- Email: [email protected]
An exploration of the correlates of grandiose and vulnerable narcissism in romantic relationships: Homophily, partner characteristics, and dyadic adjustment
Personality and Individual Differences 79 (2015) 166–171
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
An exploration of the correlates of grandiose and vulnerable narcissism in romantic relationships: Homophily, partner characteristics, and dyadic adjustment Joanna Lamkin ⇑, W. Keith Campbell, Michelle R. vanDellen, Joshua D. Miller University of Georgia, Department of Psychology, Psychology Building, Athens, GA 30602-3013, United States
a r t i c l e
i n f o
Article history: Received 5 December 2014 Received in revised form 20 January 2015 Accepted 21 January 2015
Keywords: Narcissism Romantic relationships Homophily Grandiose Vulnerable
a b s t r a c t The present study examined the characteristics of individuals (N = 104 undergraduate couples) who date grandiosely or vulnerably narcissistic individuals, including the experience of developmental trauma, general and pathological personality traits, and psychopathology, using multiple data sources. In addition, relationship duration was tested as a moderator of the relations between the narcissism dimensions and relationship adjustment. Actor–Partner Interdependence Models indicated that negative relationship adjustment was found when both partners had higher entitlement/exploitativeness traits and had been together for a longer period of time. Overall, there were no clear patterns of partner characteristics, although some evidence for homophily emerged for traits related to grandiose narcissism. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Narcissistic individuals possess a sense of self-importance, believe themselves to be superior or special, are preoccupied with fantasies of success and admiration, and expect preferential treatment (American Psychiatric Association, 2013). Narcissism is also linked to a lack of empathy, sense of entitlement, and envy. More recent conceptualizations of narcissism emphasize the need to delineate between grandiose and vulnerable dimensions of narcissism (e.g., Miller & Campbell, 2008; Pincus & Lukowitsky, 2010), as these dimensions manifest divergent nomological networks (Miller et al., 2011). Grandiose narcissism is linked to extraversion, dominance, self-assurance, exhibitionism, and aggression; vulnerable narcissism is distinguished by introversion, defensiveness, anxiety, interpersonal coldness and hostility, as well as vulnerability to stress (Wink, 1991). Both narcissism dimensions, however, are associated with interpersonal dysfunction due, in part, to the antagonistic and intrusive behaviors associated with both. In the present research, we examine both dimensions in the context of romantic relationships. It is unclear whether narcissistic individuals seek out a certain type of romantic partner. Given the interpersonal antagonism associated with narcissism, one hypothesis is that narcissistic ⇑ Corresponding author. Tel.: +1 (706) 542 1173. E-mail address: [email protected] (J. Lamkin). http://dx.doi.org/10.1016/j.paid.2015.01.029 0191-8869/Ó 2015 Elsevier Ltd. All rights reserved.
individuals might seek out meek, passive individuals who make for more pliant partners (i.e., ‘‘victims’’). For example, perhaps partners with mood- or personality-related psychopathology and family histories of conflict or maltreatment might be seen by narcissistic individuals as easier to manipulate and thus appealing romantic partners. However, evidence from studies on homophily support another hypothesis, which suggests that grandiosely narcissistic individuals will seek partners who are similarly self-centered. Homophily, the idea that individuals partner with similar others, is observed throughout a wide range of human relationships (see McPherson, Smith-Lovin, & Cook, 2001). Several studies have tested for homophily in relation to grandiose narcissism with some studies finding evidence for it (Campbell, Foster, & Finkel, 2002; Keller et al., 2014; McNulty & Widman, 2013) and others not (e.g., Ryan, Weikel, & Sprechini, 2008). To date, there has been no published work examining relational homophily as it pertains to vulnerable narcissism. Using egocentric social network analysis, Lamkin, Clifton, Campbell, and Miller (2014) found that both vulnerable and grandiose narcissistic individuals described, on average, the individuals closest to them in their social networks as self-centered, narcissistic, and disagreeable—suggesting, at the very least, that there is some perception of homophily among both grandiosely and vulnerably narcissistic individuals. Campbell’s (1999) self-orientation model of narcissism and romantic attraction posits that individuals with narcissistic tendencies employ a self-regulatory strategy in which interpersonal
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
relationships serve to enhance and maintain their self-concept. Initial research provides some support for this model; for instance, even in a collectivistic culture that emphasizes communal values over self-focused goals (i.e., Thailand), narcissism predicted attraction to individuals who provided an opportunity for self-enhancement (Tanchotsrinon, Maneesri, & Campbell, 2007). Although all participants were attracted to caring individuals, grandiose participants were more attracted to partners who were described as admiring, attractive, and high status, and less attracted to otheroriented individuals who were described as caring and emotionally needy. At this time, little is known about the manner in which vulnerable narcissism is related to romantic functioning. However, given that vulnerable narcissism shares a basic tendency towards self-absorption and entitlement with grandiose narcissism, one might hypothesize that vulnerably narcissistic individuals would also prefer romantic partners who serve as sources of self-enhancement. Actually entering relationships that provide these opportunities may be more difficult for these individuals to achieve, however, due to their more neurotic and introverted personality style (Wink, 1991). Furthermore, vulnerable narcissism has been linked to anxiety, depression, and maladaptive attachment styles, which may also impact romantic relationships (Miller et al., 2011). Narcissism often has negative interpersonal consequences, including causing pain and suffering in romantic partners, family members, and friends (Miller, Campbell, & Pilkonis, 2007). Partners of narcissistic individuals incur long-term costs as a result of the decisions and behaviors enacted by their partners (Campbell & Campbell, 2009). Relationships with narcissistic individuals may start out positively, but then decline over time. For example, when asked to rate their relationship satisfaction at the beginning and end of their relationship, those who were dating narcissistic individuals endorsed a larger reduction in satisfaction than those who were not dating narcissistic individuals (Brunell & Campbell, 2011). This change in relationship satisfaction over time has been called the chocolate cake model (Campbell, 2005). In this model, chocolate cake serves as a metaphor for a relationship with someone who is narcissistic: initially, eating chocolate cake is enjoyable, but it comes with long-term costs (e.g., weight gain, discomfort) that lead to dissatisfaction. Of note, studies related to the chocolate cake model have typically used the Narcissistic Personality Inventory (NPI), which tends to measure the primarily grandiose features of narcissism and NPD (e.g., Miller et al., 2014) and thus the generalizability of these findings to vulnerable narcissism is unclear. Based on what is known about narcissism and romantic relationships, we developed two hypotheses and one research question. First, we predicted that partners of narcissistic individuals would also exhibit narcissistic characteristics. We further anticipated that grandiose and vulnerable narcissism would manifest homophily differentially through their respective personality traits (i.e., grandiosely narcissistic individuals would be more likely to date others high in extraversion; vulnerably narcissistic individuals would be more likely to date others characterized by introversion and neuroticism). Second, we explored the characteristics of those dating narcissistic individuals. Beyond homophily, there was no clear prediction to make as to whether grandiosely or vulnerably narcissistic individuals would have clear preferences for certain types of partners. We also included variables related to the ‘‘victim’’ hypothesis to test whether narcissistic individuals might seek out more vulnerable partners (e.g., histories of maltreatment; negative emotionality; personality difficulties). Third, consistent with the chocolate cake model, we hypothesized that partners of those with higher levels of grandiose narcissism would rate a relationship of shorter duration more positively than those
167
