Homogenous scales of narcissism: Using the psychological entitlement scale, interpersonal exploitativeness scale, and narcissistic grandiosity scale to study narcissism

Personality and Individual Differences 123 (2018) 182–190

Contents lists available at ScienceDirect

Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid

Homogenous scales of narcissism: Using the psychological entitlement scale, interpersonal exploitativeness scale, and narcissistic grandiosity scale to study narcissism☆

T

⁎

Amy B. Brunell , Melissa T. Buelow Department of Psychology, The Ohio State University, United States

A R T I C L E I N F O

A B S T R A C T

Keywords: Entitlement Exploitativeness Grandiosity Narcissism Factor analysis

We examined the extent to which the Psychological Entitlement Scale (PES), the Interpersonal Exploitativeness Scale (IES), and the Narcissistic Grandiosity Scale (NGS), when taken together, assess a broader construct or three distinct facets. In Study 1, a principal components analysis was conducted, demonstrating that the PES, IES, and NGS should be considered three separate traits rather than one overall construct. In Study 2, confirmatory factor analysis (CFA) revealed that the most efficient and best fitting model contained 8 items of the PES (dropping a reverse-scored item), the 6-item IES, and a revised 6-item model of the NGS. Study 3 replicated the CFA and examined the correlates of the PES, IES, and NGS with measures of narcissism and related measures such as empathic concern and self-esteem. Implications for future assessment of narcissism traits are discussed.

There has been growing interest in understanding narcissism in recent years. As this interest in narcissism has grown, there has been a corresponding increase in the development of measures that assess narcissism and specific narcissistic traits. Some (e.g., Brown, Budzek, & Tamborski, 2009; Smith, McCarthy, & Zapolski, 2009) suggest that examining homogenous narcissistic traits might be useful because each trait may differentially predict outcomes. For example, some aspects of narcissism (e.g., grandiosity) are positively associated with mental health while other aspects (e.g., entitlement) are negatively associated with mental health (Brown et al., 2009). There are currently several measures available for assessing homogenous traits. For example, Campbell, Bonacci, Shelton, Exline, and Bushman (2004) created the Psychological Entitlement Scale (PES; Campbell et al., 2004) as a stand-alone measure of entitlement and Brunell et al. (2013) created the Interpersonal Exploitativeness Scale (IES) as a stand-alone measure of exploitativeness. More recently, Crowe, Carter, Campbell, and Miller (2016) validated a measure of grandiosity (the Narcissistic Grandiosity Scale; NGS), first developed by Rosenthal, Hooley, and Steshenko (2007), as a singular, homogenous trait associated with narcissism. Crowe et al. found that abbreviated versions of the measure (i.e., a 13-item version and a 6-item version) had strong reliability and performed similarly to the full 16-item measure. Our purpose for the present studies was to examine the extent to

which the PES, IES, and NGS are three separate, distinct constructs or represent one broader construct of narcissism. The possibility that these scales represent a latent broader construct stems from research showing that exploitativeness, entitlement, and grandiosity tend to be correlated (e.g., Brown et al., 2009; Brunell et al., 2013; Brunell & Buelow, 2017; Buelow & Brunell, 2014). Thus, the present studies sought to examine (a) the factor structure of these homogenous measures, and (b) their relationship to other measures of narcissism and related personality variables as well as mental health characteristics. To this point, some (i.e., Miller, Price, & Campbell, 2012) have argued that the NGS and PES do not sufficiently account for the variance in narcissism. We address this point by including an additional measure (the IES) and by assessing the extent to which the NPI accounts for variance of relevant measures above and beyond the homogenous measures. One reason researchers may be interested in understanding the role of specific narcissistic traits is that it may be valuable to know when the darker traits of narcissism (e.g., entitlement, exploitativeness) or the seemingly harmless traits (e.g., grandiosity) are associated with behavior. For example, in an investigation of narcissism and cheating (Brown et al., 2009, Study 3), psychological entitlement (i.e., the PES) predicted deliberative cheating (intentionally engaging in misconduct) and grandiosity (i.e., the NGS) predicted rationalized cheating (situations in which people do not explicitly intend to cheat but are able to explain away their behavior) because psychological entitlement is

☆ ⁎

The authors will share materials and data upon request. Corresponding author at: Department of Psychology, The Ohio State University, Mansfield, OH 44906, United States. E-mail address: [email protected] (A.B. Brunell).

https://doi.org/10.1016/j.paid.2017.11.029 Received 8 September 2017; Received in revised form 14 November 2017; Accepted 16 November 2017 0191-8869/ © 2017 Elsevier Ltd. All rights reserved.

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

2. Study 1: methods

linked to antisocial behavior and overtly rejecting social expectations for behavior, whereas narcissistic grandiosity is linked to having a selfserving mindset that is linked to rationalizing away negative behavior. Other scholars have also sought to examine the role of specific narcissistic traits. For example, Brunell et al. (2013) found that the IES, but not the PES, was linked to resource destruction during a commons dilemma, a situation wherein people share a common renewable resource. Their exploitative behavior during the task demonstrated a lack of cooperativeness over time and taking too much of the resource at a time to allow it to become adequately replenished. In a study of moral reasoning, Daddis and Brunell (2015) found that psychological entitlement and interpersonal exploitativeness were both linked to more reasoning about the self than a concern for others, but these dimensions diverged in predicting judgments about whether certain behaviors were acceptable or not (e.g., texting in class), and why it was acceptable or not. For example, people who were more entitled reasoned that this behavior was not acceptable because it would distract them from their own learning, whereas people who were more exploitative reasoned that texting in class was acceptable because it concerned personal choice. Neither were likely to reason that texting in class might be distracting to others. To our knowledge, only a handful of studies have considered the PES, IES, and NGS jointly. Buelow and Brunell (2014) included grandiosity, psychological entitlement, and exploitativeness in their assessment of risk. They found that the primary predictor of risk-taking behaviors was narcissistic grandiosity. Specifically, the NGS predicted reports of ethical, financial, and social risk-taking as well as reports of aggressive behavior and drug use. Exploitativeness predicted ethical and financial risk-taking and reports of risk-taking in sports. In a study of risk-taking behavior, exploitativeness was a predictor of poorer performance on the Iowa Gambling Task (IGT; Bechara, Damasio, Damasio, & Anderson, 1994), a test of affective and deliberative decision-making (Brunell & Buelow, 2017). In a study of college student volunteerism, each narcissistic trait predicted different motivations for volunteering (Brunell, Tumblin, & Buelow, 2014; Study 2). Exploitativeness was positively associated with the motivation to volunteer to gain new learning experiences or offer opportunities to exercise one's knowledge, skills, and abilities that might otherwise be underutilized. Entitlement, by contrast, was positively associated with the career (resume-building and acquiring new skills) and social (expanding one's social network and to make new connections) motives. Put another way, entitlement and exploitativeness had to do with benefiting the self rather than helping others.

