bullying presentation

Personality and Individual Differences 91 (2016) 22–26

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

‘You can't sit with us:’ Gender and the differential roles of social intelligence and peer status in adolescent relational aggression

Della C. Loflin a, Christopher T. Barry b,⁎ a Auburn University, United States b Washington State University, United States

⁎ Corresponding author at: Department of Psychology Pullman, WA 99164, United States.

E-mail address: [email protected] (C.T. Barry).

http://dx.doi.org/10.1016/j.paid.2015.11.048 0191-8869/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history: Received 14 August 2015 Received in revised form 18 November 2015 Accepted 21 November 2015 Available online 28 November 2015

Keywords: Relational aggression Gender Social intelligence

Despite common notions that relational aggression (i.e., gossip, rumors, manipulation targeting another's social status; Crick & Grotpeter, 1995) is most commonly exhibited by females, there is mixed evidence regarding gen- der differences and what factors might heighten gender-specific risk for adolescent relational aggression (RA). The present study considered perceived social intelligence and peer status as such factors. Participants were 258 adolescents (217 males, 41 females) ranging in age from 16 to 19 attending a residential program for youth who have dropped out of school. Males and females did not differ on RA according to self-report or peer nominations. Self-reported social intelligence was associated with higher self-reported RA but only for females. The implications for future research as well as efforts for understanding RA are discussed.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Whereas aggression is defined as an act intended to injure or harm a victim (Archer & Coyne, 2005), relational aggression (RA) is specifically aimed at damaging the victim's relationships or social status. Instead of targeting a victim directly (e.g., physically, verbally), a person who en- gages in RA seeks to manipulate the target's relationships or peer groups in harmful ways (e.g., spreading a rumor about someone to encourage peer exclusion; Crick & Grotpeter, 1995). A variety of risk factors for RA in youth have been identified, particularly in the realm of social development. Specifically, previous research has focused on ele- ments of adolescents' peer context, including peer-rated popularity and social competence (e.g., McQuade, Achufusi, Shoulberg, & Murray- Close, 2014), as well as peer rejection/acceptance (see Card, Stucky, Sawalani, & Little, 2008) as relevant for RA.

The factors that might differentially increase the likelihood of relational aggression for adolescent males and females are not clearly understood despite numerous investigations. Some research has indi- cated that, for females, attention to social cues and understanding of be- havioral demands in social situations are linked to relational aggression (Andreou, 2006). Furthermore, much of the previous research has ex- amined factors involved in RA in females with relatively little emphasis on males. However, the evidence suggests that both males and females engage in relational aggression to some degree (McQuade et al., 2014), with a large meta-analysis also concluding that there are minimal gen- der differences in indirect, or relational, aggression (Card et al., 2008).

, Washington State University,

Traditionally, males have been described as engaging primarily in overt aggression, whereas females are more apt to exhibit RA. One explanation for this possible difference is that girls are more relationship-oriented than boys, meaning that friendships and other close relationships are particularly valued (Malove, 2014); thus, both positive and negative behaviors for females may be directed toward advancing females' peer interactions. Differences in the socialization of boys and girls could also contribute to how they exhibit aggression (Mikami, Lee, Hinshaw, & Mullin, 2008). Specifically, when encountered with a threat (e.g., peer rejection), some girls might utilize primarily relationally aggressive tactics, whereas boys may rely on physical or overt aggression (McQuade et al., 2014; Salmivalli & Kaukiainen, 2004). These conclusions are tempered by evidence that suggests min- imal or non-significant gender differences in adolescent RA (e.g., Card et al., 2008; Golmaryami & Barry, 2010) or even that male youth exhibit more RA than their female counterparts (e.g., Tomada & Schneider, 1997). In the context of social dominance, Hawley and colleagues (Hawley, Little, & Card, 2008) found that boys who self-identified as using prosocial and coercive tactics self-reported being the most relationally aggressive; however, peers still perceived males in general as being primarily overtly aggressive. It may be that RA is not as clearly perceived by peers when utilized by males. Clearly, the issue is complex and highlights a need to further examine social factors that might denote common or differential risk for RA across males and females.