in a relationship of a longer duration. We hypothesized that this pattern would be weaker or nonexistent for individuals dating a vulnerably narcissistic individual, as these relationships may be less likely to begin with significant satisfaction. Additionally, we obtained peer reports about the relationships in order to have an alternative perspective on variables related to self-enhancement (e.g., social influence). 2. Method 2.1. Overview We used multiple measures to answers the questions of interest. Our focus was on the narcissism scores of ‘‘Partner A’’ in relation to their romantic partner’s (‘‘Partner B’’) characteristics. We collected self-report data from both partners. Due to time constraints, only one member of the dyad, recruited via a research participant pool (Partner B), completed the full battery of questionnaires; their scores are used as outcomes. Their respective partners, who were recruited via email (Partner A), completed a shorter battery; their scores are used as predictors. 2.2. Participants The sample comprised 104 undergraduate couples (50% male; 81.8% Caucasian; mean age = 19.6; SD = 1.63) consisting of participants recruited from a research pool (Partner B) and their respective romantic partners (Partner A). To participate, the dyad had to have been in their current relationship for at least 6 weeks. Of the 173 individuals who took part in the initial assessment, information on both partners was available for 104 couples. Participants who had a partner respond did not differ from participants who did not with respect to mean age, gender, ethnicity (i.e., % Caucasian vs. % non-Caucasian), dyadic adjustment, or narcissism. 2.3. Procedure Partner B filled out a series of measures pertaining to themselves, their partners, and their relationships through a secure website for research credit. Each Partner B was also asked to provide the email address of his or her romantic partner and three peers. Each Partner A was emailed a link to a secure online survey where they were asked questions about their relationship and personality. Partner A received ten dollars for participating in the study. Peers were emailed a link to a brief online survey where they were asked to answer questions about the couple and were entered into a $50 drawing. 2.4. Materials 2.4.1. Demographic and relationship information Both partners were asked to provide demographic information as well as duration of their current relationship. As there was wide variability in the length of relationships, duration was calculated as the number of weeks in the relationship. This information was reported by the participant via the question, ‘‘What is the current duration of your relationship?’’ Also, all partners were asked to rate their partners’ social status, influence, physical attractiveness, and popularity on a 1 (not at all) to 5 (very) scale. 2.4.2. Dyadic Adjustment Scale (DAS) The DAS (Spanier, 1976) consists of 32 items that assess consensus, affectional expression, satisfaction, and cohesion. Analyses use
168
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
the total score, which reflects overall relationship adjustment (Partner A: a = .92; Partner B: a = .92). 2.4.3. Hypersensitive Narcissism Scale (HSNS) Both partners completed the HSNS (Hendin & Cheek, 1997), a 10-item self-report measure of vulnerable narcissism. The items are rated on a 1 (very uncharacteristic) to 5 (very characteristic) scale (Partner A: a = .63; Partner B: a = .77). 2.4.4. Narcissistic Personality Inventory (NPI) Both partners completed the NPI (Raskin & Terry, 1988), a 40-item, forced-choice assessment of grandiose narcissism. Analyses use three subscales (Ackerman et al., 2011): Leadership/ Authority (LA – 11 items; Partner A: a = .82; Partner B: a = .77), Grandiose Exhibitionism (GE – 10 items; Partner A: a = .75; Partner B: a = .69), and Entitlement/Exploitativeness (EE – 4 items; Partner A: a = .43; Partner B: a = .53). These subscales were considered separately because they manifest divergent empirical profiles that align differentially with grandiose and vulnerable narcissism (Miller et al., 2014). 2.4.5. International Personality Item Pool (IPIP) The IPIP is a set of personality inventory questions in the public domain (Goldberg et al., 2006). Items are rated on a 1 (strongly disagree) to 5 (strongly agree) scale. Twenty questions were selected, with four questions representing each dimension of the five-factor personality model (FFM): neuroticism (Partner A: a = .55; Partner B: a = .68), extraversion (Partner A: a = .64; Partner B: a = .73), openness (Partner A: a = .75; Partner B: a = .79), agreeableness (Partner A: a = .49; Partner B: a = .66), and conscientiousness (Partner A: a = .73; Partner B: a = .69). 2.5. Partner B individual differences and developmental experiences 2.5.1. Child Abuse and Trauma Scale (CATS) The CATS (Sanders & Becker-Lausen, 1995) is a self-reported index of the frequency of negative events in childhood and adolescence. Items are rated from 0 (never) to 4 (always). The 38-item scale produces a total score (a = .94) along with three subscales that reflect sexual abuse, punishment, and negative home environment/neglect; in the current analyses we use the total score. 2.5.2. Patient-Reported Outcomes Measurement Information System, emotional distress (PROMIS) The PROMIS scales were developed for the National Institutes of Health (Pilkonis et al., 2011). In the present study, two scales were used: anxiety (PROMIS-A; a = .91) and depression (PROMIS-D; a = .95). Items were answered on a 1 (never) to 5 (always) scale. 2.5.3. Structured Clinical Interview for DSM-IV Axis II screener (SCID-II screener) The SCID-II screener (First, Gibbon, Spitzer, Williams, & Benjamin, 1997) is a self-report screening questionnaire for personality disorders with items answered in a yes/no format. A composite of all positively endorsed items was created to generate a total score (a = .95). 2.6. Peer informant ratings The email addresses of peers who knew both partners well were provided by Partner B. Ratings (143 peers responded) were received from one or more informant for 67 couples; so as to avoid differential reliability and validity that can result from using aggregated vs. non-aggregated ratings across participants, only a single informant report was used for all participants for whom a report was received (i.e., the first that was received). Each informant rated
each member of the couple on social status, influence, physical attractiveness, and popularity from 1 (very low) to 7 (very high). Informants used the same scale to rate the couple on overall attraction to each other, closeness, commitment, and satisfaction, which were used to form a composite of informant-rated relational adjustment (a = .82). An attrition analysis was performed in order to determine whether participants who had an informant respond differed from participants who did not. Groups did not differ with respect to age, ethnicity (i.e., % Caucasian vs. % non-Caucasian), dyadic adjustment, or narcissism. However, the sample with informants was composed of a higher proportion of females (77.6%) than the sample without informants (54.1% female). 3. Results 3.1. Assessing homophily by correlating narcissism scores Grandiose narcissism was measured using NPI subscales, whereas vulnerable narcissism was measured using the HSNS total score; the scores were uncorrelated. We found evidence of homophily for grandiose but not vulnerable narcissism (see Table 1). Partner A’s NPI LA narcissism was positively related to Partner B’s NPI LA (r = .32, p < .01) and Partner B’s NPI GE (r = .29, p < .01). Furthermore, Partner A’s NPI EE narcissism was positively related to Partner B’s NPI GE narcissism (r = .21, p < .05). All other Partner A to Partner B narcissism relations were statistically nonsignificant. 3.2. Additional partner characteristics 3.2.1. Self ratings Next, we examined Partner A’s self-rated narcissism in relation to Partner B’s self-reported characteristics (Table 1). Individuals who were dating partners with high NPI EE scores endorsed lower dyadic adjustment (r = .21, p < .05) and higher anxiety (r = .25, p < .05). Beyond these relations, Partner A’s grandiose and vulnerable narcissism scores were unrelated to Partner B’s scores on
Table 1 Relations between self-reported Partner A narcissism and Partner B characteristics. Partner B characteristics
Partner A NPI LA
Partner A NPI GE
Partner A NPI EE
Partner A HSNS
Narcissism scores NPI LA NPI GE NPI EE HSNS
.32** .29** .06 .02
.11 .16 .07 .15
.02 .21 .05 .08
.05 .01 .02 .06
Traits Neuroticism Extraversion Openness Agreeableness Conscientiousness
.06 .05 .02 .02 .10
.04 .14 .06 .05 .20
.05 .10 .11 .03 .06
.09 .06 .08 .05 .12
Dyadic adjustment
.01
.13
.21*
.06
Child abuse
.01
.01
.02
.13
Psychopathology Anxiety Depression DSM-IV PD symptoms
.11 .03 .13
.10 .08 .03
.25* .19 .08
.11 .07 .06
NPI LA = Leadership/Authority; NPI GE = Grandiose Exhibitionism; NPI EE = Entitlement/Exploitation; HSNS = Hypersensitive Narcissism Scale; PD = Personality Disorder. Ns range from 93 to 100. * p < .05. ** p < .01.