2.1. Participants Three hundred Introductory Psychology students (155 males, 141 females, and 4 individuals who did not indicate their sex) at a regional campus of a large Midwestern University participated in the study in exchange for partial course credit. Participants were 19.77 years old on average (SD = 3.50), and 74.8% self-identified as Caucasian. 3. Materials and procedure The study was approved by the university's Institutional Review Board, and all participants provided informed consent. Participants completed all measures in a randomized order as part of a larger study. All participants were debriefed at the end of the study. The IES is a 6-item measure that examines the extent to which individuals feel comfortable readily taking advantage of others. A sample item is, “vulnerable people are fair game.” Responses are made using 7point scales, ranging from 1 (strong disagreement) to 7 (strong agreement) (Brunell et al., 2013). Total scores are computed by summing scores across the six items (M = 13.35, SD = 7.45, α = 0.89). The PES consists of nine statements, including “I deserve more things in my life.” Responses are made using 7-point scales, ranging from 1 (strong disagreement) to 7 (strong agreement) (Campbell et al., 2004). Total scores are computed by summing responses across the nine items (M = 28.02, SD = 10.47, α = 0.87). The NGS consists of 16 grandiose adjectives, such as “superior.” Participants respond on a 7-point scale, indicating the extent to which the adjectives are self-descriptive (1 = not at all, 7 = extremely) (Rosenthal et al., 2007). Total scores are computed by summing scores across the 16 items (M = 49.53, SD = 19.43, α = 0.96). 4. Study 1: results and discussion A principal components analysis (PCA) was conducted in SPSS, and results of oblique rotations are reported. As we had a relatively large sample size and our three variables were correlated with one another, we opted for a PCA rather than an exploratory factor analysis. To determine which components to retain, a parallel analysis was conducted on a randomly generated dataset with 31 variables. Components in the PCA were retained if the eigenvalue was greater than the corresponding eigenvalue in the parallel analysis (O'Connor, 2000). In addition, Velicer's minimum average partial (MAP) test (Velicer, 1976) was conducted to confirm the number of components retained. Table 1 includes the variable means, standard deviations, eigenvalues, and component loadings. The Kaiser-Meyer-Olkin measure indicated the analysis was acceptable (KMO = 0.941), and Bartlett's test of sphericity was significant, χ2 = 6439.60, p < 0.001. Three components met criteria for retention and accounted for 59.75% of the total variance (Table 1). Supplemental tables include both the scree plot and the results of Velicer's MAP test. The first component included all of the NGS items, with factor loadings ranging from 0.599 (Item 15) to 0.835 (Item 16). The second component included all of the IES items, with factor loadings ranging from 0.675 (Item 2) to 0.904 (Item 6). Component 3 included all of the PES items, except for Item 5, with factor loadings ranging from 0.593 (Item 9) to 0.822 (Item 7). The principal components analysis was re-run with PES Item 5 included but not reverse scored, and no differences emerged in the factor loadings (i.e., three components again emerged). The results of the principal components analysis suggest the measures comprise separate components rather than loading on a single, common factor. This result provides evidence that the IES, PES, and NGS consistently measure separate but distinct facets of narcissism. That the PES Item 5 did not load with the other PES items might be that this item was subject to systematic measurement error due to

1. Present studies We report three studies that examine of the use of the PES, IES, and NGS in the investigation of narcissism. The aim of Study 1 was to examine the factor structure of these three measures to determine if they are three separate traits or should be considered one broader trait. We conducted a principal components analysis on the NGS, PES, and IES to examine this question. Study 2 followed-up on this initial analysis with a confirmatory factor analysis, comparing model fits across competing models based on Study 1 results as well as condensed versions of the NGS (see Crowe et al., 2016). Lastly, the aim for Study 3 was to examine the convergent and discriminant validity of these measures. We examined associations between the NGS, PES, and IES and other established measures of trait narcissism (as opposed to clinical narcissism or Narcissistic Personality Disorder) well as constructs related to narcissism (e.g., empathy, mental health, extraversion). We also assess the extent to which the NPI accounts for variance above and beyond the three homogenous traits. Taken together, these studies aim to understand the extent to which the independent narcissistic traits represent one construct, remove redundancy among the items, and improve efficiency for psychologists seeking to investigate these traits. 183

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Table 1 Study 1: Means, standard deviations, eigenvalues, and factor loadings for the narcissism facets. Variable

Item

M

SD

Factor 1

Factor 2

Factor 3

NGS 1 NGS 2 NGS 3 NGS 4 NGS 5 NGS 6 NGS 7 NGS 8 NGS 9 NGS 10 NGS 11 NGS 12 NGS 13 NGS 14 NGS 15 NGS 16 IES 1 IES 2 IES 3 IES 4 IES 5 IES 6 PES 1 PES 2 PES 3 PES 4 PES 6 PES 7 PES 8 PES 9 PES 5-r Eigenvalue % Variance

Perfect Extraordinary Superior Heroic Omnipotent Unrivalled Authoritative Glorious Prestigious Acclaimed Prominent High-Status Brilliant Dominant Envied Powerful It doesn't bother me to benefit at someone else's expense. Only weak people worry about fairness. Using other people doesn't bother me very much. Vulnerable people are fair game. I'm less interested in fairness than getting what I want. I'm perfectly willing to profit at the expense of others. I honestly feel I'm just more deserving than others. Great things should come to me. If I were on the Titanic, I would deserve to be on the first lifeboat! I demand the best because I'm worth it. I deserve more things in my life. People like me deserve an extra break now and then. Things should go my way. I feel entitled to more of everything. I do not necessarily deserve special treatment.

2.61 3.86 3.08 3.25 2.40 2.65 3.51 2.97 3.11 2.94 3.18 2.88 3.60 3.45 2.99 3.02 2.55 2.25 2.04 2.07 2.24 2.19 2.79 3.72 2.52 3.28 3.36 3.27 3.49 2.52 3.10

1.52 1.53 1.56 1.67 1.46 1.47 1.57 1.61 1.52 1.48 1.49 1.59 1.56 1.64 1.54 1.69 1.65 1.58 1.41 1.46 1.58 1.53 1.66 1.66 1.64 1.76 1.66 1.68 1.60 1.57 1.71

0.651 0.748 0.766 0.809 0.756 0.771 0.800 0.803 0.815 0.802 0.815 0.726 0.640 0.778 0.599 0.835 − 0.069 0.102 0.032 0.018 − 0.031 0.020 0.030 0.061 0.011 0.146 0.073 − 0.047 − 0.064 0.082 0.030 12.66 40.83%

0.021 −0.062 0.117 −0.020 0.075 0.003 −0.004 0.018 −0.073 0.007 0.068 0.075 0.023 0.025 −0.001 0.072 0.709 0.675 0.878 0.793 0.846 0.904 0.179 −0.011 0.030 −0.051 −0.046 −0.035 0.091 0.180 −0.020 3.62 11.69%

0.032 − 0.071 0.013 − 0.131 − 0.007 0.011 − 0.065 0.084 0.135 0.087 0.020 0.138 0.143 0.025 0.141 − 0.042 0.134 0.021 − 0.077 − 0.005 0.025 − 0.052 0.606 0.768 0.706 0.672 0.775 0.822 0.730 0.593 0.089 2.24 7.23%

Note: NGS = Narcissistic Grandiosity Scale, IES = Interpersonal Exploitativeness Scale; PES = Psychological Entitlement Scale. The highest-loading factor is indicated by boldface.

