, whereas HSE individuals display both an egoistic and a moralistic bias. Narcissists perceive themselves as intelligent and outgo- ing but not as caring or conscientious. HSE individuals perceive themselves as both intelligent and caring.
Our research paradigm is derived primarily from work on the better-than-average effect (Alicke, 1985; Alicke, Klotz, Breitenbecher, Yurak, & Vredenberg, 1995). We ask participants to compare themselves to the average other on a range of theoretically relevant traits. We rely on two widely used personality instruments, the Narcis- sistic Personality Inventory (NPI) (Raskin & Hall, 1979)
Authors’ Note: We would like to thank Ahmie Polak for assistance in conducting these studies. We would also like to thank Roy Baumeister, Dawn Dhavale, and Katie Vohs for helpful comments. Correspondence concerning this article should be addressed to W. Keith Campbell, De- partment of Psychology, University of Georgia, Athens, GA 30602- 3013; e-mail: firstname.lastname@example.org.
PSPB, Vol. 28 No. 3, March 2002 358-368 © 2002 by the Society for Personality and Social Psychology, Inc.
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
and the Rosenberg Self-Esteem Scale (RSE) (Rosenberg, 1965). Before presenting our methodological proce- dures in detail, we will review briefly the relevant litera- ture on self-concept biases, narcissism, and self-esteem.
Researchers have identified two primary types of self- deceptive biases, an egoistic bias and a moralistic bias (Paulhus & John, 1998). These two biases reflect either an agentic or a communal value system—loosely speak- ing, a concern either with social dominance or social connection. An egoistic bias is part of an agentic value system and includes inflated self-views in the domains of extraversion, openness, and intelligence. A moralistic bias is part of a communal value system and includes inflated self-views in the domains of agreeableness, con- scientiousness, and morality. Paulhus and John (1998) described these two patterns of value systems and biases at a more general level as alpha and gamma constella- tions. The present research is an extension of this theo- retical approach for the comparison of narcissists and HSE individuals.
Characterization. The personality dimension of narcis- sism is derived from the clinical criteria for narcissistic personality disorder, but as applied to a normal popula- tion (for reviews, see Emmons, 1987; Morf & Rhodewalt, in press; Rhodewalt & Morf, 1995). Narcissists are char- acterized by a highly positive or inflated self-concept. Narcissists use a range of intrapersonal and interper- sonal strategies for maintaining positive self-views. For example, narcissists fantasize about fame or power (Raskin & Novacek, 1991), respond to critical feedback with anger and self-enhancing attributions (Campbell, Reeder, Sedikides, & Elliot, 2000; Farwell & Wohlwend-Lloyd, 1998; Rhodewalt & Morf, 1996), and derogate those who provide threatening feedback (Kernis & Sun, 1994). In addition, narcissists have interpersonal relationships that lack in commitment and caring (Campbell, 1999; Camp- bell & Foster, 2001). On the Five Factor Model (FFM) of personality, narcissism is related most consistently to extraversion. However, there is also some evidence that narcissism is related positively to openness/intellectance and negatively to neuroticism and agreeableness (Bradlee & Emmons, 1992; Costa & Widiger, 1994; Hendin & Cheek, 1997; Rhodewalt & Morf, 1995).
Self-concept positivity. As noted above, narcissists’ self- views should reflect high agency and low communion. Past research is largely consistent with this view. Narcis- sists perceive themselves to be more intelligent (Gabriel, Critelli, & Ee, 1994) and creative (Raskin & Shaw, 1988) than nonnarcissists. They exhibit defensive self-esteem such that they seek admiration but not acceptance (Raskin,
Novacek, & Hogan, 1991a). They also manifest a some- what unstable self-esteem (Rhodewalt, Madrian, & Cheney, 1998). Furthermore, narcissists score highly on the Self- Attributes Questionnaire (SAQ) (Rhodewalt & Morf, 1995), although this measure makes it difficult to distin- guish egoistic and moralistic biases. This pattern of self- aggrandizement also can be observed in experiments that involve interdependent (i.e., joint outcomes) tasks and experimenter-provided feedback on agentic traits (e.g., creativity) (Sedikides, Campbell, Reeder, & Elliot, 1998). On such tasks, narcissists report that their perfor- mance is superior to that of their partners, regardless of whether they work in dyads (Campbell et al., 2000; Farwell & Wohlwend-Lloyd, 1998) or small groups (Gos- ling, John, Craik, & Robins, 1998; John & Robins, 1994). Finally, in the self-deception literature, narcissism has been linked to an egoistic bias and, specifically, the FFM traits of extraversion and openness (Paulhus & John, 1998). In summary, the key theme underlying these find- ings is an agency orientation on that part of narcissists.
The interpersonal dimension. The relation between nar- cissism and variables associated with interpersonal relat- edness is negative. Narcissists express a relatively low desire for many aspects of interpersonal relatedness. This is evident in a lower need for intimacy (Carroll, 1987) and succorance (Raskin & Terry, 1988). Narcissists are also less empathetic in their relationships (Watson, Grisham, Trotter, & Biderman, 1984). Likewise, narcis- sists report enhanced levels of agency (Bradlee & Emmons, 1992), dominance (Bradlee & Emmons, 1992; Emmons, 1984; Raskin, Novacek, & Hogan, 1991b; Raskin & Terry, 1988), power (e.g., Carroll, 1987), machiavellianism (McHoskey, 1995), and competitiveness (Raskin & Terry, 1988). Clearly, narcissists are unlikely to desire relation- ships as a source of intimacy (Sedikides, Campbell, Reeder, Elliot, & Gregg, in press). Indeed, narcissists are low on communal orientation, a pattern that reflects less self- deception on such traits as agreeableness, conscientious- ness, and morality (Paulhus & John, 1998).