169
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
general personality traits, adverse developmental experiences, and psychopathology. 3.2.2. Peer ratings Partner A’s narcissism was examined in relation to peer ratings of the two partners’ individual characteristics (Table 2). Partner A’s narcissism scores were unrelated to peers’ ratings of Partner A and B with regard to social status, influence, and attractiveness. Partner A’s narcissism scores were also unrelated to peer ratings of Partner B’s popularity. However, Partner A’s grandiose narcissism scores were positively related to being rated as popular by the informant (NPI LA: r = .32, p < .05; NPI GE: r = .34, p < .01), whereas Partner A’s vulnerable narcissism scores were negatively related to being rated as popular (r = .27, p < .05). 3.3. Relationship adjustment and partner narcissism 3.3.1. Narcissism correlations with relationship adjustment For most of the narcissism scores (NPI LA, NPI GE, and HSNS), Partner A’s narcissism was unrelated to Partner B’s relationship adjustment (Table 1) as well as peer ratings of the couples’ relationship qualities (Table 2). However, Partner A’s NPI EE was
Table 2 Relations between Partner A narcissism and peer reports. Peer ratings Partner A social status Partner A influence Partner A attractiveness Partner A popularity Partner B social status Partner B influence Partner B attractiveness Partner B popularity Couples’ adjustment
Partner A NPI LA
Partner A NPI GE
Partner A NPI EE
Partner A HSNS
.16 .11 .08
.06 .05 .05
.06 .09 .07
.02 .09 .10
.32* .21 .09 .23
.34** .20 .10 .14
.03 .12 .08 .12
.27* .06 .10 .07
.02 .03
.14 .02
.01 .14
.04 .08
NPI LA = Leadership/Authority; NPI GE = Grandiose Exhibitionism; NPI EE = Entitlement/Exploitation; HSNS = Hypersensitive Narcissism Scale. Ns range from 62 to 66. Couples’ adjustment refers to the couples’ overall relationship quality. * p < .05. ** p < .01.
negatively related to Partner B’s self-reported relationship adjustment (r = .21, p < .05). Partner B’s total dyadic adjustment score was significantly positively correlated with the informant-reported relationship adjustment composite (r = .40, p < .01), suggesting some level of consensus regarding relationship quality.
3.3.2. Actor–Partner Interdependence Models The Actor–Partner Interdependence Model (APIM) is a statistical framework for analyzing dyadic data that acknowledges mutual influence, or the fact that one partner’s behaviors or attitudes can affect the other partner’s outcomes (Kenny, Kashy, & Cook, 2006). This model reflects both actor effects (e.g., an individual’s characteristics influence his own outcome) and partner effects (i.e., an individual’s outcome is influenced by his partner’s characteristics). In the present study, APIMs were constructed to determine whether having a partner with narcissism affected relationship ratings, and whether these effects were moderated by the duration of the relationship. Tests of distinguishability (Kenny, 2008) revealed that dyads were distinguishable by gender. This necessitated the exclusion of five same-sex couples so that gender could be used in the model to distinguish between dyad members. A model was constructed for each narcissism score (NPI LA, NPI GE, NPI EE, and HSNS), with dyadic adjustment (reported by both members of the couple) as the outcome variable. Models included main effects of gender, duration, actor narcissism and partner narcissism as well as cross-products between these variables (see Table 3). Significant interactions were graphed using Aiken and West’s (1991) procedure for simple slopes for visual interpretation. Levels were defined as low ( 1 standard deviation) and high (+1 standard deviation). The NPI LA model did not manifest any significant coefficients. In the NPI GE model, relationship duration predicted lower relationship adjustment (B = 3.30; p < .05; Table 3). When NPI EE was used to predict relationship adjustment, duration predicted lower adjustment (i.e., DAS Total; B = 2.88; p < .05). Two interactions were present in this model: partner EE and duration (B = 3.67; p < .01), and a three-way interaction between actor EE, partner EE, and duration (B = 4.02; p < .05). In the three-way interaction, duration was negatively related to relationship adjustment for the combination of high actor EE and high partner EE (B = 3.81; p < .01), while the other combinations of EE narcissism had slopes that did not differ from zero (Fig. 1). The HSNS model generated one significant coefficient. Relationship duration predicted lower relationship adjustment (B = 3.61; p < .01; Table 3).
Table 3 Actor–Partner Interdependence Models estimating Total Adjustment. NPI LA
Intercept Sex Duration Actor narcissism Partner narcissism Sex ⁄ actor narcissism Sex ⁄ partner narcissism Actor narcissism ⁄ partner narcissism Duration ⁄ actor narcissism Duration ⁄ partner narcissism Duration ⁄ actor narcissism ⁄ partner narcissism
NPI GE
NPI EE
HSNS
Estimate
SE
Estimate
SE
Estimate
SE
Estimate
SE
108.53 3.97 2.56 4.69 0.73 2.04 1.58 0.91 0.62 1.16 0.39
3.00 1.72 1.41 4.18 4.28 2.68 2.72 1.53 1.03 1.05 1.31
109.23 3.79 3.30* 2.39 4.23 1.16 2.09 0.24 1.34 1.40 0.67
2.82 1.63 1.43 3.70 3.72 2.32 2.34 1.23 1.31 1.33 1.66
112.49 2.58 2.88* 6.25 4.77 2.21 4.44 2.05 2.61 3.67** 4.02*
3.00 1.74 1.20 3.72 4.52 2.62 2.66 1.59 1.36 1.39 1.78
109.54 3.55 3.61** 0.78 6.29 3.65 2.33 1.28 1.63 0.19 1.18
2.64 1.58 1.19 3.26 3.27 2.08 2.08 1.24 1.06 1.09 1.02
Total Adjustment = Dyadic Adjustment Scale Total Adjustment Score; NPI LA = Narcissistic Personality Inventory Leadership/Authority Scale; NPI GE = Narcissistic Personality Inventory Grandiose Exhibitionism Scale; NPI EE = Narcissistic Personality Inventory Entitlement/Exploitativeness Scale; HSNS = Hypersensitive Narcissism Scale; SE = Standard Error. N = 104 dyads. * p < .05. ** p < .01.
170
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
125
Self-Rated Adjustment
120
High Actor EE + High Partner EE
115
High Actor EE + Low Partner EE
110
Low Actor EE + High Partner EE 105 Low Actor EE + Low Partner EE 100
95 Low Duration
High Duration
Fig. 1. Three-way interaction using duration, actor EE narcissism, and partner EE narcissism to predict relationship adjustment. Note: Low and high levels of variables are represented by 1 SD and +1 SD, respectively. Self-Rated Adjustment = Dyadic Adjustment Scale Total Score. EE = Entitlement/Exploitativeness Subscale of the Narcissistic Personality Inventory.