SD = 19.36, α = 0.95) in a random order as part of a larger study, and were debriefed at the end of the study.

incorrect or confused responses resulting from the reverse-scoring process (e.g., Hughes, 2009), or that the reverse of entitlement is not the opposite of entitlement. Others have also made the case for removing reverse-scored items. For example, when researching the Social Interaction Anxiety Scale (SIAS), Rodebaugh, Woods, and Heimberg (2007) found that the reverse-scored items on the SIAS created problematic psychometric properties for the measure; they recommended not using reverse-scored items for the measurement of social anxiety. In Study 2, we sought to confirm the factor structure of the IES, PES, and NGS in a larger sample by including and removing PES Item 5. We also tested alternate models (13 item and 6 item versions) of the NGS based on recent research by Crowe et al. (2016). We tested these alternative models due to the finding that the condensed NGS showed adequate reliability and represents a more parsimonious version of the instrument.

6. Study 2: results and discussion A confirmatory factor analysis was conducted using AMOS 19, testing the model based on the Study 1 analysis. In this initial model (Model 1), interpersonal exploitativeness (6 items), grandiosity (16 items), and psychological entitlement (all 9 items included) were used in the analysis. As this model did not adequately fit the data (see below), several alternative models were then tested to find the bestfitting and most parsimonious model. Model 2 was tested based on Study 1 PCA results (which found PES item 5 did not load along with the remaining items). The remaining models (Model 3, Model 4a, and Model 4b) tested the Crowe et al. (2016) finding that condensed versions of the NGS (13 items and 6 items) were reliable and valid versions of the full NGS. Specifically, Model 3 was based on Crowe et al. (2016) in which only 13 items from the NGS were retained (Items 2, 6, and 13 removed and the PES Item 5 removed); Model 4A was based on Crowe et al. (2016) in which only 6 items from the NGS were retained (only Items 8, 9, 10, 11, 12, 16 retained, PES Item 5 removed); and Model 4B, which is based on Model 4A, but with the NGS, PES, and IES treated as components of a latent narcissism variable (Model 4B). Finally, our last model, Model 5, reflects a relationship between exploitativeness, entitlement, and grandiosity where they are all part of the overall construct narcissism and do not reflect distinct but correlated constructs. In this last model, there were no facets of narcissism, and PES item 5 was removed). For each model, the following fit indices were used to estimate model fit: Normed Fit Index (NFI) ≥ 0.95, Comparative Fit Index (CFI) ≥ 0.95, Tucker Lewis Index (TLI) ≥0.95, and the Root Mean Square of Approximation (RMSEA) ≤ 0.06 (Hu & Bentler, 1999). The Akaike Information Criterion (AIC) was used to compare fit across the tested

5. Study 2: methods 5.1. Participants A sample of 842 Introductory Psychology students (362 males, 455 females, and 25 individuals who did not indicate their sex) at a regional campus of a large Midwestern University participated in the study in exchange for partial course credit. Participants were 18.87 years old on average (SD = 2.26), and 67.2% self-identified as Caucasian. 5.2. Materials and procedure The study was approved by the university's Institutional Review Board, and all participants provided informed consent. Participants again completed the standard IES (M = 12.89, SD = 7.20, α = 0.88), PES (M = 27.86, SD = 10.46, α = 0.85), and NGS (M = 47.61, 184

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Table 2 Summary of model fit indices. Study 2

Study 3

Model 1

Model 2

Model 3

Model 4A

Model 4B

Model 5B

Model 4A

Model 4B

IES PES NGS Latent Narc

6 items 9 items 16 items No

6 items 8 items 16 items No

6 items 8 items 13 items No

6 items 8 items 6 items No

6 items 8 items 6 items Yes

6 items 8 items 6 items Yes

6 items 8 items 6 items No

6 items 8 items 6 items Yes

Statistic NFI CFI TLI RMSEA 95% CI AIC

0.881 0.905 0.897 0.064 0.061–0.067 2030.826

0.884 0.907 0.899 0.065 0.062–0.068 1961.751

0.894 0.915 0.907 0.065 0.062–0.068 1575.187

0.934 0.950 0.943 0.058 0.053–0.062 717.562

0.934 0.950 0.943 0.058 0.053–0.062 717.562

0.575 0.584 0.536 0.165 0.160–0.169 4128.276

0.911 0.946 0.938 0.060 0.053–0.068 490.749

0.911 0.946 0.938 0.060 0.053–0.068 490.749

Note: IES = Interpersonal Exploitativeness Scale; PES = Psychological Entitlement Scale; NGS = Narcissistic Grandiosity Scale; Latent Narc = yes/no was narcissism included as a latent variable; NFI = Normed Fit Index; CFI = Comparative Fit Index; TLI = Tucker Lewis Index; RMSEA = Root Mean Square of Approximation with 95% confidence interval; AIC = Akaike Information Criterion.

models, with lower values indicating a better fitting model. Table 2 includes the fit indices for the six models. Model 1 was not an adequate fit for the data, as NFI, CFI, and TLI fell below 0.95 and RMSEA was above 0.06. Removing PES Item 5 (Model 2) improved the fit indices and lowered the AIC value, but NFI and TLI remained below 0.95 and RMSEA above 0.06. Model 3, in which only 13 NGS items were retained, again showed an improved fit from Model 2 but all values were under the suggested cut-off values. Model 4A, in which only six NGS items were retained, showed an improved fit on all indices and the lowest AIC value of the models tested. NFI, CFI, and TLI approximated 0.95, and RMSEA was below 0.06. The model testing a broader latent structure of these three scales (Model 4B) did not increase or decrease model fit from Model 4A. The last model, Model 5, removed the narcissism facets and instead examined one latent narcissism variable. This model was an overall poor fit for the data, with fit indices falling well below 0.95. Taken together, Models 4A and 4B best fit the data as they reflect the same model but with the addition of a latent narcissism variable (4B). This final model included the six items from the IES, eight items from the PES, and six items from the NGS (consistent with Crowe et al., 2016). Fig. 1 depicts Model 4A. Correlations among the IES, PES, and NGS6 for all studies are displayed in Table 3. Studies 1 and 2 provided further evidence that entitlement, exploitativeness, and grandiosity represent three distinct, but correlated, psychometrically sound constructs. Furthermore, these three constructs contain content and face validity. The models that provided the best fit for the data in Study 2 were also the most efficient, as it reduced the total number of items to 20 from 31. Using these 20 items would enable researchers to more quickly assess narcissistic traits, and the information about these three narcissistic traits can in turn inform us about narcissism. In Study 3, we used these 20 items to examine the convergent and discriminant validity of the IES, PES, and NGS6 by correlating them with other relevant constructs, including alternative measures of trait narcissism. We utilized three measures of grandiose narcissism, two measures of vulnerable narcissism, and a measure of the Dark Triad that is composed of narcissism and the related constructs of Machiavellianism and psychopathy. In addition, we examined the extent to which the PES, IES, and NGS6 were related to other relevant constructs. Narcissism is associated with a lack of empathy for others (e.g., Foster, McCain, Hibberts, Brunell, & Johnson, 2015; Hepper, Hart, & Sedikides, 2014; Pincus et al., 2009) and low agreeableness (Miller & Maples, 2011). Grandiose narcissists are described as having high self-esteem (Brown, Freis, Carroll, & Arkin, 2016) and psychological well-being (Ng, Cheung, & Tam, 2014), being approach-oriented rather than avoidance-oriented (Foster & Brennan, 2011), and a tendency to be impulsive (Vazire &