Does this mean that narcissists are loners or recluses? This is likely not the case. Narcissists do desire contact with others; however, the purpose behind this contact is largely the enhancement of the narcissists’ self via admi- ration, dominance, and competitiveness (Sedikides et al., in press). Narcissists are judged as sociable (e.g., “entertaining” and “not boring;” Paulhus, 1998, Study 2, Time 1) and energetic (Raskin & Terry, 1988). Also, nar- cissists report relatively low levels of social anxiety (Wat- son & Biderman, 1994) and they do not differ reliably from nonnarcissists on loneliness (Rudich & Sedikides, 2001). In addition, narcissists are high in sensation-seek- ing (Emmons, 1991) and report (and are judged to have) elevated levels of exhibitionism and attention-
Campbell et al. / NARCISSISM AND SELF-ESTEEM 359
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
seeking (Buss & Chiodo, 1991; Raskin & Terry, 1988; Rudich, 1999).
This approach to interpersonal relationships is well illustrated in narcissists’ romantic relationships. Narcis- sists are attracted to admiring and highly positive individ- uals who will enhance the narcissists’ sense of self-worth either directly via praise or indirectly via identification (e.g., a “trophy spouse”). Narcissists are less attracted to caring individuals (Campbell, 1999). Once in a romantic relationship, a similar self-serving pattern can be observed. Relative to nonnarcissists, narcissists report less commit- ment in ongoing romantic relationships. This is largely a result of narcissists’ increased attention to alternative dating partners (Campbell & Foster, 2001). Likewise, narcissists’ love styles reflect greater game-playing and more selfishness (Campbell, Foster, & Finkel, 2001). These particular patterns of behavior in romantic rela- tionships are not characteristic of HSE individuals.
To summarize, we anticipate that narcissists will have positive self-views in domains reflecting agency (e.g., extraversion, openness, intellectance). In contrast, nar- cissists will not report inflated self-views in domains reflecting a communal orientation (e.g., agreeableness, conscientiousness, and morality). This pattern will be evident in narcissists’ romantic relationships. Specifically, they are likely to rate themselves as better than their romantic partners.
Characterization and self-concept positivity. By definition, HSE individuals evaluate themselves positively. Further- more, this positive self-evaluation will be reflected in both agentic and communal domains. HSE individuals are generally confident, gravitating toward leadership positions (Rosenberg, 1965). In contrast, low-self-esteem individuals have a lesser (although not necessarily highly negative) opinion of themselves (Baumeister, Tice, & Hutton, 1989). In fact, low-self-esteem individuals have certain areas in which they believe that they excel but are otherwise somewhat lacking in confidence (Pelham, 1993). On the FFM, self-esteem is correlated positively with the factors of extraversion, conscientiousness, and openness/intellectance. Self-esteem also correlates neg- atively with neuroticism (Jackson & Gerard, 1996). Fur- thermore, self-esteem has been linked to a general self- enhancement bias derived from self-ratings on traits rep- resenting the FFM (Sinha & Krueger, 1998). Interest- ingly, narcissism did not correlate with this self-enhance- ment bias in the Sinha and Krueger (1998) study when self-esteem was controlled.
The interpersonal dimension. High self-esteem is linked to several positive relational outcomes. For example, the positive link between self-esteem and interpersonal relat- edness is a central tenet of the sociometer model of self-
esteem (Leary, Tambor, Terdal, & Downs, 1995). Individ- uals with low self-esteem also may be more socially needy than HSE individuals (Rudich & Vallacher, 1999).
When examining romantic relationships, the influ- ence of self-esteem is complex (Campbell & Baumeister, 2001). HSE individuals typically have positive evalua- tions of their romantic partners that may, in turn, result in relationship satisfaction (Murray, Holmes, & Griffin, 1996a, 1996b). HSE individuals also report less mania or “lovesickness” in their romantic relationships (Camp- bell et al., 2001; Hendrick & Hendrick, 1986). In con- trast, low-self-esteem individuals engage in reassurance- seeking behaviors in romantic relationships, especially when they are depressed (Joiner, Alfano, & Metalsky, 1992). In short, HSE is related to some positive relation- ship-oriented outcomes and HSE individuals are not as socially needy as low-self-esteem individuals.
To summarize, we anticipate that HSE individuals will have positive self-views in domains reflecting agency (e.g., extraversion, openness, intellectance) as well as those reflecting a communal orientation (e.g., agree- ableness, conscientiousness, morality). This pattern will be evident in HSE individuals’ romantic relationships. Specifically, HSE individuals are likely to refrain from rating themselves as better than their romantic partners.
The Present Research
The primary goal of the present research is to distin- guish between the self-concepts of narcissists and HSE individuals. It is clear from the research literature that both narcissists and HSE individuals have positive self- views. However, we propose that the two groups differ in the specific self-views that they deem to be positive.
If there are differences in the positivity of narcissists’ and HSE individuals’ self-concepts, where would these differences likely be found? We hypothesize that narcis- sists view themselves positively primarily in domains reflect- ing agency (e.g., extraversion, openness, intelligence). That is, they will manifest an egoistic bias. However, nar- cissists will not display inflated self-views in the domain of communion (e.g., agreeableness, conscientiousness, moral- ity). That is, they will not manifest a moralistic bias. In contrast, HSE individuals perceive themselves to be posi- tive on a range of traits. These will include both agentic and communal traits. Stated otherwise, these individuals will display both an egoistic and a moralistic bias.