4. Discussion The current study examined who grandiosely and vulnerably narcissistic individuals date and whether the presence of these traits in romantic partners is related to relationship quality that differs depending on relationship duration. Overall, there was some support for the presence of homophily (i.e., similarity) across romantic partners, but only in relation to aspects of grandiose narcissism. Self-reports of grandiose narcissism scores from both romantic partners were positively related for NPI LA (and to some extent, NPI GE). No similar pattern was found for vulnerable narcissism. More broadly, however, there was limited evidence that either grandiosely or vulnerably narcissistic individuals have a specific type of romantic partner – for example, they do not tend to date partners with particular traits or characteristics that are indicative of a preference for ‘‘submissive’’ romantic partners. In fact, there were remarkably few correlations between Partner A’s narcissism scores, grandiose and vulnerable, and Partner B’s development experiences (i.e., abuse), general personality traits, psychopathology, or relationship satisfaction (both self and peer-rated). Although friends of the romantic couple reported few consistencies in who dates narcissistic individuals, these peer raters did confirm some of the positive and negative characteristics reported by Partner As (i.e., grandiosely narcissistic individuals were perceived by peers of the couple as more popular; vulnerably narcissistic individuals were perceived as less popular). Throughout these findings, there appears to be a difference between NPI subscales. Participants with high scores on the scales that are the strongest and purest markers of grandiose narcissism – Leadership/Authority and Grandiose Exhibitionism (Miller et al., 2014) were perceived more positively by peers. This pattern did not hold true for the NPI scale measuring entitlement and exploitativeness, which is associated with both grandiose and vulnerable narcissism, as it was associated with having partners who reported lower relationship adjustment and greater anxiety. Portions of the chocolate cake model of narcissism (Campbell, 2005) in the dating context were confirmed as well, such that certain narcissism traits such as entitlement and exploitativeness were negatively related to dyadic adjustment in relationships of a longer duration when both partners exhibited these traits. There was weaker evidence, however, that grandiose narcissism was related to positive outcomes in shorter duration relationships
(the relations were positive but not statistically significant). In general, there was no substantial relation between vulnerable narcissism and dyadic adjustment either as a main effect or moderated by duration. 4.1. Limitations and future considerations An important consideration regarding the present study is the representativeness of the sample, which consisted of college students reporting on their romantic relationships halfway through the school year. Replications in older, more diverse couples with more well-established relationships are needed. Additionally, many questions about narcissism in the romantic context remain to be answered. The cross-sectional design used here is limited in that we could only study intact romantic relationships and thus could not examine the relations between narcissism and relationship dissolution or how narcissism affects romantic relationships over time. Ultimately, longitudinal studies will be necessary to gather a more fine-grained examination of the manner in which narcissism affects romantic relationships. 4.2. Conclusion Information about narcissism in the context of relationships has the potential to add to the growing body of literature aimed at distinguishing between grandiose and vulnerable narcissism. While it is tempting to believe narcissistic individuals have a ‘‘type’’ of romantic partner who they pursue, particularly those who are equally self-centered and grandiose or those who may be easily manipulated, we found relatively little consistent evidence of this. More specifically, no evidence was found to support the ‘‘victim preference’’ hypothesis and only modest support for the homophily hypothesis was found. Relationship adjustment did manifest a temporal pattern: as expected, the negative effects of grandiose narcissism appear to emerge in relationships of longer duration, which mirrors the shift in likability and popularity found when peers rate grandiose individuals over time (Paulhus, 1998). 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.
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171 Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage Publications. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author. Brunell, A. B., & Campbell, W. K. (2011). Narcissism and romantic relationships. In W. K. Campbell & J. D. Miller (Eds.), The handbook of narcissism and narcissistic personality disorder: Theoretical approaches, empirical findings, and treatments (pp. 344–350). Hoboken, NJ: John Wiley & Sons Inc. Campbell, W. (1999). Narcissism and romantic attraction. Journal of Personality and Social Psychology, 77, 1254–1270. Campbell, W. K. (2005). When you love a man who loves himself: How to deal with a one-way relationship. Chicago, IL: Sourcebooks Casablanca. Campbell, W., & Campbell, S. M. (2009). On the self-regulatory dynamics created by the particular benefits and costs of narcissism: A contextual reinforcement model and examination of leadership. Self and Identity, 8, 214–232. Campbell, W., Foster, C. A., & Finkel, E. J. (2002). Does self-love lead to love for others? A story of narcissistic game playing. Journal of Personality and Social Psychology, 83, 340–354. First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., & Benjamin, L. S. (1997). Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Washington, D.C.: American Psychiatric Press Inc. 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 publicdomain personality measures. Journal of Research in Personality, 40, 84–96. Hendin, H. M., & Cheek, J. M. (1997). Assessing hypersensitive narcissism: A reexamination of Murray’s Narcissism Scale. Journal of Research in Personality, 31, 588–599. Keller, P. S., Blincoe, S., Gilbert, L. R., Dewall, C. N., Haak, E. A., & Widiger, T. (2014). Narcissism in romantic relationships: A dyadic perspective. Journal of Social and Clinical Psychology, 33, 25–50. Kenny, D. A. (2008). Dyadic analysis. (internet web site). Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. New York, NY: The Guilford Press. Lamkin, J., Clifton, A., Campbell, W. K., & Miller, J. (2014). An examination of the social network characteristics associated with grandiose and vulnerable narcissism. Personality Disorders: Theory, Research, and Treatment, 5, 137–145. McNulty, J. K., & Widman, L. (2013). The implications of sexual narcissism for sexual and marital satisfaction. Archives of Sexual Behavior, 42, 1021–1032.
171
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 415–444. Miller, J. D., & Campbell, W. K. (2008). Comparing clinical and social-personality conceptualizations of narcissism. Journal of Personality, 76, 449–476. Miller, J. D., Campbell, W., & Pilkonis, P. A. (2007). Narcissistic personality disorder: Relations with distress and functional impairment. Comprehensive Psychiatry, 48, 170–177. Miller, J. D., Hoffman, B. J., Gaughan, E. T., Gentile, B., Maples, J., & Campbell, W. K. (2011). Grandiose and vulnerable narcissism: A nomological network analysis. Journal of Personality, 79, 1013–1042. Miller, J. D., McCain, J., Lynam, D. R., Few, L. R., Gentile, B., MacKillop, J., et al. (2014). A comparison of the criterion validity of popular measures of narcissism and narcissistic personality disorder via the use of expert ratings. Psychological Assessment, 26, 958–969. Paulhus, D. L. (1998). Interpersonal and intrapsychic adaptiveness of trait selfenhancement: A mixed blessing? Journal of Personality and Social Psychology, 74, 1197–1208. Pilkonis, P. A., Choi, S. W., Reise, S. P., Stover, A. M., Riley, W. T., & Cella, D. (2011). Item banks for measuring emotional distress from the Patient-Reported Outcomes Measurement Information System (PROMIS): Depression, anxiety, and anger. Assessment, 18, 263–283. Pincus, A. L., & Lukowitsky, M. R. (2010). Pathological narcissism and narcissistic personality disorder. Annual Review of Clinical Psychology, 6, 421–446. Raskin, R., & 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. Ryan, K. M., Weikel, K., & Sprechini, G. (2008). Gender differences in narcissism and courtship violence in dating couples. Sex Roles, 58, 802–813. Sanders, B., & Becker-Lausen, E. (1995). The measurement of psychological maltreatment: Early data on the child abuse and trauma scale. Child Abuse and Neglect, 19, 315–323. Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15–28. Tanchotsrinon, P., Maneesri, K., & Campbell, W. (2007). Narcissism and romantic attraction: Evidence from a collectivistic culture. Journal of Research in Personality, 41, 723–730. Wink, P. (1991). Two faces of narcissism. Journal of Personality and Social Psychology, 61, 590–597.