The present study not only sought to extend previous research on social risk factors for RA but also to examine whether similar (or different) aspects of social functioning are tied to RA in male and fe- male adolescents. Specifically, the present study investigated whether adolescent self-reported social intelligence and peer status have

23D.C. Loflin, C.T. Barry / Personality and Individual Differences 91 (2016) 22–26

implications for RA and whether the role of such factors differs for fe- males and males.

1.1. RA and social intelligence

Social intelligence encompasses several cognitive, behavioral, and emotional concepts pertaining to one's perceived effectiveness in social interactions. According to Kaukiainen et al. (1999), social intelligence is the ability to use social skills to accomplish interpersonal tasks and/or an understanding of oneself in relation to others. Social intelligence can be further conceptualized as multidimensional, involving a) social information processing abilities; b) social skills; and c) social awareness (Silvera, Martinussen, & Dahl, 2001). Social information processing in- volves cognitive processes tied to encoding social cues and generating appropriate behavioral responses (Li, Fraser, & Wike, 2013). Thus, social information processing skills could allow one to know when or how to use RA effectively.

In the context of social intelligence, social skills capture one's capac- ity to engage in situation-appropriate social behaviors toward others, whereas social awareness involves one's ability to stay attuned to, and unsurprised by, what transpires in social situations (Delič, Novak, Kovačič, & Avsec, 2011). An individual with such social awareness may perceive him or herself capable of not only predicting what might happen in social interactions but also as adept at recognizing his/her influence on those interactions. Thus, an individual who per- ceives him/herself as having such awareness may utilize RA in situations where he/she believes it will enable the attainment of social goals. Similarly, an individual with relevant social skills may be effective at de- termining what behaviors, including RA, will meet his/her social goals.

As noted above, success in peer relationships may be defined some- what differently for males and females, with females being particularly concerned about connectedness with others (Crick & Grotpeter, 1995) and using RA in retaliation for perceived violations of relational norms or to achieve connections with some peers through their diminishment of the social status of other peers. Such a transaction could provide fe- males with what could be perceived as an effective way to bolster their social status. Viewing RA as an effective way to reach relationship goals might be relatively common for females who perceive themselves as socially intelligent. Among boys, Andreou (2006) noted a relation be- tween RA and the ability to predict others' feelings and reactions, whereas among girls, RA was connected to better interpreting social cues. All variables in that study were based, however, on peer percep- tions. Therefore, the examination of self-reported social intelligence could provide further insight into how RA is utilized by males and fe- males and whether gender differences exist in this regard. Perceived ability to master social situations could contribute to the use of RA as one way to achieve social goals, particularly as RA is more covert than physical or verbal assault. Traditional views suggest that such covert forms of aggression would be particularly appealing to females given the socialization of females to refrain from more overt aggression. The appeal might be particularly evident for females who perceive them- selves as socially intelligent and thus skilled at recognizing and seizing opportunities to engage in less direct aggression.

1.2. Perceived peer status and RA

In general, peer rejection has been tied to youth aggression (Crain, Finch, & Foster, 2005). However, the nature of this relation changes as children move into adolescence, with the association between peer re- jection becoming stronger for RA but lessening for overt aggression (Mayeux & Cillessen, 2008). Furthermore, some evidence suggests that for males, peer acceptance is related to indirect (relational) aggres- sion (Salmivalli, Kaukiainen, & Lagerspetz, 2000), whereas other evi- dence indicates that inaccurately high or low perceived social status is associated with male RA (McQuade et al., 2014). Thus, the precise role of peer status in RA remains unclear.