In the present research, we relied on a standard and a modified better-than-average effect procedure. This pro- cedure requires individuals to describe their self- concept by comparing themselves to others on a range of trait terms. We used converging methods to examine self-concept positivity on agentic and communal traits. In Study 1A, we examined the better-than-average effect on a list of positive and negative trait terms taken from
360 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
past research (Alicke, 1985) as well as trait terms derived from the FFM (John, 1990). An agentic bias will be reflected in elevated extraversion and openness to expe- rience/intellectance ratings. A communal bias will be reflected in elevated agreeableness and conscientious- ness ratings. In Study 1B, we replicated these findings and also examined the positivity attached to these trait words. In Study 2, we examined self-concept positivity directly in the interpersonal realm. Individuals rated their views of themselves, their romantic partner, and themselves relative to their romantic partner. An agentic bias will be reflected in an inflated view of self compared to the romantic partner. In Study 3, we adopted a proce- dure associated with the “Muhammad Ali effect” (Allison, Messick, & Goethals, 1989). Participants described their self-views regarding intelligence and morality. An agentic bias will be reflected in enhanced intelligence ratings, whereas a communal bias will be reflected in enhanced morality ratings. To summarize our hypotheses, narcis- sists’ better-than-average self-views will fall squarely in the domain of agency (i.e., egoistic bias), whereas HSE individuals’ self-views will extend into both agency and communal domains (i.e., egoistic and moralistic biases).
Do narcissists and HSE individuals report having posi- tive yet distinct self-views? In what aspects of the self- concept, agency or communion, do these positive self- views reside? We approached these questions by examin- ing traits relevant to aspects of the FFM.
Participants. In Study 1A, 113 undergraduate students (27 men, 86 women) from the University of North Carolina at Chapel Hill (UNC-CH) participated. In Study 1B, par- ticipants were 85 UNC-CH undergraduate students. Due to an error in data collection, participant gender was not recorded in Study 1B and Study 3. In all studies, volun- teers received Introductory Psychology course credit and were thoroughly debriefed at the end of the experi- mental session.
Materials and procedure. After arriving at the experi- mental room, participants in Study 1A completed the RSE scale and the NPI. The form of the RSE that we used contained 10 items that were responded to on a 9-point scale (potential range 10-90). The NPI contained 40 forced-choice items with a potential range of 0 to 40. Next, participants reported the extent to which they pos- sessed certain traits relative to the average person. Partic- ipants rated themselves on 80 traits using a 9-point scale with endpoints at 0 (much less than the average person) and 8 (much more than the average person). We adapted this pro- cedure from Alicke (1985). We took 40 of these traits
directly from Alicke (1985), with 20 traits expressing posi- tive characteristics (e.g., intelligent, dependable) and another 20 traits expressing negative characteristics (e.g., insecure, complaining). We derived an additional 40 traits from the FFM (John, 1990) factors of extraversion (e.g., energetic, assertive), agreeableness (e.g., cold [reverse-scored], cooperative), conscientiousness (e.g., efficient, organized), neuroticism (e.g., tense, nervous), and openness to experience/intellectance (e.g., clever, intelligent).
In Study 1B, participants completed the same mea- sures and trait ratings as in Study 1A. In addition, Study 1B participants rated the positivity of each of these traits using a 9-point scale with anchors at 0 (very negative) and 8 (very positive). We hypothesized that the traits on which narcissists and HSE individuals rate themselves as better than average also will be the traits that they deem to be positive (Alicke, 1985; Sedikides, 1993; Sedikides & Green, 2000).
Descriptive statistics. In Study 1A, means and standard deviations for the variables of interest were as follows: RSE (M = 69.38, SD = 13.82, α = .89), NPI (M = 15.30, SD = 6.67, α = .84), positive traits (M = 5.81, SD = .81, α = .90), negative traits (M = 2.93, SD = .94, α = .80), extraversion (M = 4.71, SD = 1.17, α = .87), agreeableness (M = 5.40, SD = 1.86, α = .90), conscientiousness (M = 4.99, SD = .92, α = .70), neuroticism (M = 3.58, SD = 1.03, α = .81), and openness (M = 5.07, SD = .78, α = .79). The RSE and the NPI were correlated, r = .22, p < .05.
In Study 1B, means and standard deviations for the variables of interest were as follows: RSE (M = 73.52, SD = 13.05, α = .88), NPI (M = 16.72, SD = 6.59, α = .83), posi- tive traits (M = 5.93, SD = .93, α = .89), negative traits (M = 2.71, SD = 1.05, α = .90), extraversion (M = 4.85, SD = 1.23, α = .87), agreeableness (M = 5.85, SD = .96, α = .78), conscientiousness (M = 5.18, SD = 1.03, α = .80), neuroticism (M = 3.40, SD = 1.16, α = .84), and openness (M = 5.26, SD = .88, α = .83). The RSE and the NPI were correlated, r = .24, p < .05.
Positive and negative traits. We present all results in Table 1. This table also contains a combined correlation representing the results from both Study 1A and 1B. As hypothesized, both narcissism and HSE were related sig- nificantly to perceiving the self as above average on posi- tive trait terms across both samples. Likewise, self-esteem was related inversely to perceiving the self as above aver- age on negative trait terms (as expected), whereas there was no relation between narcissism and negative trait terms. Thus, both constructs predicted the better-than- average effect on positive traits, but only HSE predicted the better-than-average effect on negative traits. (These
Campbell et al. / NARCISSISM AND SELF-ESTEEM 361
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
findings were not qualified by gender, with the exception that the link between self-esteem and positive traits and neuroticism was stronger for women than for men.)
Next, we compared the correlations involving narcis- sism and self-esteem both for positive and negative traits (Cohen & Cohen, 1983). In both cases, HSE individuals reported a more positive self-concept than narcissists. Finally, the trait positivity rating paralleled the better- than-average ratings.
FFM traits. We display the FFM results also in Table 1. As hypothesized, narcissism was associated significantly with the better-than-average effect for extraversion and
openness to experience. Both of these factors reflect an agency orientation. Likewise, there was no relation between narcissism and the factors representing communal ori- entation (i.e., agreeableness and conscientiousness). Nar- cissists did not think that they were better than average on these traits. Finally, there was only a small negative correlation between narcissism and reported better- than- average neuroticism. This overall pattern of results was consistent across the two samples. Finally, the trait positivity rating paralleled the better-than-average rat- ings in all but one instance.