Funder, 2006). Vulnerable narcissists are described as having low selfesteem (Brown et al., 2016) and psychological well-being (Ng et al., 2014). Thus, in the present study, we examine how the PES, IES, and NGS6 relate to constructs such as empathy, Big Five personality, and psychological well-being (comprising global self-esteem, depression, and anxiety). In a previous test of the NGS and the PES, Miller et al. (2012) found that these two measures failed to capture relevant variance in narcissism-relevant measures. They argue, for example, that the Narcissistic Personality Inventory (NPI; Raskin & Terry, 1988) correlates with the Five Factor Model traits that are consistent with what experts deem as prototypical of Narcissistic Personality Disorder (Lynam & Widiger, 2001; Miller & Campbell, 2008; Miller et al., 2009) whereas using the NGS and PES, even in combination with other measures that assess grandiosity (i.e., the grandiose narcissism subscale of the Pathological Narcissism Inventory, PNI, Pincus et al., 2009), did not adequately capture all the narcissism-related variance in FFM trait ratings. Thus, in Study 3, we also examine the extent to which the NPI predicts these external correlates above and beyond the IES, PES, and NGS6. 7. Study 3: methods 7.1. Participants Participants were 448 Introductory Psychology students (146 males, 207 females, and 95 individuals who did not indicate their sex) at a small regional campus of a large Midwestern University who participated in exchange for partial course credit. Participants were 18.96 years old on average (SD = 2.79; only 317 participants indicated their age), and 64.7% self-identified as Caucasian (44 participants did not indicate their ethnicity). 8. Materials and procedure The study was approved by the university's Institutional Review Board, and all participants provided informed consent. As in Studies 1 and 2, participants completed the IES (M = 10.04, SD = 8.35, α = 0.87), PES (M = 28.10, SD = 9.43, α = 0.84), and NGS (MNGS6 = 17.50, SD = 8.24, α = 0.93) in a randomized order as part of a larger study. As part of the study battery, additional measures of narcissism and related constructs were assessed in a randomized order. 8.1. Narcissism and related measures 8.1.1. Narcissistic personality inventory (NPI) The NPI is the most commonly used measure of grandiose 185

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Fig. 1. Confirmatory factor analysis of Model 4A (Study 2). IES = Interpersonal Exploitativeness Scale, PES = Psychological Entitlement Scale, NGS6 = the 6-item version of the Narcissistic Grandiosity Scale.

levels of grandiose narcissism. Mean NPI score in the present sample was 14.92 (SD = 7.34). Internal consistency was high (α = 0.86).

narcissism. The NPI contains 40 items (Raskin & Terry, 1988) that ask participants to choose between two options (e.g., “I think I am a special person” versus “I am no better or worse than most people”). Narcissistic responses are assigned scores of 1 and non-narcissistic responses are assigned a score of 0. Scores are summed; higher scores reflect higher

8.1.2. Grandiose narcissism scale (GNS) The GNS (Foster et al., 2015) is a 33-item measure of grandiose 186

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Table 3 Correlations between the IES, PES, NGS6, in Studies 1–3. Study 1

IES PES (item 5 deleted)

Study 2

Study 3

Studies 1–3 combined

PES

NGS6

PES

NGS6

PES

NGS6

PES

NGS6

0.477⁎ –

0.370⁎ 0.515⁎

0.445⁎ –

0.364⁎ 0.487⁎

0.238⁎ –

0.318⁎ 0.472⁎

0.385⁎ –

0.348⁎ 0.488⁎

Note: IES = Interpersonal Exploitativeness Scale; PES = Psychological Entitlement Scale; NGS = Narcissistic Grandiosity Scale. ⁎ p < 0.001

commonly used to measure Neuroticism (e.g., “I see myself as someone who worries a lot;” α = 0.79; M = 4.06, SD = 1.37), Extraversion (e.g., “I see myself as someone who is talkative;” α = 0.78; M = 4.32, SD = 1.38), Openness to Experience (e.g., “I see myself as someone who is curious about many different things;” α = 0.71; M = 4.73, SD = 1.11), Conscientiousness (e.g., “I see myself as someone who does a thorough job;” α = 0.72; M = 4.76, SD = 1.12), and Agreeableness (e.g., “I see myself as someone who is considerate and kind to almost everyone;” α = 0.83; M = 5.38, SD = 1.24) using 5-point scales, such that 1 = disagree strongly and 5 = agree strongly. Scores are computed by averaging the items on each subscale; higher scores represent higher levels of each personality trait.

narcissism that was developed to more reliably measure the global- and facet-levels of grandiose narcissism. Responses are made using a 6-point scale such that 1 = strongly disagree and 6 = strongly agree. A sample item is “I'm more talented than most other people.” Total scores achieved high internal consistency (α = 0.94, M = 93.38, SD = 28.06). 8.1.3. Pathological narcissism inventory (PNI) The PNI assesses grandiose and vulnerable narcissism ((Pincus et al., 2009). Responses are made on 6-point scales ranging from 0 (not at all like me) to 5 (very much like me). A sample items assessing grandiose narcissism is: “I often fantasize about being recognized for my accomplishments.” (For the present sample, α = 0.90; M = 66.13, SD = 20.21.) A sample items assessing vulnerable narcissism is: “It's hard for me to feel good about myself unless I know other people like me.” (For the present sample, α = 0.91; M = 55.26, SD = 24.00.)