The present study examined the association between one's per- ceived peer status and RA, with the expectation that this interpersonal variable may be particularly important in males. Most studies provide little insight into the mechanisms driving males to utilize RA, and some findings may be an artifact of the assessment method (i.e., self- vs. peer-report). It is possible that peers admire, or perhaps fear, those who successfully use RA, contributing to a positive relation between peer nominations of acceptance and RA (Salmivalli et al., 2000). Howev- er, theoretically, based on previous findings concerning peer rejection and RA (e.g., McQuade et al., 2014; Tomada & Schneider, 1997) males who report engaging in RA may perceive themselves as rejected by peers or as needing to enact strategies (e.g., RA) to raise their status among peers. Males, in particular, may be more likely to use RA in cir- cumstances in which they perceive that their peer status is low and that this form of aggression might be useful for elevating that status. This view is consistent with the idea that RA is utilized to some males to attain social dominance (Hawley et al., 2008). For adolescent females, elevated self-perceptions of peer status appear to be tied to RA, although such perceptions may not be accurate relative to peer appraisals (McQuade et al., 2014).

1.3. The present study

As noted above, numerous studies have directly examined gender differences in RA with equivocal findings. This study investigated this issue with an emphasis not on gender differences in levels of RA but rather to determine if aspects of self-perceived social functioning would play a differential role in RA for male and female adolescents. In this way, the present study sought to complement the recent work by McQuade et al. (2014) by focusing particularly on adolescents' per- ceived social status and social competencies (i.e., social intelligence) as factors involved in RA. The present study is also one of the first known studies to include social awareness as it relates to adolescent RA.

1.4. Hypotheses

It was hypothesized that self-reported and peer-nominated RA would be moderately interrelated based on previous research (e.g., Golmaryami & Barry, 2010). It was expected that the association between social intelligence and RA would be moderated by gender. Spe- cifically, it was predicted that high levels of social intelligence would be associated with relatively high RA but especially for females. Further- more, it was hypothesized that for males, self-perceptions of low peer status would be positively related to self-reported RA but negatively re- lated to peer-nominated RA. For females, higher perceived peer status would be associated with higher RA based on the findings of McQuade et al. (2014) across informant methods.

2. Method

2.1. Participants

Participants were 258 adolescents (217 males, 41 females) ranging in age from 16 to 19 (M = 16.77 years, SD = .79), attending a voluntary 22-week military-style residential program for youth who have dropped out of school. Adolescents were not presently involved in the legal system. The sample was recruited from two consecutive cohorts from the program. The majority of participants (58.1%) were White, 37.7% were Black, and 4.2% of participants were from other racial/ethnic backgrounds.

An at-risk sample was selected with the anticipation that partici- pants from a residential, but non-behaviorally severe sample, would demonstrate greater variability on aggression than community or detained samples of adolescents (see Marsee et al., 2011). More impor- tantly, a residential sample affords a unique opportunity to gain further insight as to how aggression, particularly RA, plays out in a setting in

24 D.C. Loflin, C.T. Barry / Personality and Individual Differences 91 (2016) 22–26

which peers are in close contact (Golmaryami & Barry, 2010). Thus, peer report of RA was thought to be particularly informative for such a sam- ple of adolescents.

2.2. Materials

2.2.1. Peer Conflict Scale (PCS; Marsee, Kimonis, & Frick, 2004) The PCS is a 40-item self-report questionnaire designed to measure

different forms (i.e., overt, relational) of aggression. Participants rated each item (e.g., “Sometimes I gossip about others when I am angry with them”) on a scale from 0 (“not at all true”) to 3 (“definitely true”). The Relational Aggression scale, which includes reactive relation- al and proactive relational aggression subscales, was of particular inter- est in the present study. Previous research has demonstrated the validity of conceptualizing this scale as separate from items assessing overt aggression (Marsee et al., 2011). The internal consistency of the Relational Aggression scale was α = .93 in this sample.

2.2.2. Peer nomination of RA (Crick & Grotpeter, 1995) The peer nomination procedure described by Crick and Grotpeter

(1995) was used in the present study. Specifically, participants nominat- ed up to 3 peers from their platoon (see below) on each of 15 items which included attributes such as “is a leader” and “is shy” along with behaviors involving aggression. The 4 RA items (e.g., “When mad, gets even by keeping the person from being in their group of friends”) were of interest. The internal consistency across these items was α = .84. Following the procedures of Crick and Grotpeter, the number of nominations participants received for each item was summed and then z-scored within each group (i.e., platoon) to form the score for peer-nominated RA.