Also in line with the hypotheses, correlations revealed that self-esteem was related positively to perceiving the self as (a) better than average on the two communal fac- tors (i.e., agreeableness and conscientiousness) and one of the two agency factors (i.e., extraversion) and (b) below average on neuroticism. HSE individuals consid- ered themselves better than average on both communal and agentic traits.
Next, we compared the self-views of narcissists and HSE individuals (Table 1). Narcissists, relative to HSE individuals, displayed a better-than-average effect on agentic traits. In contrast, HSE individuals, relative to narcissists, showed a better-than-average effect on com- munal traits.
Both narcissists and HSE individuals have positive— yet distinct—self-views. When measured with a list of unspecified traits, HSE individuals appear to have more positive self-concepts than narcissists. The reasons for this difference become clear when a researcher exam- ines the better-than-average effect on traits that reflect the FFM. Narcissists’ self-superiority beliefs were limited primarily to those traits that reflect agency, specifically extraversion and openness. On these traits, narcissists displayed a larger better-than-average effect than HSE individuals. In contrast, HSE individuals’ self-superiority beliefs expanded on the complete range of traits: agentic, communal, and in between (i.e., neuroticism). Indeed, HSE individuals reported a larger better-than- average effect than narcissists on all but the two agentic traits.
Two findings were unexpected. Narcissism was unre- lated to the better-than-average effect on general nega- tive traits. Perhaps this finding is an artifact of the com- position of the word list. In particular, the positive traits contained several items regarding intelligence (i.e., agentic traits), whereas the negative traits were more communal. Also unexpected was the lack of correlation between HSE and self-rating on traits that denoted openness to experience. Nevertheless, this correlation was in the expected positive direction. A replication is needed.
362 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
TABLE 1: Narcissism, Self-Esteem, and the Better-Than-Average Effect: Study 1, Samples A and B
More Agency/ NPI RSE t Value Enhancing Communion
Positive items Rating A .28* .37** Rating B .15 .40** Combined .22** .38** –1.97* RSE Both Positivity B .29** .42**
Negative items Rating A –.01 –.40** Rating B .08 –.48** Combined .03 –.43** 5.80** RSE Both Positivity B –.06 –.26*
Extraversion Rating A .46** .34** Rating B .58** .35** Combined .51** .34** 2.31* NPI Agency Positivity B .43** .17
Agreeableness Rating A –.04 .47** Rating B –.17 .47** Combined –.10 .47** –7.49** RSE Communion Positivity B –.11 .29**
Conscientiousness Rating A .13 .30** Rating B .10 .33** Combined .12 .31** –2.25* RSE Communion Positivity B .17 .31**
Neuroticism Rating A –.13 –.48** Rating B –.17 –.65** Combined –.15* –.56** 5.57** RSE Both Positivity B –.14 –.30**
Openness Rating A .44** .14 Rating B .37** .10 Combined .41** .12 3.38** NPI Agency Positivity B .44** .18
NOTE: RSE = Rosenberg Self-Esteem Scale, NPI = Narcissistic Person- ality Inventory. Combined refers to the combined r across samples A and B. More enhancing is the group (narcissists, high-self-esteem indi- viduals) that exhibited more self-enhancement. Agency/communion refers to the type of trait. *p < .05. **p < .01.
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
In Study 2, we examined the self-views of narcissists and HSE individuals using a different approach. Spe- cifically, we examined the degree to which narcissists and HSE individuals in ongoing romantic relationships were willing to (a) rate their current romantic partner as better than average and (b) rate themselves as better than their current romantic partner. We hypothesized that narcissists will express a positive view of themselves compared to the average other and, more important, that they will perceive the self as superior to their current romantic partner. In contrast, we hypothesized that HSE individuals will express a positive view of themselves compared to average others but that they will not rate the self as superior to their romantic partner. These hypotheses reflect the difference in orientations between narcissists and HSE individuals. If narcissists have an agentic but noncommunal orientation, their positive self-views will not be attenuated when the comparison is a romantic partner. In contrast, if HSE individuals have both agentic and communal self-views, they will be will- ing to temper their positive self-views when the compari- son is a close other.
Participants. One hundred undergraduate students (50 men, 50 women) from Case Western Reserve Univer- sity volunteered for the study. (None of the results was qualified by gender.) All participants were currently involved in a romantic relationship. We collected these data as part of a larger study of narcissism and romantic relationships.
Procedure. First, participants completed the RSE and NPI. Next, they rated themselves compared to the aver- age person on 10 positive and 10 negative personality traits, which represented a subset of those that we used in Studies 1 and 2. Participants also compared their roman- tic partner to the average person on these 20 traits. Finally, participants rated themselves in comparison to their romantic partner on the 20 traits. Unlike Study 1, the 9-point rating scale in Study 2 ranged from –4 to +4, with 0 as the midpoint. We modified the range of the scale to make salient the comparison between the self and the partner.
Results and Discussion
Descriptive statistics. Means and standard deviations for the variables of interest were as follows: RSE (M = 73.95, SD= 11.73, α = .86), NPI (M = 17.55, SD = 7.73, α = .88), self versus average person on positive traits (M = 1.65, SD = 1.10, α = .73) and on negative traits (M = –1.02, SD = 1.04, α = .72), romantic partner versus average person on positive traits (M = 1.65, SD = .94, α = .82) and on negative
traits (M = –1.19, SD = .96, α = .63), and self versus roman- tic partner on positive traits (M = .63, SD = .80, α = .69) and on negative traits (M = –.00, SD = .61, α = .46). The RSE and the NPI were correlated, r = .22, p < .05. These values are consistent with those of Study 1.