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
An exploration of the correlates of grandiose and vulnerable narcissism in romantic relationships: Homophily, partner characteristics, and dyadic adjustment Joanna Lamkin ⇑, W. Keith Campbell, Michelle R. vanDellen, Joshua D. Miller University of Georgia, Department of Psychology, Psychology Building, Athens, GA 30602-3013, United States
a r t i c l e
i n f o
Article history: Received 5 December 2014 Received in revised form 20 January 2015 Accepted 21 January 2015
Keywords: Narcissism Romantic relationships Homophily Grandiose Vulnerable
a b s t r a c t The present study examined the characteristics of individuals (N = 104 undergraduate couples) who date grandiosely or vulnerably narcissistic individuals, including the experience of developmental trauma, general and pathological personality traits, and psychopathology, using multiple data sources. In addition, relationship duration was tested as a moderator of the relations between the narcissism dimensions and relationship adjustment. Actor–Partner Interdependence Models indicated that negative relationship adjustment was found when both partners had higher entitlement/exploitativeness traits and had been together for a longer period of time. Overall, there were no clear patterns of partner characteristics, although some evidence for homophily emerged for traits related to grandiose narcissism. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Narcissistic individuals possess a sense of self-importance, believe themselves to be superior or special, are preoccupied with fantasies of success and admiration, and expect preferential treatment (American Psychiatric Association, 2013). Narcissism is also linked to a lack of empathy, sense of entitlement, and envy. More recent conceptualizations of narcissism emphasize the need to delineate between grandiose and vulnerable dimensions of narcissism (e.g., Miller & Campbell, 2008; Pincus & Lukowitsky, 2010), as these dimensions manifest divergent nomological networks (Miller et al., 2011). Grandiose narcissism is linked to extraversion, dominance, self-assurance, exhibitionism, and aggression; vulnerable narcissism is distinguished by introversion, defensiveness, anxiety, interpersonal coldness and hostility, as well as vulnerability to stress (Wink, 1991). Both narcissism dimensions, however, are associated with interpersonal dysfunction due, in part, to the antagonistic and intrusive behaviors associated with both. In the present research, we examine both dimensions in the context of romantic relationships. It is unclear whether narcissistic individuals seek out a certain type of romantic partner. Given the interpersonal antagonism associated with narcissism, one hypothesis is that narcissistic ⇑ Corresponding author. Tel.: +1 (706) 542 1173. E-mail address: [email protected] (J. Lamkin). http://dx.doi.org/10.1016/j.paid.2015.01.029 0191-8869/Ó 2015 Elsevier Ltd. All rights reserved.
individuals might seek out meek, passive individuals who make for more pliant partners (i.e., ‘‘victims’’). For example, perhaps partners with mood- or personality-related psychopathology and family histories of conflict or maltreatment might be seen by narcissistic individuals as easier to manipulate and thus appealing romantic partners. However, evidence from studies on homophily support another hypothesis, which suggests that grandiosely narcissistic individuals will seek partners who are similarly self-centered. Homophily, the idea that individuals partner with similar others, is observed throughout a wide range of human relationships (see McPherson, Smith-Lovin, & Cook, 2001). Several studies have tested for homophily in relation to grandiose narcissism with some studies finding evidence for it (Campbell, Foster, & Finkel, 2002; Keller et al., 2014; McNulty & Widman, 2013) and others not (e.g., Ryan, Weikel, & Sprechini, 2008). To date, there has been no published work examining relational homophily as it pertains to vulnerable narcissism. Using egocentric social network analysis, Lamkin, Clifton, Campbell, and Miller (2014) found that both vulnerable and grandiose narcissistic individuals described, on average, the individuals closest to them in their social networks as self-centered, narcissistic, and disagreeable—suggesting, at the very least, that there is some perception of homophily among both grandiosely and vulnerably narcissistic individuals. Campbell’s (1999) self-orientation model of narcissism and romantic attraction posits that individuals with narcissistic tendencies employ a self-regulatory strategy in which interpersonal
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
relationships serve to enhance and maintain their self-concept. Initial research provides some support for this model; for instance, even in a collectivistic culture that emphasizes communal values over self-focused goals (i.e., Thailand), narcissism predicted attraction to individuals who provided an opportunity for self-enhancement (Tanchotsrinon, Maneesri, & Campbell, 2007). Although all participants were attracted to caring individuals, grandiose participants were more attracted to partners who were described as admiring, attractive, and high status, and less attracted to otheroriented individuals who were described as caring and emotionally needy. At this time, little is known about the manner in which vulnerable narcissism is related to romantic functioning. However, given that vulnerable narcissism shares a basic tendency towards self-absorption and entitlement with grandiose narcissism, one might hypothesize that vulnerably narcissistic individuals would also prefer romantic partners who serve as sources of self-enhancement. Actually entering relationships that provide these opportunities may be more difficult for these individuals to achieve, however, due to their more neurotic and introverted personality style (Wink, 1991). Furthermore, vulnerable narcissism has been linked to anxiety, depression, and maladaptive attachment styles, which may also impact romantic relationships (Miller et al., 2011). Narcissism often has negative interpersonal consequences, including causing pain and suffering in romantic partners, family members, and friends (Miller, Campbell, & Pilkonis, 2007). Partners of narcissistic individuals incur long-term costs as a result of the decisions and behaviors enacted by their partners (Campbell & Campbell, 2009). Relationships with narcissistic individuals may start out positively, but then decline over time. For example, when asked to rate their relationship satisfaction at the beginning and end of their relationship, those who were dating narcissistic individuals endorsed a larger reduction in satisfaction than those who were not dating narcissistic individuals (Brunell & Campbell, 2011). This change in relationship satisfaction over time has been called the chocolate cake model (Campbell, 2005). In this model, chocolate cake serves as a metaphor for a relationship with someone who is narcissistic: initially, eating chocolate cake is enjoyable, but it comes with long-term costs (e.g., weight gain, discomfort) that lead to dissatisfaction. Of note, studies related to the chocolate cake model have typically used the Narcissistic Personality Inventory (NPI), which tends to measure the primarily grandiose features of narcissism and NPD (e.g., Miller et al., 2014) and thus the generalizability of these findings to vulnerable narcissism is unclear. Based on what is known about narcissism and romantic relationships, we developed two hypotheses and one research question. First, we predicted that partners of narcissistic individuals would also exhibit narcissistic characteristics. We further anticipated that grandiose and vulnerable narcissism would manifest homophily differentially through their respective personality traits (i.e., grandiosely narcissistic individuals would be more likely to date others high in extraversion; vulnerably narcissistic individuals would be more likely to date others characterized by introversion and neuroticism). Second, we explored the characteristics of those dating narcissistic individuals. Beyond homophily, there was no clear prediction to make as to whether grandiosely or vulnerably narcissistic individuals would have clear preferences for certain types of partners. We also included variables related to the ‘‘victim’’ hypothesis to test whether narcissistic individuals might seek out more vulnerable partners (e.g., histories of maltreatment; negative emotionality; personality difficulties). Third, consistent with the chocolate cake model, we hypothesized that partners of those with higher levels of grandiose narcissism would rate a relationship of shorter duration more positively than those
167