2.2.3. Tromsø Social Intelligence Scale (TSIS; Silvera et al., 2001) The TSIS consists of 21 items and provides a self-report account of

one's ability to identify social cues and effectively manage social situa- tions. The measure consists of 3 subscales: Social Information Processing, Social Skills, and Social Awareness. Respondents rate their perceived competence (e.g., “I can predict how others will react to my behavior”) on a 7-point Likert-type scale, ranging from “Completely disagree” to “Completely agree.” In the present sample, the internal consistency of the Social Intelligence composite was good (α = .83), and the internal consistencies of the Social Information Processing scale (α = .82) and the Social Awareness scale (α = .70) were adequate. However, the inter- nal consistency of the Social Skills scale (α = .42) was poor; consequent- ly, this subscale was not specifically considered in any subsequent analyses.

2.2.4. Personality Inventory for Youth (PIY; Lachar & Gruber, 1995) The PIY is an omnibus youth self-report rating scale of functioning in

a variety of behavioral, emotional, and social domains, with responses made in a True/False format. The Limited Peer Status subscale (e.g., “I'm often rejected by other kids”) was used in this study and dem- onstrated adequate internal consistency (α = .77) in this sample.

2.3. Procedure

The present study was approved by the Institutional Review Board (IRB) at the authors' affiliated university at the time data were collected. The director of the residential program, who serves as guardian ad litem for adolescents during their program enrollment, provided consent for them to be contacted about the study. Adolescents then provided writ- ten assent/consent if they elected to participate. Participation did not af- fect an individual's status in the residential program. Adolescents attending the program are organized into platoons of 20–35 individuals with whom they attend classes, participate in activities, and live in the same barracks. Self-report measures for this study and a larger research project were completed in classrooms separately by platoon in a series

of three to four 45-minute sessions over the course of approximately two weeks. The peer nomination procedure involved a separate as- sent/consent process and was conducted approximately one month fol- lowing the self-report phase of the study (i.e., approximately 18 weeks into the 22-week program). For the first cohort, 4 of 7 platoons were randomly selected for participation in the peer nomination procedure to reduce response burden as part of a larger project. All 7 platoons in the second cohort were invited to participate in this portion of the study. Thus, peer nomination data were only available for a subset of participants.

3. Results

Analyses concerning self-report variables were conducted for indi- viduals who completed all self-report measures (n = 258), whereas analyses that included peer nominations were conducted for partici- pants who completed the self-report and peer nomination phases of the study (n = 184). Descriptive statistics for the variables of interest are shown in Table 1. As shown in Table 1, the distribution for self- reported RA was positively skewed. As a result, scores for the two partic- ipants who scored at least 4 standard deviations above the sample mean on self-reported RA were transformed such that their scores were set to be one value above the next highest score in the sample. Analyses were conducted with and without these adjusted scores. Results are reported for non-transformed scores unless otherwise noted.

Correlations among study variables are shown in Table 2 and include all participants, with correlations shown separately for males and fe- males in Table 3. The first hypothesis was not supported, as self- reported RA and peer-nominated RA were not significantly related, r = .14, p = .07. However, as shown in Table 3, this relation was mod- erate in magnitude for females (r = .24), although not statistically sig- nificant. Peer-nominated RA was actually not significantly related to any of the other variables in this study. Correlational analyses revealed no significant gender differences on RA or social intelligence, with fe- males tending to report higher scores on limited peer status. Self- reported RA was associated with the Social Awareness component of social intelligence, r = .14, p = .02. Results of correlational analyses were virtually identical after transforming scores of the two outliers on self-reported RA.

To test the remaining hypotheses, a series of multiple regression analyses were conducted using PROCESS for SPSS (Hayes, 2013). First, the social intelligence composite from the TSIS and gender were entered simultaneously in a model predicting self-reported RA, with their inter- action term entered in a subsequent step. In the first step, there were no main effects for gender or social intelligence, R2 = .00. In the next step, the interaction term between gender and social intelligence predicted unique variance in RA, b = 4.24, se = 1.91, p = .03 R2 change = .03. Further inspection of this interaction according to the procedure outlined by Hayes (2013) indicated that higher perceived social intelli- gence was associated with self-reported RA for females, b = 4.78, se = 1.79, p = .008, but not for males, b = .54, se = .67, p = .42. This inter- action is depicted in Fig. 1. Follow-up regression models were examined for the components of social intelligence, but there were no significant interactions from these models.