8.2.4. Impulsive sensation seeking (ImpSS) Individual differences in impulsivity and sensation seeking were assessed with the 19-item ImpSS subscale of the Zuckerman-Kuhlman Personality Questionnaire (Zuckerman, Kuhlman, Joireman, Teta, & Kraft, 1993). Eight items assess impulsivity (e.g., “I often do things on impulse,” α = 0.67; M = 0.36, SD = 0.26) and eleven items assess sensation seeking (e.g., “I'll try anything once,” α = 0.74; M = 0.60, SD = 0.25). Average scores are calculated for each subscale, with higher scores indicating greater levels of each trait.

8.1.4. Hypersensitive narcissism scale (HSNS) Vulnerable narcissism was also assessed with the HSNS (Hendin & Cheek, 1997). A sample item is: “I dislike sharing the credit of an achievement with others.” Responses are made on 5-point scales such that 1 = very uncharacteristic or untrue/strongly disagree and 5 = very characteristic or true/strongly agree. Responses are summed to create a total score. For the present sample, α = 0.76; M = 28.46, SD = 6.58.

8.2.5. Behavioral inhibition/behavioral activation scales (BIS/BAS) Behavioral inhibition (sensitivity to punishment) and activation (sensitivity to reward) were assessed with the 24-item BIS/BAS scale (Carver & White, 1994). Performance on four subscales was calculated: BIS (e.g., “Criticism or scolding hurts me quite a bit,” α = 0.70; M = 21.02, SD = 3.68), BAS-Drive (e.g., “I go out of my way to get things I want,” α = 0.78; M = 11.31, SD = 2.57), BAS-Fun Seeking (e.g., “I often act on the spur of the moment,” α = 0.70; M = 12.45, SD = 2.28), and BAS-Reward Responsiveness (e.g., “It would excite me to win a contest,” α = 0.59; M = 17.36, SD = 2.20). Summed total scores are calculated for each subscale, with higher scores indicating greater levels of each trait.

8.1.5. Dirty dozen The Dirty Dozen (Jonason & Webster, 2010) is a 12-item measure of the Dark Triad personality traits of Grandiose Narcissism (e.g., “I tend to want others to admire me,” α = 0.86; M = 15.39, SD = 7.86), Machiavellianism (e.g., “I have used deceit or lied to get my way,” α = 0.88; M = 12.22, SD = 7.04), and Psychopathy (e.g., “I tend to lack remorse,” α = 0.82; M = 6.60, SD = 6.02). Responses are made on 5-point scales such that 1 = not at all and 5 = very much, with summed totals created for each of the three traits. 8.2. External correlates

8.2.6. Adult manifest anxiety scale (AMAS), college student version The AMAS (Reynolds, Richmond, & Lowe, 2003) was utilized to assess levels of current anxiety. A sample item is “I am often described as restless.” Participants respond to each item with a yes or no response. Total scores are calculated; with higher scores indicating greater levels of anxiety (α = 0.93; M = 26.80, SD = 10.22).

8.2.1. Interpersonal reactivity index (IRI) The empathic concern subscale of the IRI (Davis, 1980) was used to assess empathy. A sample item is “When I see someone being taken advantage of, I feel kind of protective toward them” (α = 0.77; M = 19.51, SD = 4.51). Responses are assessed with 5-point scales such that 0 = does not describe me well and 4 = describe me very well. Summed total scores were calculated for each scale.

8.2.7. Beck depression inventory-II (BDI-II) The BDI-II was utilized to assess current symptoms of depression (Beck, Steer, & Brown, 1996). Participants respond to a series of 21 items (“Sadness,” α = 0.91; M = 12.08, SD = 9.84) on a scale from 0 to 3, with higher scores indicating greater depressive symptoms. Total scores were calculated.

8.2.2. Rosenberg self-esteem scale (RSES) The RSES (Rosenberg, 1965) was used to assess global self-esteem. A sample item is “I feel that I have a number of good qualities.” Respondents use 5-point scales to agree or disagree with each statement (1 = strong disagreement; 5 = strong agreement). For the present sample total scores, α = 0.85; M = 36.86, SD = 7.71.

9. Study 3: results and discussion

8.2.3. The big five inventory (BFI) The BFI (Benet-Martinez & John, 1998) consists of 44 items and is

Although our sample for Study 3 was significantly smaller in size than our Study 2 sample, we first sought to replicate our CFA Models 4A 187

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

2015) modeled the GNS after the original seven-factor solution of the NPI (Raskin & Terry, 1988). It is also noteworthy that the IES, PES, and NGS6 account for variance even when the NPI is entered into the equation. For example, the IES uniquely predicted greater psychopathy, lower empathic concern, and lower agreeableness when the PES, NGS6 and NPI were also entered into the equation whereas the NPI either did not uniquely predict these variables while the IES, PES, and NGS6 were in the equation, or accounted for little variance. It is also noteworthy that the NPI sometimes predicted related variables in opposing directions than the IES, PES, or NGS6. For example, the NPI, IES, and PES predicted greater Machiavelliansism, but the NGS6 predicted less Machiavellianism. The NPI and the NGS6 predicted higher self-esteem and lower depression, but the IES predicted slightly lower self-esteem and higher depression. The NPI predicted higher conscientiousness and openness to experience, whereas the IES was associated with less conscientiousness and openness to experience. Taken together, while we agree with Miller et al. (2012) that the NPI is still relevant, we also believe that these lower-order trait measures capture valuable information that gets lost when relying only on NPI total scores.

Table 4 Study 3: Correlations between the IES, PES, NGS6, and other variables. Correlations

Narcissism measures NPI GNS PNI grandiose narcissism PNI vulnerable narcissism HSNS Dirty dozen narcissism Dirty dozen mach Dirty dozen psychopathy External correlates IRI empathic concern RSES Extraversion Neuroticism Agreeableness Openness Conscientiousness ImpSS impulsivity ImpSS sensation seeking BIS BAS drive BAS fun seeking BAS reward responsiveness Beck depression inventory AMAS total

IES

PES

NGS6

0.287⁎⁎⁎a 0.514⁎⁎⁎a 0.203⁎⁎⁎a 0.072a 0.138⁎⁎a 0.208⁎⁎⁎ab 0.389⁎⁎⁎a 0.625⁎⁎⁎a

0.394⁎⁎⁎ab 0.414⁎⁎⁎a 0.275⁎⁎⁎a 0.114⁎a 0.177⁎⁎⁎a 0.324⁎⁎⁎a 0.325⁎⁎⁎a 0.239⁎⁎b

0.474⁎⁎⁎b 0.478⁎⁎⁎a 0.208⁎⁎⁎a −0.083b 0.080a 0.187⁎⁎⁎b 0.181⁎⁎⁎b 0.213⁎⁎⁎b

− 0.430⁎⁎⁎a 0.030a − 0.009a − 0.039a − 0.298⁎⁎⁎a − 0.117a − 0.139a 0.248⁎⁎⁎a 0.131⁎a − 0.163⁎⁎⁎ab 0.102⁎a 0.108⁎a − 0.005a 0.029a −.024a