in a relationship of a longer duration. We hypothesized that this pattern would be weaker or nonexistent for individuals dating a vulnerably narcissistic individual, as these relationships may be less likely to begin with significant satisfaction. Additionally, we obtained peer reports about the relationships in order to have an alternative perspective on variables related to self-enhancement (e.g., social influence). 2. Method 2.1. Overview We used multiple measures to answers the questions of interest. Our focus was on the narcissism scores of ‘‘Partner A’’ in relation to their romantic partner’s (‘‘Partner B’’) characteristics. We collected self-report data from both partners. Due to time constraints, only one member of the dyad, recruited via a research participant pool (Partner B), completed the full battery of questionnaires; their scores are used as outcomes. Their respective partners, who were recruited via email (Partner A), completed a shorter battery; their scores are used as predictors. 2.2. Participants The sample comprised 104 undergraduate couples (50% male; 81.8% Caucasian; mean age = 19.6; SD = 1.63) consisting of participants recruited from a research pool (Partner B) and their respective romantic partners (Partner A). To participate, the dyad had to have been in their current relationship for at least 6 weeks. Of the 173 individuals who took part in the initial assessment, information on both partners was available for 104 couples. Participants who had a partner respond did not differ from participants who did not with respect to mean age, gender, ethnicity (i.e., % Caucasian vs. % non-Caucasian), dyadic adjustment, or narcissism. 2.3. Procedure Partner B filled out a series of measures pertaining to themselves, their partners, and their relationships through a secure website for research credit. Each Partner B was also asked to provide the email address of his or her romantic partner and three peers. Each Partner A was emailed a link to a secure online survey where they were asked questions about their relationship and personality. Partner A received ten dollars for participating in the study. Peers were emailed a link to a brief online survey where they were asked to answer questions about the couple and were entered into a $50 drawing. 2.4. Materials 2.4.1. Demographic and relationship information Both partners were asked to provide demographic information as well as duration of their current relationship. As there was wide variability in the length of relationships, duration was calculated as the number of weeks in the relationship. This information was reported by the participant via the question, ‘‘What is the current duration of your relationship?’’ Also, all partners were asked to rate their partners’ social status, influence, physical attractiveness, and popularity on a 1 (not at all) to 5 (very) scale. 2.4.2. Dyadic Adjustment Scale (DAS) The DAS (Spanier, 1976) consists of 32 items that assess consensus, affectional expression, satisfaction, and cohesion. Analyses use
168
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
the total score, which reflects overall relationship adjustment (Partner A: a = .92; Partner B: a = .92). 2.4.3. Hypersensitive Narcissism Scale (HSNS) Both partners completed the HSNS (Hendin & Cheek, 1997), a 10-item self-report measure of vulnerable narcissism. The items are rated on a 1 (very uncharacteristic) to 5 (very characteristic) scale (Partner A: a = .63; Partner B: a = .77). 2.4.4. Narcissistic Personality Inventory (NPI) Both partners completed the NPI (Raskin & Terry, 1988), a 40-item, forced-choice assessment of grandiose narcissism. Analyses use three subscales (Ackerman et al., 2011): Leadership/ Authority (LA – 11 items; Partner A: a = .82; Partner B: a = .77), Grandiose Exhibitionism (GE – 10 items; Partner A: a = .75; Partner B: a = .69), and Entitlement/Exploitativeness (EE – 4 items; Partner A: a = .43; Partner B: a = .53). These subscales were considered separately because they manifest divergent empirical profiles that align differentially with grandiose and vulnerable narcissism (Miller et al., 2014). 2.4.5. International Personality Item Pool (IPIP) The IPIP is a set of personality inventory questions in the public domain (Goldberg et al., 2006). Items are rated on a 1 (strongly disagree) to 5 (strongly agree) scale. Twenty questions were selected, with four questions representing each dimension of the five-factor personality model (FFM): neuroticism (Partner A: a = .55; Partner B: a = .68), extraversion (Partner A: a = .64; Partner B: a = .73), openness (Partner A: a = .75; Partner B: a = .79), agreeableness (Partner A: a = .49; Partner B: a = .66), and conscientiousness (Partner A: a = .73; Partner B: a = .69). 2.5. Partner B individual differences and developmental experiences 2.5.1. Child Abuse and Trauma Scale (CATS) The CATS (Sanders & Becker-Lausen, 1995) is a self-reported index of the frequency of negative events in childhood and adolescence. Items are rated from 0 (never) to 4 (always). The 38-item scale produces a total score (a = .94) along with three subscales that reflect sexual abuse, punishment, and negative home environment/neglect; in the current analyses we use the total score. 2.5.2. Patient-Reported Outcomes Measurement Information System, emotional distress (PROMIS) The PROMIS scales were developed for the National Institutes of Health (Pilkonis et al., 2011). In the present study, two scales were used: anxiety (PROMIS-A; a = .91) and depression (PROMIS-D; a = .95). Items were answered on a 1 (never) to 5 (always) scale. 2.5.3. Structured Clinical Interview for DSM-IV Axis II screener (SCID-II screener) The SCID-II screener (First, Gibbon, Spitzer, Williams, & Benjamin, 1997) is a self-report screening questionnaire for personality disorders with items answered in a yes/no format. A composite of all positively endorsed items was created to generate a total score (a = .95). 2.6. Peer informant ratings The email addresses of peers who knew both partners well were provided by Partner B. Ratings (143 peers responded) were received from one or more informant for 67 couples; so as to avoid differential reliability and validity that can result from using aggregated vs. non-aggregated ratings across participants, only a single informant report was used for all participants for whom a report was received (i.e., the first that was received). Each informant rated
each member of the couple on social status, influence, physical attractiveness, and popularity from 1 (very low) to 7 (very high). Informants used the same scale to rate the couple on overall attraction to each other, closeness, commitment, and satisfaction, which were used to form a composite of informant-rated relational adjustment (a = .82). An attrition analysis was performed in order to determine whether participants who had an informant respond differed from participants who did not. Groups did not differ with respect to age, ethnicity (i.e., % Caucasian vs. % non-Caucasian), dyadic adjustment, or narcissism. However, the sample with informants was composed of a higher proportion of females (77.6%) than the sample without informants (54.1% female). 3. Results 3.1. Assessing homophily by correlating narcissism scores Grandiose narcissism was measured using NPI subscales, whereas vulnerable narcissism was measured using the HSNS total score; the scores were uncorrelated. We found evidence of homophily for grandiose but not vulnerable narcissism (see Table 1). Partner A’s NPI LA narcissism was positively related to Partner B’s NPI LA (r = .32, p < .01) and Partner B’s NPI GE (r = .29, p < .01). Furthermore, Partner A’s NPI EE narcissism was positively related to Partner B’s NPI GE narcissism (r = .21, p < .05). All other Partner A to Partner B narcissism relations were statistically nonsignificant. 3.2. Additional partner characteristics 3.2.1. Self ratings Next, we examined Partner A’s self-rated narcissism in relation to Partner B’s self-reported characteristics (Table 1). Individuals who were dating partners with high NPI EE scores endorsed lower dyadic adjustment (r = .21, p < .05) and higher anxiety (r = .25, p < .05). Beyond these relations, Partner A’s grandiose and vulnerable narcissism scores were unrelated to Partner B’s scores on
Table 1 Relations between self-reported Partner A narcissism and Partner B characteristics. Partner B characteristics
Partner A NPI LA
Partner A NPI GE
Partner A NPI EE
Partner A HSNS
Narcissism scores NPI LA NPI GE NPI EE HSNS
.32** .29** .06 .02
.11 .16 .07 .15
.02 .21 .05 .08
.05 .01 .02 .06
Traits Neuroticism Extraversion Openness Agreeableness Conscientiousness
.06 .05 .02 .02 .10
.04 .14 .06 .05 .20
.05 .10 .11 .03 .06
.09 .06 .08 .05 .12
Dyadic adjustment
.01
.13
.21*
.06
Child abuse
.01
.01
.02
.13
Psychopathology Anxiety Depression DSM-IV PD symptoms
.11 .03 .13
.10 .08 .03
.25* .19 .08
.11 .07 .06
NPI LA = Leadership/Authority; NPI GE = Grandiose Exhibitionism; NPI EE = Entitlement/Exploitation; HSNS = Hypersensitive Narcissism Scale; PD = Personality Disorder. Ns range from 93 to 100. * p < .05. ** p < .01.