The same procedure was followed for limited peer status as a predic- tor of self-reported RA. No significant main effects or interactions were evident in these models. The above regression models were analyzed with peer-nominated RA as the criterion variable with no significant in- teractions emerging. Lastly, with transformed scores on RA, the above interaction for overall social intelligence (Fig. 1) was marginally signifi- cant (i.e., p = .05).

4. Discussion

Social intelligence and RA were differentially related for males and females despite their overall lack of association in the present study.

Table 1 Descriptive statistics.

Variable (possible range) M SD Range Skew

Self-reported RA (0–60) 6.82 (6.81/7.28)

8.91 (8.26/11.46)

0–60 (0–60/0–59)

2.71 (2.38/2.87)

Peer-nominated RA .01 (.00/.01)

.96 (.97/.93)

−1.61–4.02 (−1.61/3.79)

1.46 (1.35/2.19)

Social intelligence (1–7) 4.05 (4.01/4.26)

.88 (.88/.76)

1–7 (1–7/2.52–5.62)

−.22 (−.16/−.19)

Social information processing (1–7) 4.29 (4.27/4.46)

1.26 (1.27/1.24)

1–7 (1–7/2.14–7)

−.12 (−.12/.10)

Social awareness (1–7) 3.84 (3.77/4.15)

1.16 (1.13/1.25)

1–7 (1–7/2–6.86)

.26 (.27/.18)

Limited peer status (0–12) 5.16 (5.07/5.27)

3.06 (3.06/2.79)

0–11 (0–11/1–11)

.23 (.20/.45)

Note: Statistics for the overall sample are shown, with those for males followed by females in parentheses underneath (males/females). Peer-nominated RA z-scored within platoon. Social intelligence and its components are based on mean-item ratings on a scale ranging from 1 to 7.

25D.C. Loflin, C.T. Barry / Personality and Individual Differences 91 (2016) 22–26

That is, females who reported higher levels of social intelligence report- ed engaging in significantly higher RA than their male counterparts or females with low perceived social intelligence. Despite similar levels of self-reported RA for males and females, perceived social intelligence only heightened RA for females, whereas none of the variables investi- gated in this study appeared to be specifically associated with RA for males.

Differences in gender socialization and the methods of assessing so- cial intelligence and RA may offer some insight concerning our findings. Based on the present findings, perceiving that one is adept at recogniz- ing ways to manipulate social situations or opportunities to retaliate for perceived relationship slights seems to increase the likelihood that fe- males would use such a strategy. As noted above, females perhaps value connectedness with others more than males and use RA when the desired affiliation is not reciprocated. Males who perceive them- selves as socially intelligent may not be as driven to use such skills for engaging in RA. Males may value and engage in RA, but again, the self- perception factors that contribute to such behavior may not include so- cial intelligence. A small, significant relation existed between social awareness and RA independent of gender, indicating that perceived ability to predict social interactions is a factor in one's self-reported RA, yet the relative importance of this construct for RA overall appears minimal.

Methodology may also have contributed to the findings or lack of support for some hypotheses. Specifically, social intelligence as assessed in this study was inherently a self-perception construct. We lacked a way to objectively evaluate participants' actual effectiveness at under- standing, reading, and predicting social interactions vis-á-vis their re- ported capacity to do so. Further, the lack of association between self- reported peer status and RA could be partly a function of self-reports of these constructs. Perhaps some adolescents who engage in RA are not particularly accurate in evaluating the quality of their peer interac- tions. Interestingly, many of the previous studies that found support

Table 2 Correlations among main study variables.

1. 2. 3. 4. 5. 6. 7.