Self versus average other. We display the results in Table 2. The comparisons of the self to the average other repli- cated those of Study 1. Both narcissists and HSE individ- uals reported positive self-views, with HSE individuals being more positive on the negative traits.
Romantic partner versus average other. How positively do narcissists and HSE individuals view their romantic part- ners? Narcissists did not view their romantic partners as better than average on either the positive or the negative traits. In contrast, HSE individuals did view their roman- tic partners in a positive light, although only when respond- ing to the negative traits. That is, they rated their part- ners as being below average on negative traits.
Self versus romantic other. We asked participants to rate themselves in relation to their romantic partners. For narcissists, changing the comparison had little effect. Narcissists perceived themselves as better than their romantic partners on positive traits (r = .48 vs. .41 for self better-than-average other). In contrast, HSE individuals’ better-than-average effect disappeared when the target was their romantic partner rather than an average other.
In summary, this examination of self-views in roman- tic relationships reveals an interesting theme. Befitting a primarily agentic orientation, narcissists hold positive self-views in limited areas (i.e., those represented by the positive trait terms) and are willing to maintain these self-views even at the cost of derogating their own roman- tic partners. In contrast, consistently with a less agentic and a more communal orientation, HSE individuals’ positive self-views are shared with a romantic partner. We should note that these findings, although in line with much empirical work on “normal” narcissists, may coun- ter some clinical research that has noted romantic ideal- ization on the part of narcissists (e.g., Kernberg, 1974). Assuming that the clinical insights are valid, there are at least two resolutions to this discrepancy. First, it is possi- ble that that there is an early and highly transitory ideal- ization stage on the part of narcissists that our measures did not detect. Second, narcissists’ idealized beliefs about romantic partners may be evident in only a small sub- group of narcissists or in a subgroup of relationships. The presence of such subgroups may not have been suffi- cient in our sample to affect our results.
In Study 3, we attempted to gain another perspective on the self-views of narcissists and HSE individuals by
Campbell et al. / NARCISSISM AND SELF-ESTEEM 363
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
examining the “Muhammad Ali effect.” According to this effect, people believe that they are more moral, but not more intelligent, than the average person (Van Lange, 1991; Van Lange & Sedikides, 1998; Sedikides & Strube, 1997). For the purpose of the present investiga- tion, this effect serves as a technique to pit directly agentic (i.e., intelligence) against communal (i.e., moral- ity) aspects of the self.
Indeed, the Muhammad Ali effect is particularly appro- priate for examining agentic and communal self-views. Consistently with Paulhus and John’s (1998) theorizing, intelligence will clearly reflect an egoistic bias, whereas morality will by definition reflect a moralistic bias. As such, we hypothesize that narcissists will rate themselves as better than average on intelligence (an agentic trait) but not on morality (a communal trait). In contrast, HSE individuals will rate themselves as better than average on both traits.
Materials and procedure. Participants were 109 UNC- CH undergraduates who completed the RSE and NPI and then reported the degree to which they thought they were better than average on traits that described intelli- gence and morality. The response format was the same as the one used in Studies 1 and 2. We assessed the degree to which participants reported that they possessed better-than-average intelligence by using 11 traits, such as intelligent, smart, and bright. We assessed the degree to which participants reported that they possessed better-than-average morality by using 17 traits, such as moral, honest, and deceptive (reverse scored). We com- bined scores from these scales to form indices of better- than-average intelligence and morality. We also asked participants to rate the positivity of these traits.
Results and Discussion
Descriptive statistics. Means and standard deviations for the variables of interest were as follows: RSE (M = 74.00, SD = 10.01, α = .81), NPI (M = 16.05, SD = 6.31, α = .82), intelligence traits (M = 6.07, SD = 1.04, α = .90), and moral traits (M = 6.85, SD = .77, α = .84). The RSE and the NPI were correlated, r = .36, p < .05.
Better-than-average intelligence. We display all relevant correlations in Table 3. As hypothesized, both narcissism and self-esteem were associated positively with better- than-average intelligence. This latter result suggests that the lack of significance found in the correlation between self-esteem and openness to experience/intellectance in Study 1 may reflect a small effect size and lack of statis- tical power. A statistical test showed that narcissism, when compared to self-esteem, was related to a margin- ally larger better-than-average effect on intelligence traits.
Better-than-average morality. We observed a different pattern of findings on the morality items (Table 3). There was no significant relation between narcissism and better-than-average morality. This pattern is consis- tent with the hypothesis that narcissists will not report being better than average on communal traits. In con- trast, there was a significant positive relation between self-esteem and self-reported morality. Individuals high on self-esteem believed that they were more moral than the average person. When we compared the narcissism and self-esteem correlations, we found that HSE individ- uals reported a significantly higher better-than-average effect on morality.
In summary, the results of Study 3 supported the hypotheses. Narcissists perceive themselves as being more intelligent (an agentic trait) but not as more moral (a communal trait) than the average person. HSE individu- als tend to see themselves as both more intelligent (although not to the degree that narcissists do) and more moral than the average person.
Both narcissists and HSE individuals have positive self-views, as our investigation confirms. More impor- tant, it is now clear that the particular self-views on which narcissists and HSE individuals perceive themselves as being better than average differ reliably. To communi- cate this notion statistically, we meta-analyzed the results of our three studies (Table 4). For each sample, we pres- ent the average correlation between narcissism and the agency and communion variables as well as for self- esteem and the agency and communion variables. We operationalize agency in terms of extraversion and open- ness (Study 1A and 1B), self versus romantic partner (Study 2), and intelligence (Study 3). We operationalize
364 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
TABLE 2: Narcissism, Self-Esteem, and the Better-Than-Average Effect in Romantic Relationships: Study 2
More NPI RSE t Value Enhancing
Self versus average other Positive items .41** .27** 1.24 — Negative items –.05 –.45** 3.54* RSE
Romantic partner versus average other Positive items –.03 .12 –1.19 — Negative items –.03 –.35** 2.70* RSE
Self versus romantic partner Positive items .48** .06 3.78** NPI Negative items –.06 –.18 .96 —
NOTE: RSE = Rosenberg Self-Esteem Scale, NPI = Narcissistic Person- ality Inventory. More enhancing is the group (narcissists, high-self- esteem individuals) that exhibited more self-enhancement. *p < .05. **p < .01.