− 0.263⁎⁎⁎b 0.172⁎⁎⁎b 0.041a − 0.038a − 0.225⁎⁎a 0.025b 0.010b 0.159⁎a 0.111⁎a − 0.151⁎⁎a 0.259⁎⁎⁎b 0.122⁎a 0.105⁎ab − 0.075a − 0.045a

−0.204⁎⁎b 0.339⁎⁎⁎c 0.140⁎b −0.229⁎⁎b −0.209⁎a 0.030b 0.079b 0.140⁎⁎a 0.100⁎a −0.276⁎⁎⁎b 0.370⁎⁎⁎b 0.204⁎⁎⁎a 0.149⁎⁎b −0.265⁎⁎⁎b −0.127⁎a

10. General discussion The IES, PES, and NGS are reliable measures of three separate, distinct constructs—as they were originally designed—rather than one overarching construct of narcissism. Although not all aspects (e.g., vanity, exhibitionism) of grandiose narcissism are covered in these three dimensions, they reliably measure what are arguably the most important aspects of the “dark side” of grandiose narcissism (i.e., entitlement, exploitativeness). In addition, although our data suggest that each of the three traits measure aspects of narcissism, they also tend to correspond to different external measures. Grandiosity was associated with higher extraversion, but exploitativeness and entitlement were not. Exploitativeness and entitlement were associated with lower agreeableness, but grandiosity was not. Exploitativeness and entitlement were not associated with depression, but grandiosity was associated with less depression and anxiety. Thus, it appears that grandiosity is the buffer against threats to mental health among narcissists. These findings are consistent with other findings using a dimensional approach. For example, Brunell et al. (2013) found that exploitativeness corresponded to less cooperative behavior over time in a commons dilemma, but entitlement did not. In a similar vein, Tamborski, Brown, and Chowning (2012) found that grandiosity, but not entitlement, was associated with unrealistic optimism (i.e., expecting positive events to happen to a greater extent and negative events to happen to a lesser extent), but entitlement, but not grandiosity, was associated with unethical decision-making. Buelow and Brunell (2014) found that grandiosity and exploitativeness, but not entitlement, were associated with greater self-reports of risk-taking, but entitlement was associated with a preference for smaller, more immediate gains over larger, temporally distant rewards. Grandiose narcissism is understood as a multifaceted construct (Raskin & Terry, 1988) and as such should be explored with a dimensional approach for those interested in how specific dimensions are associated with behaviors and outcomes. The present results have important implications for the assessment of narcissism by personality-social and clinical psychologists alike. Assessment of grandiose narcissism with only the NPI has been frequently critiqued (Ackerman et al., 2011; Brown et al., 2009; Brunell et al., 2013; Campbell et al., 2004; Rosenthal & Hooley, 2010), and our results indicate that utilizing the IES, PES, and NGS—potentially in conjunction with other narcissism measures—can provide additional information about the “dark side” of grandiose narcissism. We agree with Miller and colleagues (e.g., Miller et al., 2012) that the NPI is still relevant as it captures variance that is not captured with narrower trait measures, such as the agentic, approach and reward-oriented aspects of

Note: IES = Interpersonal Exploitativeness Scale; PES = Psychological Entitlement Scale; NGS = Narcissistic Grandiosity Scale; NPI = Narcissistic Personality Inventory; GNS = Grandiose Narcissism Scale; PNI = Pathological Narcissism Inventory; HSNS = Hypersensitive Narcissism Scale; IRI = Interpersonal Reactivity Index; RSES = Rosenberg Self-Esteem Scale; IMPSS = Impulsive Sensation Seeking; BIS = Behavioral Inhibition; BAS = Behavioral Activation Scale; AMAS = Adult Manifest Anxiety Scale. Correlations with different subscripts are significantly different from one another. ⁎ p ≤ 0.05. ⁎⁎ p ≤ 0.01. ⁎⁎⁎ p ≤ 0.001.

(no broader latent structure) and 4B (assessing broader latent structure) from Study 2. See Table 2 for a summary of results. Although the values are slightly below the recommended cut-off values, the fit indices suggest adequate fit of the data. We next computed the intercorrelations among the IES, the PES (removing the reversed-scored item 5), and the NGS6 in Table 3. We also present these correlations from Studies 1 and 2, as well as across all three studies. Intercorrelations among these three variables are small to moderate. Correlations between the NGS, PES, and IES and narcissism and narcissism-related measures are reported in Table 4. Fisher's r to z transformations were computed to determine if the correlations significantly differ from each other in their prediction of the outcome variables. As expected, the IES, PES, and NGS were negatively associated with agreeableness, empathic concern, and avoidance motivation. The IES, PES, and NGS were positively associated with approach motivation and impulsivity. Grandiosity was negatively associated with the measure of anxiety and depression, replicating that grandiosity is the dimension associated with more positive mental health (Brown et al., 2009). Lastly, we predicted narcissism measures and related measures by entering the IES, PES, and NGS6 into regression models at Step 1 and then adding the NPI the models at Step 2. We examined if collinearity was a problem with the analyses, but found that it was not (VIF < 2 for all analyses). The summary of the regression analyses are presented in Table 5. We note that the NPI frequently accounts for variance in measures above and beyond the IES, PES, and NGS6. For the GNS, specifically, this is to be expected as the measure authors (Foster et al., 188

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Table 5 Study 3: Regressions predicting narcissism and narcissism-related constructs from the IES, PES, NGS6, and the NPI. Betas Step 1

Betas Step 2

IES

PES

NGS6

R2

IES

PES

NGS6

NPI

R2

ΔR2

Narcissism measures NPI GNS PNI GN PNI VN HSNS Dirty Dozen Narcissism Dirty Dozen Mach Dirty Dozen Psychopathy

0.116⁎ 0.366⁎⁎⁎ 0.129⁎ 0.072 0.097# 0.115⁎ 0.325⁎⁎⁎ 0.611⁎⁎⁎

0.196⁎⁎⁎ 0.170⁎⁎⁎ 0.204⁎⁎⁎ 0.181⁎⁎ 0.160⁎⁎ 0.280⁎⁎⁎ 0.240⁎⁎⁎ 0.047

0.345⁎⁎⁎ 0.274⁎⁎⁎ 0.080 − 0.186⁎⁎ − 0.027 0.015 − 0.038 − 0.005

0.274 0.389 0.099 0.041 0.039 0.118 0.198 0.392

___ 0.319⁎⁎⁎ 0.090# 0.077 0.108⁎ 0.079 0.284⁎⁎⁎ 0.598⁎⁎⁎

___ 0.078# 0.147⁎⁎ 0.187⁎⁎ 0.182⁎⁎ 0.216⁎⁎⁎ 0.168⁎⁎⁎ 0.023

___ 0.090⁎ − 0.036 − 0.175⁎⁎ 0.017 − 0.089 − 0.156⁎⁎ − 0.044

___ 0.476⁎⁎⁎ 0.325⁎⁎⁎ − 0.033 − 0.115# 0.315⁎⁎⁎ 0.354⁎⁎⁎ 0.117⁎

___ 0.547 0.177 0.042 0.049 0.190 0.289 0.402

___ 0.158⁎⁎⁎ 0.078⁎⁎⁎ 0.001 0.009# 0.072⁎⁎⁎ 0.092⁎⁎⁎ 0.010⁎

External correlates IRI Empathic Concern RSES Extraversion Neuroticism Agreeableness Openness Conscientiousness ImpSS Impulsivity ImpSS Sensation Seeking BIS BAS Drive BAS Fun Seeking BAS Reward BDI AMAS