169
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
general personality traits, adverse developmental experiences, and psychopathology. 3.2.2. Peer ratings Partner A’s narcissism was examined in relation to peer ratings of the two partners’ individual characteristics (Table 2). Partner A’s narcissism scores were unrelated to peers’ ratings of Partner A and B with regard to social status, influence, and attractiveness. Partner A’s narcissism scores were also unrelated to peer ratings of Partner B’s popularity. However, Partner A’s grandiose narcissism scores were positively related to being rated as popular by the informant (NPI LA: r = .32, p < .05; NPI GE: r = .34, p < .01), whereas Partner A’s vulnerable narcissism scores were negatively related to being rated as popular (r = .27, p < .05). 3.3. Relationship adjustment and partner narcissism 3.3.1. Narcissism correlations with relationship adjustment For most of the narcissism scores (NPI LA, NPI GE, and HSNS), Partner A’s narcissism was unrelated to Partner B’s relationship adjustment (Table 1) as well as peer ratings of the couples’ relationship qualities (Table 2). However, Partner A’s NPI EE was
Table 2 Relations between Partner A narcissism and peer reports. Peer ratings Partner A social status Partner A influence Partner A attractiveness Partner A popularity Partner B social status Partner B influence Partner B attractiveness Partner B popularity Couples’ adjustment
Partner A NPI LA
Partner A NPI GE
Partner A NPI EE
Partner A HSNS
.16 .11 .08
.06 .05 .05
.06 .09 .07
.02 .09 .10
.32* .21 .09 .23
.34** .20 .10 .14
.03 .12 .08 .12
.27* .06 .10 .07
.02 .03
.14 .02
.01 .14
.04 .08
NPI LA = Leadership/Authority; NPI GE = Grandiose Exhibitionism; NPI EE = Entitlement/Exploitation; HSNS = Hypersensitive Narcissism Scale. Ns range from 62 to 66. Couples’ adjustment refers to the couples’ overall relationship quality. * p < .05. ** p < .01.
negatively related to Partner B’s self-reported relationship adjustment (r = .21, p < .05). Partner B’s total dyadic adjustment score was significantly positively correlated with the informant-reported relationship adjustment composite (r = .40, p < .01), suggesting some level of consensus regarding relationship quality.
3.3.2. Actor–Partner Interdependence Models The Actor–Partner Interdependence Model (APIM) is a statistical framework for analyzing dyadic data that acknowledges mutual influence, or the fact that one partner’s behaviors or attitudes can affect the other partner’s outcomes (Kenny, Kashy, & Cook, 2006). This model reflects both actor effects (e.g., an individual’s characteristics influence his own outcome) and partner effects (i.e., an individual’s outcome is influenced by his partner’s characteristics). In the present study, APIMs were constructed to determine whether having a partner with narcissism affected relationship ratings, and whether these effects were moderated by the duration of the relationship. Tests of distinguishability (Kenny, 2008) revealed that dyads were distinguishable by gender. This necessitated the exclusion of five same-sex couples so that gender could be used in the model to distinguish between dyad members. A model was constructed for each narcissism score (NPI LA, NPI GE, NPI EE, and HSNS), with dyadic adjustment (reported by both members of the couple) as the outcome variable. Models included main effects of gender, duration, actor narcissism and partner narcissism as well as cross-products between these variables (see Table 3). Significant interactions were graphed using Aiken and West’s (1991) procedure for simple slopes for visual interpretation. Levels were defined as low ( 1 standard deviation) and high (+1 standard deviation). The NPI LA model did not manifest any significant coefficients. In the NPI GE model, relationship duration predicted lower relationship adjustment (B = 3.30; p < .05; Table 3). When NPI EE was used to predict relationship adjustment, duration predicted lower adjustment (i.e., DAS Total; B = 2.88; p < .05). Two interactions were present in this model: partner EE and duration (B = 3.67; p < .01), and a three-way interaction between actor EE, partner EE, and duration (B = 4.02; p < .05). In the three-way interaction, duration was negatively related to relationship adjustment for the combination of high actor EE and high partner EE (B = 3.81; p < .01), while the other combinations of EE narcissism had slopes that did not differ from zero (Fig. 1). The HSNS model generated one significant coefficient. Relationship duration predicted lower relationship adjustment (B = 3.61; p < .01; Table 3).
Table 3 Actor–Partner Interdependence Models estimating Total Adjustment. NPI LA
Intercept Sex Duration Actor narcissism Partner narcissism Sex ⁄ actor narcissism Sex ⁄ partner narcissism Actor narcissism ⁄ partner narcissism Duration ⁄ actor narcissism Duration ⁄ partner narcissism Duration ⁄ actor narcissism ⁄ partner narcissism
NPI GE
NPI EE
HSNS
Estimate
SE
Estimate
SE
Estimate
SE
Estimate
SE
108.53 3.97 2.56 4.69 0.73 2.04 1.58 0.91 0.62 1.16 0.39
3.00 1.72 1.41 4.18 4.28 2.68 2.72 1.53 1.03 1.05 1.31
109.23 3.79 3.30* 2.39 4.23 1.16 2.09 0.24 1.34 1.40 0.67
2.82 1.63 1.43 3.70 3.72 2.32 2.34 1.23 1.31 1.33 1.66
112.49 2.58 2.88* 6.25 4.77 2.21 4.44 2.05 2.61 3.67** 4.02*
3.00 1.74 1.20 3.72 4.52 2.62 2.66 1.59 1.36 1.39 1.78
109.54 3.55 3.61** 0.78 6.29 3.65 2.33 1.28 1.63 0.19 1.18
2.64 1.58 1.19 3.26 3.27 2.08 2.08 1.24 1.06 1.09 1.02
Total Adjustment = Dyadic Adjustment Scale Total Adjustment Score; NPI LA = Narcissistic Personality Inventory Leadership/Authority Scale; NPI GE = Narcissistic Personality Inventory Grandiose Exhibitionism Scale; NPI EE = Narcissistic Personality Inventory Entitlement/Exploitativeness Scale; HSNS = Hypersensitive Narcissism Scale; SE = Standard Error. N = 104 dyads. * p < .05. ** p < .01.
170
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171
125
Self-Rated Adjustment
120
High Actor EE + High Partner EE
115
High Actor EE + Low Partner EE
110
Low Actor EE + High Partner EE 105 Low Actor EE + Low Partner EE 100
95 Low Duration
High Duration
Fig. 1. Three-way interaction using duration, actor EE narcissism, and partner EE narcissism to predict relationship adjustment. Note: Low and high levels of variables are represented by 1 SD and +1 SD, respectively. Self-Rated Adjustment = Dyadic Adjustment Scale Total Score. EE = Entitlement/Exploitativeness Subscale of the Narcissistic Personality Inventory.