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
communion in terms of agreeableness and conscien- tiousness (Study 1A and 1B) and morality (Study 3). (Across the three studies, the average correlation between narcissism and self-esteem was .26.)
The pattern of results is remarkably consistent across studies. Narcissists perceive themselves as better than average on traits reflecting agency but do not perceive themselves as better than average on traits reflecting communion. In contrast, HSE individuals perceive them- selves as better than average on both agency and commu- nion traits. Furthermore, an inspection of the confi- dence intervals around the combined correlations sug-
gests that narcissists’ better-than-average views on agency traits (combined r = .41) are higher than those of HSE individuals (combined r = .21). Likewise, narcissists’ self- views on communal traits (combined r = –.06) are lower than those of HSE individuals (combined r = .33).
Put more simply, narcissists’ positive self-opinions rest squarely and strongly in the agency domain, whereas HSE individuals allocate their positive self-opinions equally to the agency and communion domains.
What do these findings tell us about the relation between narcissism and HSE? First, narcissism does not appear simply to reflect exceptionally high self-esteem. Indeed, HSE individuals viewed themselves equally to or in a more positive light than did narcissists. Rather, the key differences between these two groups are the facets of the self that each group holds in high regard. Narcis- sists view themselves as highly outgoing and competent on certain cognitive skills (i.e., agency). These positive beliefs do not transfer to their romantic partners. Also, narcissists are relatively unconcerned with being nice or moral (i.e., communion); that is, they display an exten- sive egoistic bias but not a moralistic bias. In contrast, HSE individuals view themselves as highly positive on communal traits, such as nice, considerate, conscien- tious, well-adjusted, and moral. Also, they hold positive views of their romantic partners. In addition, they per- ceive themselves as more intelligent than others but not to the same extent as do narcissists. In summary, HSE individuals display both an egoistic and a moralistic bias.
These differences in self-conceptions have implica- tions for the interpersonal self-regulatory strategies that the two groups use. Narcissists’ positive agentic self-views are expressed through efforts to win admiration and attention from others, often by comparing and compet- ing with others; narcissists see themselves as willing to assert and defend their competence interpersonally. For example, a narcissist who believes that he is intelligent may actively maintain this self-view by publicly exclaim- ing his own superior skills, derogating the success of oth- ers (including even a close other), and seeking situations in which he can compete intellectually with others. Why are narcissists willing to derogate others in the pursuit of individual self-enhancement? Probably because narcis- sists are not burdened by communal concerns (Sedikides et al., in press).
In contrast, HSE individuals report both egoistic and moralistic biases. Self-regulatory efforts on the part of HSE individuals will thus be aimed at enhancing both agentic and communal traits. This will make it problem- atic to enhance the self by, for example, comparing the self positively to close others (e.g., Study 2). HSE individ-
Campbell et al. / NARCISSISM AND SELF-ESTEEM 365
TABLE 3: Narcissism, Self-Esteem, and the Muhammad Ali Effect: Study 3
More Agency/ NPI RSE t Value Enhancing Communion
Intelligence Rating .41** .23** 1.80† NPI Agency Positivity .27** .02
Morality Rating –.17 .21* –3.67** RSE Communion Positivity –.06 –.08
NOTE: RSE = Rosenberg Self-Esteem Scale, NPI = Narcissistic Person- ality Inventory. More enhancing is the group (narcissists, high-self-es- teem individuals) that exhibited more self-enhancement. Agency/ communion refers to the type of trait. †p < .10. *p < .05. **p < .01.
TABLE 4: Synthesis of Agency and Communion Results Across Studies
Study Study Study Study Combined, 1A 1B 2 3 ci 95%
Narcissism (NPI) Agency .45** .48** .28** .41** .41
Communion .04 –.04 –.17 –.06 (–.16,
.04) High self-esteem (RSE) Agency .24** .23* .12 .23** .21
Communion .39** .40** .21* .33 (.24, .41)
NOTE: RSE = Rosenberg Self-Esteem Scale, NPI = Narcissistic Person- ality Inventory. Combined n = 407; ci refers to 95% confidence interval. Agency is represented by extraversion and openness (Study 1A and 1B), self versus romantic partner (Study 2), and intelligence (Study 3). Communion is represented by agreeableness and conscientiousness (Study 1A and 1B) and morality (Study 3). *p < .05. **p < .01.
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
uals would like refrain from self-regulatory strategies that inflate egoistic biases at the expense of moralistic biases.
This self-regulatory quandary that is faced by HSE individuals but not by narcissists may be why society smiles on the former and frowns on the latter. Narcissists like themselves in unlikable ways and HSE individuals like themselves in likable ways. One may dislike the nar- cissist for placing importance on outdoing others and not placing importance on interpersonal caring or moral- ity. In contrast, the HSE individual may be admired for placing importance on prosocial traits. The one domain in which society may admire narcissists is achievement. Individuals may not mind a narcissist on the team if he or she is focused on gaining praise by performing well. Unfortunately, narcissism is problematic even in this domain because the narcissist may view success where there is none or even steal success from his or her col- leagues (Farwell & Wohlwend-Lloyd, 1998; John & Robins, 1994; Sedikides & Gregg, 2001). Perhaps another way to distill the difference between narcissists and HSE indi- viduals is that narcissists want to be admired, whereas HSE individuals want to be popular. The latter is less tax- ing from an interpersonal or societal perspective.