− 0.385⁎⁎⁎ − 0.092# 0.169⁎ 0.041 − 0.228⁎⁎ − 0.169⁎ − 0.209⁎ 0.213⁎⁎⁎ 0.100# − 0.093# − 0.016 0.053 − 0.057 0.098# 0.025

− 0.101 0.036 − 0.003 0.063 − 0.099 0.064 0.031 0.075 0.061 − 0.008 0.110# 0.023 0.055 0.045 0.015

− 0.007 0.351⁎⁎⁎ − 0.074 − 0.272⁎⁎ − 0.077 0.068 0.147 0.037 0.040 − 0.247⁎⁎⁎ 0.322⁎⁎⁎ 0.179⁎⁎ 0.139⁎ − 0.312⁎⁎⁎ − 0.144⁎

0.194 0.122 0.024 0.058 0.106 0.023 0.041 0.070 0.024 0.085 0.146 0.045 0.027 0.082 0.017

− 0.386⁎⁎⁎ − 0.120⁎ − 0.160⁎ 0.078–0.235⁎⁎

− 0.102 − 0.011 − 0.128 0.116 − 0.109 − 0.012 − 0.027 0.033 − 0.004 0.027 0.039 − 0.045 0.014 0.072 0.067

− 0.008 0.269⁎⁎⁎ − 0.073 − 0.169# − 0.096 − 0.080 0.035 − 0.037 − 0.075 − 0.166⁎⁎ 0.157⁎⁎ 0.023 0.044 − 0.257⁎⁎⁎ − 0.046

0.003 0.239⁎⁎⁎ 0.597⁎⁎⁎ − 0.253⁎⁎ 0.047 0.364⁎⁎⁎ 0.276⁎⁎ 0.215⁎⁎⁎ 0.333⁎⁎⁎ − 0.207⁎⁎ 0.423⁎⁎⁎ 0.401⁎⁎⁎ 0.242⁎⁎⁎ − 0.157⁎⁎ − 0.240⁎⁎⁎

0.194 0.164 0.247 0.098 0.108 0.106 0.089 0.103 0.104 0.115 0.274 0.159 0.068 0.101 0.056

0.000 0.042⁎⁎⁎ 0.223⁎⁎⁎ 0.040⁎⁎ 0.001 0.083⁎⁎⁎ 0.048⁎⁎ 0.033⁎⁎⁎ 0.081⁎⁎⁎ 0.031⁎⁎ 0.127⁎⁎⁎ 0.114⁎⁎⁎ 0.042⁎⁎⁎ 0.018⁎⁎ 0.039⁎⁎⁎

− 0.221⁎⁎ − 0.249⁎⁎ 0.188⁎⁎⁎ 0.061 − 0.070 − 0.063 0.008 − 0.085 0.117⁎ 0.038

Note: IES = Interpersonal Exploitativeness Scale; PES = Psychological Entitlement Scale; NGS = Narcissistic Grandiosity Scale; NPI = Narcissistic Personality Inventory; GNS = Grandiose Narcissism Scale; PNI = Pathological Narcissism Inventory; HSNS = Hypersensitive Narcissism Scale; IRI = Interpersonal Reactivity Index; RSES = Rosenberg SelfEsteem Scale; IMPSS = Impulsive Sensation Seeking; BIS = Behavioral Inhibition; BAS = Behavioral Activation Scale; AMAS = Adult Manifest Anxiety Scale. ⁎ p ≤ 0.05. ⁎⁎ p ≤ 0.01. ⁎⁎⁎ p ≤ 0.001.

11. Conclusions

narcissism. Also as Miller and colleagues (e.g., Miller, Lynam, Hyatt, & Campbell, 2017) have demonstrated, despite its limitations, the NPI is associated with traits associated with personality disorders and other psychopathological constructs and is also consistent with expert ratings of Narcissistic Personality Disorder and grandiose narcissism. However, we recommend that researchers continue to develop and test measures that adequately measure the facets of narcissism because such investigations can potentially illuminate ways in which grandiose and vulnerable narcissism intersect (e.g., they both share entitlement) and the ways that they diverge (e.g., grandiose narcissism is associated with higher self-esteem and vulnerable narcissism is associated with lower self-esteem). Likewise, it is not uncommon for narcissistic traits to predict different outcomes. As Study 3 demonstrated, both the NPI and the NGS6 predicted lower levels of depression (and higher self-esteem), but the IES predicted higher levels of depression (and lower selfesteem). A better understanding of these dimensions would allow researchers and clinicians to gain a more solid understanding of how narcissism affects daily life. One limitation of the present study was that the factor analytic results could be affected by method variance. For example, the NGS uses a singular adjective rating approach that is considerably different from agreeing or disagreeing with statements as is done with the PES and IES. A second limitation was the college student sample. This factor structure therefore may or may not replicate in an older, non-college student sample. Future research is needed to address this issue. In addition, we did not assess all external correlates shown to be associated with narcissism. Future studies should examine how entitlement, exploitativeness, and grandiosity are associated with factors such as aggression, cheating, and self-enhancement to name a few. Future research should also continue to examine how the dimensions of narcissism predict behavioral outcomes.

The PES, IES, and NGS reliably measure three distinct traits of narcissism rather than one overall construct. Use of an eight-item version of the PES (removing the reverse-scored item), IES, and the sixitem version of the NGS efficiently measure these traits. These traits correspond with established measures of narcissism and uniquely predict relevant variables, such as empathy, extraversion, agreeableness, anxiety, and depression, demonstrating convergent and discriminant validity. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.paid.2017.11.029. References Ackerman, R. A., Witt, E. A., Donnellan, M. B., Trzesniewski, K. H., Robins, R. W., & Kashy, D. A. (2011). What does the narcissistic personality inventory really measure? Assessment, 18, 67–87. Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994). Insensitivity to future consequences following damage to human prefrontal cortex. Cognition, 50, 7–15. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck depression inventory-II. San Antonio, TX: Psychological Corporation. Benet-Martinez, V., & John, O. P. (1998). Los cinco grandes across cultures and ethnic groups: Multitrait multimethod analyses of the big five in Spanish and English. Journal of Personality and Social Psychology, 75, 729–750. Brown, A. A., Freis, S. D., Carroll, P. J., & Arkin, R. M. (2016). Perceived agency mediates the link between the narcissistic subtypes and self-esteem. Personality and Individual Differences, 90, 124–129. Brown, R. P., Budzek, K., & Tamborski, M. (2009). On the meaning and measure of narcissism. Personality and Social Psychology Bulletin, 35, 951–964. Brunell, A. B., & Buelow, M. T. (2017). Narcissism and performance on behavioral decision-making tasks. Journal of Behavioral Decision Making, 30, 3–14.