4. Discussion The current study examined who grandiosely and vulnerably narcissistic individuals date and whether the presence of these traits in romantic partners is related to relationship quality that differs depending on relationship duration. Overall, there was some support for the presence of homophily (i.e., similarity) across romantic partners, but only in relation to aspects of grandiose narcissism. Self-reports of grandiose narcissism scores from both romantic partners were positively related for NPI LA (and to some extent, NPI GE). No similar pattern was found for vulnerable narcissism. More broadly, however, there was limited evidence that either grandiosely or vulnerably narcissistic individuals have a specific type of romantic partner – for example, they do not tend to date partners with particular traits or characteristics that are indicative of a preference for ‘‘submissive’’ romantic partners. In fact, there were remarkably few correlations between Partner A’s narcissism scores, grandiose and vulnerable, and Partner B’s development experiences (i.e., abuse), general personality traits, psychopathology, or relationship satisfaction (both self and peer-rated). Although friends of the romantic couple reported few consistencies in who dates narcissistic individuals, these peer raters did confirm some of the positive and negative characteristics reported by Partner As (i.e., grandiosely narcissistic individuals were perceived by peers of the couple as more popular; vulnerably narcissistic individuals were perceived as less popular). Throughout these findings, there appears to be a difference between NPI subscales. Participants with high scores on the scales that are the strongest and purest markers of grandiose narcissism – Leadership/Authority and Grandiose Exhibitionism (Miller et al., 2014) were perceived more positively by peers. This pattern did not hold true for the NPI scale measuring entitlement and exploitativeness, which is associated with both grandiose and vulnerable narcissism, as it was associated with having partners who reported lower relationship adjustment and greater anxiety. Portions of the chocolate cake model of narcissism (Campbell, 2005) in the dating context were confirmed as well, such that certain narcissism traits such as entitlement and exploitativeness were negatively related to dyadic adjustment in relationships of a longer duration when both partners exhibited these traits. There was weaker evidence, however, that grandiose narcissism was related to positive outcomes in shorter duration relationships
(the relations were positive but not statistically significant). In general, there was no substantial relation between vulnerable narcissism and dyadic adjustment either as a main effect or moderated by duration. 4.1. Limitations and future considerations An important consideration regarding the present study is the representativeness of the sample, which consisted of college students reporting on their romantic relationships halfway through the school year. Replications in older, more diverse couples with more well-established relationships are needed. Additionally, many questions about narcissism in the romantic context remain to be answered. The cross-sectional design used here is limited in that we could only study intact romantic relationships and thus could not examine the relations between narcissism and relationship dissolution or how narcissism affects romantic relationships over time. Ultimately, longitudinal studies will be necessary to gather a more fine-grained examination of the manner in which narcissism affects romantic relationships. 4.2. Conclusion Information about narcissism in the context of relationships has the potential to add to the growing body of literature aimed at distinguishing between grandiose and vulnerable narcissism. While it is tempting to believe narcissistic individuals have a ‘‘type’’ of romantic partner who they pursue, particularly those who are equally self-centered and grandiose or those who may be easily manipulated, we found relatively little consistent evidence of this. More specifically, no evidence was found to support the ‘‘victim preference’’ hypothesis and only modest support for the homophily hypothesis was found. Relationship adjustment did manifest a temporal pattern: as expected, the negative effects of grandiose narcissism appear to emerge in relationships of longer duration, which mirrors the shift in likability and popularity found when peers rate grandiose individuals over time (Paulhus, 1998). 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.
J. Lamkin et al. / Personality and Individual Differences 79 (2015) 166–171 Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage Publications. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author. Brunell, A. B., & Campbell, W. K. (2011). Narcissism and romantic relationships. In W. K. Campbell & J. D. Miller (Eds.), The handbook of narcissism and narcissistic personality disorder: Theoretical approaches, empirical findings, and treatments (pp. 344–350). Hoboken, NJ: John Wiley & Sons Inc. Campbell, W. (1999). Narcissism and romantic attraction. Journal of Personality and Social Psychology, 77, 1254–1270. Campbell, W. K. (2005). When you love a man who loves himself: How to deal with a one-way relationship. Chicago, IL: Sourcebooks Casablanca. Campbell, W., & Campbell, S. M. (2009). On the self-regulatory dynamics created by the particular benefits and costs of narcissism: A contextual reinforcement model and examination of leadership. Self and Identity, 8, 214–232. Campbell, W., Foster, C. A., & Finkel, E. J. (2002). Does self-love lead to love for others? A story of narcissistic game playing. Journal of Personality and Social Psychology, 83, 340–354. First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., & Benjamin, L. S. (1997). Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Washington, D.C.: American Psychiatric Press Inc. 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 publicdomain personality measures. Journal of Research in Personality, 40, 84–96. Hendin, H. M., & Cheek, J. M. (1997). Assessing hypersensitive narcissism: A reexamination of Murray’s Narcissism Scale. Journal of Research in Personality, 31, 588–599. Keller, P. S., Blincoe, S., Gilbert, L. R., Dewall, C. N., Haak, E. A., & Widiger, T. (2014). Narcissism in romantic relationships: A dyadic perspective. Journal of Social and Clinical Psychology, 33, 25–50. Kenny, D. A. (2008). Dyadic analysis.
171
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 415–444. Miller, J. D., & Campbell, W. K. (2008). Comparing clinical and social-personality conceptualizations of narcissism. Journal of Personality, 76, 449–476. Miller, J. D., Campbell, W., & Pilkonis, P. A. (2007). Narcissistic personality disorder: Relations with distress and functional impairment. Comprehensive Psychiatry, 48, 170–177. Miller, J. D., Hoffman, B. J., Gaughan, E. T., Gentile, B., Maples, J., & Campbell, W. K. (2011). Grandiose and vulnerable narcissism: A nomological network analysis. Journal of Personality, 79, 1013–1042. Miller, J. D., McCain, J., Lynam, D. R., Few, L. R., Gentile, B., MacKillop, J., et al. (2014). A comparison of the criterion validity of popular measures of narcissism and narcissistic personality disorder via the use of expert ratings. Psychological Assessment, 26, 958–969. Paulhus, D. L. (1998). Interpersonal and intrapsychic adaptiveness of trait selfenhancement: A mixed blessing? Journal of Personality and Social Psychology, 74, 1197–1208. Pilkonis, P. A., Choi, S. W., Reise, S. P., Stover, A. M., Riley, W. T., & Cella, D. (2011). Item banks for measuring emotional distress from the Patient-Reported Outcomes Measurement Information System (PROMIS): Depression, anxiety, and anger. Assessment, 18, 263–283. Pincus, A. L., & Lukowitsky, M. R. (2010). Pathological narcissism and narcissistic personality disorder. Annual Review of Clinical Psychology, 6, 421–446. Raskin, R., & 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. Ryan, K. M., Weikel, K., & Sprechini, G. (2008). Gender differences in narcissism and courtship violence in dating couples. Sex Roles, 58, 802–813. Sanders, B., & Becker-Lausen, E. (1995). The measurement of psychological maltreatment: Early data on the child abuse and trauma scale. Child Abuse and Neglect, 19, 315–323. Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15–28. Tanchotsrinon, P., Maneesri, K., & Campbell, W. (2007). Narcissism and romantic attraction: Evidence from a collectivistic culture. Journal of Research in Personality, 41, 723–730. Wink, P. (1991). Two faces of narcissism. Journal of Personality and Social Psychology, 61, 590–597.