There are several caveats that we must note when dis- cussing the implications of the present research. First, before reaching too far into the behavioral realm, it is important to restate that the focus of the present article is on self-views, not behaviors. Although the self certainly is linked to behavior (Fleury, Sedikides, & Lunsford, 2001; Sedikides & Gregg, in press), the self-views of inter- est may or may not be born out in actual behaviors. Future research may examine behavioral differences between narcissists and HSE individuals on various agentic and communal behavioral self-enhancement strategies.
Likewise, we should note that our use of the better- than-average effect has limitations. Foremost, it was not possible to distinguish clearly between accurate and illu- sory self-beliefs. In the past, researchers have confirmed self-inflation on certain beliefs by comparing self-reports to objective measures (e.g., Farwell & Wohlwend-Lloyd, 1998). Similar approaches would be useful in further clarifying accuracy versus inflation in the self-views of narcissists and HSE individuals.
The degree of self-enhancement that participants report on the better-than-average effect paradigm depends on the ambiguity versus specificity of the traits measured (Dunning & McElwee, 1995). We used a high proportion of ambiguous traits in the present research, and it would be useful to know the extent to which our findings are
replicated with specific traits. For example, do narcissists use idiosyncratic definitions of agentic traits?
Finally, our work would have benefited from a more “pure” measure of communal bias. We relied on several traits (e.g., morality, agreeableness) as proxies for com- munal self-beliefs. Future research will need to examine the link between narcissism, self-esteem, and self-beliefs on communion.
There are different ways to love oneself. By compar- ing the self-views of narcissists and HSE individuals, two of these differences become clear. Seeing the self as extremely outgoing and clever (but not as moral or nice) portrays a very different individual than seeing the self as nice and moral as well as somewhat clever or intelligent. Those who adopt the former view are narcissists, whereas those who adopt the latter view have high self-esteem.
Alicke, M. D. (1985). Global self-evaluations as determined by the desirability and controllability of trait adjectives. Journal of Person- ality and Social Psychology, 49, 1621-1630.
Alicke, M. D., Klotz, M. L., Breitenbecher, D. L., Yurak, T. J., & Vredenberg, D. S. (1995). Personal contact, individuation, and the better-than-average effect. Journal of Personality and Social Psy- chology, 68, 804-825.
Allison, S. T., Messick, D. M., & Goethals, G. R. (1989). On being better but not smarter than others: The Muhammad Ali effect. Social Cognition, 7, 275-295.
Baumeister, R. F., Tice, D. M., & Hutton, D. G. (1989). Self-presenta- tional motivations and personality differences in self-esteem. Jour- nal of Personality, 57, 547-579.
Bradlee, P. M., & Emmons, R. A. (1992). Locating narcissism within the interpersonal circumplex and the five-factor model. Personal- ity and Individual Differences, 13, 821-830.
Buss, D. M., & Chiodo, L. M. (1991). Narcissistic acts in everyday life. Journal of Personality, 59, 179-215.
Campbell, W. K. (1999). Narcissism and romantic attraction. Journal of Personality and Social Psychology, 77, 1254-1279.
Campbell, W. K. (2001). A meta-analysis of the narcissism–self-esteem link. Unpublished manuscript, University of Georgia.
Campbell, W. K., & Baumeister, R. F. (2001). Is loving the self neces- sary for loving another? An examination of identity and intimacy. In M. Clark & G. Fletcher (Eds.), Blackwell handbook of social psychol- ogy: Vol. 2. Interpersonal processes (pp. 437-456). London: Blackwell.
Campbell, W. K., & Foster, C. A. (in press). Narcissism and commit- ment in romantic relationships: An investment model analysis. Personality and Social Psychology Bulletin.
Campbell, W. K., Foster, C. A., & Finkel, E. (2001). Does self-love lead to love for others? A story of narcissistic game-playing. Unpublished manu- script, University of Georgia.
Campbell, W. K., Reeder, G. D., Sedikides, C., & Elliot, A. T. (2000). Narcissism and comparative self-enhancement strategies. Journal of Research in Personality, 34, 329-347.
Carroll, L. (1987). A study of narcissism, affiliation, intimacy, and power motives among students in business administration. Psycho- logical Reports, 61, 355-358.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.
Costa, P. T., & Widiger, T. A. (1994). Personality disorders and the five- factor model of personality. Washington, DC: American Psychological Association.
366 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at LIBERTY UNIV LIBRARY on November 4, 2016psp.sagepub.comDownloaded from
Dunning, D., & McElwee, R. O. (1995). Idiosyncratic trait definitions: Implications for self-description and social judgment. Journal of Personality and Social Psychology, 68, 936-946.
Emmons, R. A. (1984). Factor analysis and construct validity of the narcissistic personality inventory. Journal of Personality Assessment, 48, 291-300.
Emmons, R. A. (1987). Narcissism: Theory and measurement. Journal of Personality and Social Psychology, 52, 11-17.
Emmons, R. A. (1991). Relationship between narcissism and sensa- tion seeking. Journal of Social Behavior and Personality, 6, 943-954.
Farwell, L., & Wohlwend-Lloyd, R. (1998). Narcissistic processes: Optimistic expectations, favorable self-evaluations, and self- enhancing attribution. Journal of Personality, 66, 65-83.
Fleury, J., Sedikides, C., & Lunsford, V. (2001). Women’s experience following a cardiac event: The role of the self in healing. Journal of Cardiovascular Nursing, 15, 71-82.
Gabriel, M. T., Critelli, J. W., & Ee, J. S. (1994). Narcissistic illusions in self-evaluations of intelligence and attractiveness. Journal of Person- ality, 62, 143-155.