189

Personality and Individual Differences 123 (2018) 182–190

A.B. Brunell, M.T. Buelow

Miller, J. D., Lynam, D. R., Hyatt, C. S., & Campbell, W. K. (2017). Controversies in narcissism. Annual Review of Clinical Psychology, 13, 291–315. Miller, J. D., & Maples, J. (2011). Trait personality models of narcissistic personality disorder, grandiose narcissism, and vulnerable narcissism. In W. K. Campbell, & J. D. Miller (Eds.). The handbook of narcissism and narcissistic personality disorder (pp. 71– 88). Hoboken, NJ: John Wiley & Sons, Inc. Miller, J. D., Price, J., & Campbell, W. K. (2012). Is the narcissistic personality inventory still relevant? A test of independent grandiosity and entitlement scales in the assessment of narcissism. Assessment, 19, 8–13. Ng, H. K. S., Cheung, R. Y., & Tam, K. (2014). Unraveling the link between narcissism and psychological health: New evidence from coping flexibility. Personality and Individual Differences, 70, 7–10. O'Connor, B. P. (2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer's MAP test. Behavior Research Methods, Instrumentation, and Computers, 32, 396–402. Pincus, A. L., Ansell, E. B., Pimentel, C. A., Cain, N. M., Wright, A. G. C., & Levy, K. N. (2009). Initial construction and validation of the pathological narcissism inventory. Psychological Assessment, 21, 365–379. Raskin, R. N., & Terry, H. (1988). A principal components analysis of the narcissistic personality inventory and further evidence of its construct validity. Journal of Personality and Social Psychology, 54, 890–902. Reynolds, C. R., Richmond, B. O., & Lowe, P. A. (2003). The adult manifest anxiety scale manual. Los Angeles: Western Psychological Services. Rodebaugh, T. L., Woods, C. M., & Heimberg, R. G. (2007). The reverse of social anxiety is not always the opposite: The reverse-scored items of the social interaction anxiety scale do not belong. Behavior Therapy, 38, 192–206. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Rosenthal, S. A., & Hooley, J. M. (2010). Narcissism assessment in social-personality research: Does the association between narcissism and psychological health result from a confound with self-esteem? Journal of Research in Personality, 44, 453–465. Rosenthal, S. A., Hooley, J. M., & Steshenko, Y. (2007). Distinguishing grandiosity from selfesteem: Development of the narcissistic grandiosity scale. (Unpublished manuscript). Smith, G. T., McCarthy, D. M., & Zapolski, T. C. B. (2009). On the value of homogeneous constructs for construct validation, theory testing, and the description of psychopathology. Psychological Assessment, 21, 272–284. Tamborski, M., Brown, R. P., & Chowning, K. (2012). Self-serving bias or simply serving the self? Evidence for a dimensional approach to narcissism. Personality and Individual Differences, 52, 942–946. Vazire, S., & Funder, D. C. (2006). Impulsivity and the self-defeating behavior of narcissists. Personality and Social Psychology Review, 10, 154–165. Velicer, W. F. (1976). Determining the number of components from the matrix of partial correlations. Psychometrika, 41, 321–327. Zuckerman, M., Kuhlman, D., Joireman, J., Teta, P., & Kraft, M. (1993). A comparison of three structural models for personality: The big three, the big five, and the alternative five. Journal of Personality and Social Psychology, 65, 757–768.

Brunell, A. B., Davis, M. S., Schley, D. R., Eng, A., van Dulmen, M. H. M., Wester, K. L., & Flannery, D. J. (2013). A new measure of interpersonal exploitativeness. Frontiers in Psychology, 4, 299. Brunell, A. B., Tumblin, L., & Buelow, M. T. (2014). Narcissism and the motivation to engage in volunteerism. Current Psychology, 33, 365–376. Buelow, M. T., & Brunell, A. B. (2014). Facets of grandiose narcissism predict involvement in health-risk behaviors. Personality and Individual Differences, 69, 193–198. Campbell, W. K., Bonacci, A. M., Shelton, J., Exline, J. J., & Bushman, B. J. (2004). Psychological entitlement: Interpersonal consequences and validation of a new selfreport measure. Journal of Personality Assessment, 83, 29–45. Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and Social Psychology, 67, 319–333. Crowe, M., Carter, N. T., Campbell, W. K., & Miller, J. D. (2016). Validation of the narcissistic grandiosity scale and creation of reduced item variants. Psychological Assessment, 28, 1550–1560. Daddis, C., & Brunell, A. B. (2015). Entitlement, exploitativeness and reasoning about everyday transgressions: A social domain analysis. Journal of Research in Personality, 58, 115–126. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Foster, J. D., & Brennan, J. C. (2011). Narcissism, the agency model, and approachavoidance motivation. In W. K. Campbell, & J. D. Miller (Eds.). The handbook of narcissism and narcissistic personality disorder (pp. 89–100). Hoboken, NJ: John Wiley & Sons, Inc. Foster, J. D., McCain, J. L., Hibberts, M. F., Brunell, A. B., & Johnson, R. B. (2015). The grandiose narcissism scale. Personality and Individual Differences, 73, 12–16. Hendin, H. M., & Cheek, J. M. (1997). Assessing hypersensitive narcissism: A reexamination of Murray's narcissism scale. Journal of Research in Personality, 31(4), 588–599. Hepper, E. G., Hart, C. M., & Sedikides, C. (2014). Moving narcissus: Can narcissists be empathic? Personality and Social Psychology Bulletin, 40, 1079–1091. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Hughes, G. D. (2009). The impact to incorrect response to reverse-coded survey items. 16, MidSouth Educational Research Association76–88. Jonason, P. K., & Webster, G. D. (2010). The dirty dozen: A concise measure of the dark triad. Psychological Assessment, 22, 420–432. Lynam, D. R., & Widiger, T. A. (2001). Using the five factor model to represent the DSMIV personality disorders: An expert consensus approach. Journal of Abnormal Psychology, 110, 401–412. Miller, J. D., & Campbell, W. K. (2008). Comparing clinical and social-personality conceptualizations of narcissism. Journal of Personality, 76, 449–476. Miller, J. D., Gaughan, E. T., Pryor, L. R., Kamen, C., & Campbell, W. K. (2009). Is research using the NPI relevant for understanding Narcissistic Personality Disorder? Journal of Research in Personality, 43, 482–488.

190