Gosling, S. D., John, O. P., Craik, K. H., & Robins, R. W. (1998). Do people know how they behave? Self-reported act frequencies com- pared with on-line codings by observers. Journal of Personality and Social Psychology, 112, 107-118.
Hendin, H. M., & Cheek, J. M. (1997). Assessing hypersensitive narcis- sism: A reexamination of Murray’s Narcism Scale. Journal of Research in Personality, 31, 588-599.
Hendrick, C., & Hendrick, S. S. (1986). A theory and method of love. Journal of Personality and Social Psychology, 50, 392-402.
Jackson, L. A., & Gerard, D. A. (1996). Diurnal types, the “big five” personality factors, and other personal characteristics. Journal of Social Behavior and Personality, 11, 273-283.
John, O. P. (1990). The “Big Five” factor taxonomy: Dimensions of personality in the natural language and in questionnaires. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 66- 100). New York: Guilford.
John, O. P., & Robins, R. W. (1994). Accuracy and bias in self-percep- tion: Individual differences in self-enhancement and the role of narcissism. Journal of Personality and Social Psychology, 66, 206-219.
Joiner, T. E., Alfano, M. S., & Metalsky, G. I., (1992). When depression breeds contempt: Reassurance seeking, self-esteem, and rejection of depressed college students and their roommates. Journal of Abnormal Psychology, 101, 165-173.
Kernberg, O. (1974). Barriers to falling and remaining in love. Jour- nal of the American Psychoanalytic Association, 22, 486-511.
Kernis, M. H., & Sun, C. (1994). Narcissism and reactions to interper- sonal feedback. Journal of Research in Personality, 28, 4-13.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self- esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68, 518-530.
McHoskey, J. (1995). Narcissism and machiavellianism. Psychological Reports, 77, 755-759.
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcis- sism: A dynamic self-regulatory processing model. Psychological Inquiry, 12,177-196.
Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996a). The benefit of positive illusions: Idealization and the construction of satisfaction in close relationships. Journal of Personality and Social Psychology, 70, 79-98.
Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996b). The self-fulfill- ing nature of positive illusions in romantic relationships: Love is not blind, but prescient. Journal of Personality and Social Psychology, 71, 1155-1180.
Paulhus, D. L. (1998). Interpersonal and intrapsychic adaptiveness of trait self-enhancement: A mixed blessing? Journal of Personality and Social Psychology, 74, 1197-1208.
Paulhus, D. L., & John, O. P. (1998). Egoistic and moralistic biases in self-perception: The interplay of self-deceptive styles with basic traits and motives. Journal of Personality, 66, 1025-1060.
Pelham, B. W. (1993). On the highly positive thoughts of the highly depressed. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 183-200). New York: Plenum.
Raskin, R., & Hall, C. S. (1979) A narcissistic personality inventory. Psychological Reports, 45, 590.
Raskin, R. N., & Novacek, J. (1991). Narcissism and the use of fantasy. Journal of Clinical Psychology, 47, 490-499.
Raskin, R. N., Novacek, J., & Hogan, R. (1991a). Narcissistic, self- esteem and defensive self-enhancement. Journal of Personality, 59, 19-38.
Raskin, R. N., Novacek, J., & Hogan, R. (1991b). Narcissistic self- esteem management. Journal of Personality and Social Psychology, 60, 911-918.
Raskin, R. N., & Shaw, R. (1988). Narcissism and the use of personal pronouns. Journal of Personality, 56, 393-404.
Raskin, R. N., & Terry, H. (1988). A principle components analysis of the Narcissistic Personality Inventory and further evidence of its construct validity. Journal of Personality and Social Psychology, 54, 890-902.
Rhodewalt, F., Madrian, J. C., & Cheney, S. (1998). Narcissism, self- knowledge organization, and emotional reactivity: The effect of daily experiences on self-esteem and affect. Personality and Social Psychology Bulletin, 24, 75-87.
Rhodewalt, F., & Morf, C. C. (1995). Self and interpersonal correlates of the narcissistic personality inventory. Journal of Research in Per- sonality, 29, 1-23.
Rhodewalt, F., & Morf, C. C. (1996). On self-aggrandizement and anger: A temporal analysis of narcissism and affective reactions. Journal of Personality and Social Psychology, 74, 672-685.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
Rudich, E. A. (1999). The Need for Attention Scale. Unpublished manu- script, University of North Carolina at Chapel Hill.
Rudich, E. A., & Vallacher, R. R. (1999). To belong or to self-enhance? Motivational bases for choosing interaction partners. Personality and Social Psychology Bulletin, 25, 1389-1406.
Sedikides, C. (1993). Assessment, enhancement, and verification determinants of the self-evaluation process. Journal of Personality and Social Psychology, 65, 317-338.
Sedikides, C., Campbell, W. K., Reeder, G., & Elliot, A. J. (1998). The self-serving bias in relational context. Journal of Personality and Social Psychology, 74, 378-386.
Sedikides, C., Campbell, W. K., Reeder, G., Elliot, A. J., & Gregg, A. P. (in press). Do others bring out the worst in narcissists? The “Oth- ers Exist for Me” illusion. In Y. Kashima, M. Foddy, & M. Platow (Eds.), Self and identity: Personal, social, and symbolic. Mahwah, NJ: Lawrence Erlbaum.
Sedikides, C., & Green, J. D. (2000). On the self-protective nature of inconsistency/negativity management: Using the person memory paradigm to examine self-referent memory. Journal of Personality and Social Psychology, 79, 906-922.
Sedikides, C., & Gregg, A. P. (2001). Narcissists and feedback: Motivational surfeits and motivational deficits. Psychological Inquiry, 12, 237-239.
Sedikides, C., & Gregg, A. P. (in press). Self matters. In M. A. Hogg & J. Cooper (Eds.), Sage handbook of social psychology. London: Sage Ltd.