1. Relational aggression (self-report) 2. Relational aggression (peer-report)

.14

3. Social intelligence .10 −.02 4. Social information-processing .01 −.05 .79⁎⁎⁎

5. Social awareness .14⁎ .02 .77⁎⁎⁎ .35⁎⁎⁎

6. Limited peer status −.09 −.10 −.02 −.02 .00 7. Gender .09 −.01 .10 .07 .10 .13⁎

Note: Peer-report: n = 184, all other analyses: n = 258; gender coded as 1 = male, 2 = female. ⁎ p b .05.

⁎⁎⁎ p b .001.

for peer status as a factor in RA used peer-referenced assessment of peer status.

Finally, peer-nominated RA did not significantly relate to the other study variables, including gender. Prior research suggests that other self-perception variables (e.g., narcissism, self-esteem; Golmaryami & Barry, 2010) may be key in understanding attributes that peers perceive as being tied to RA. Additionally, adolescents, particularly females, with higher social intelligence may be quite adept at obscuring RA from peers.

The present study had a number of limitations. The specificity of the sample (i.e., residential) and the predominance of males may limit the generalizability of findings to adolescents in general. The lack of gender differences in relational aggression, although not unique to this study, may speak to the atypicality of this sample of adolescents in that their pattern of aggression may be different or more severe than that of many adolescents. This sample provided a unique opportunity for peer-reporting, yet further understanding of gender differences in risk factors for RA is needed in broader, more diverse samples. Contrary to expectations, self-reported RA and peer-reported RA were positively, but non-significantly, related (r = .14). These two indices did appear to relate differently for males and females which is consistent with the pattern of findings in previous research (e.g., Hawley et al., 2008). The lack of overall relation may also be a function our sample being mostly male, the varied social behaviors that might have factored into peer's perceptions of RA, or the breadth of content in the self-report measure relative to that used for peer nominations. Although the peer nomina- tion approach has been widely used in previous research, only four items are involved. Because this measure likely does not capture the full scope of RA, further refinement of peer assessments of RA is needed. Finally, the poor internal consistency of the TSIS Social Skills scale pre- cluded its examination and suggests some inconsistency or difficulty in responding to self-report of social adeptness in this sample. Still, the overall social intelligence composite had good internal consistency and was the focus of the main analyses.

Table 3 Correlations among variables by gender.

1. 2. 3. 4. 5. 6.

1. RA (self-report) – .02 .11 .03 .16⁎ −.07 2. RA (peer-report) .24 – −.06 −.11 .00 −.10 3. Social intelligence .29 .20 – .82⁎⁎⁎ .77⁎⁎⁎ −.03 4. Social information-processing .08 .18 .58⁎⁎⁎ – .34⁎⁎⁎ −.02 5. Social awareness .20 .07 .76⁎⁎⁎ .03 – −.05 6. Limited peer status −.08 .08 .06 .00 .20 –

Note: Correlations for males are above the diagonal; correlations for females are below the diagonal. ⁎ p b .05.

⁎⁎⁎ p b .001.

Fig. 1. Interaction between social intelligence and gender in predicting self-reported RA.

26 D.C. Loflin, C.T. Barry / Personality and Individual Differences 91 (2016) 22–26

The moderation between gender and social intelligence was rela- tively modest in magnitude. Future research should consider an array of specific intra- or inter-personal variables (e.g., empathy) involved in RA or that could explain for whom or in what situations social intel- ligence is related to aggression. Considering the present findings, it could be argued that social intelligence is undesirable for females based on its association with RA. However, the presence of empathy coupled with social intelligence could reduce utilization of RA. That is, if an individual perceives him/herself as skilled at understanding social situations and also feels concern for the impact of social actions on an- other, RA would be less likely. Furthermore, a specific examination of social goals and motives may provide context for understanding how social intelligence might be used maladaptively.

The developmental processes involved also need further attention, particularly as they pertain to gender socialization and its potential role in RA. The central questions of this study may also be applicable to wider forms of RA (e.g., cyber/electronic), as perceived social intelli- gence might further the appeal of engaging in RA that is potentially more anonymous. In short, continued efforts to determine factors that pertain specifically to young males and females in their use of an array of relationally aggressive behaviors hold the most promise for promot- ing efforts to prevent or reduce this covert form of aggression.

References

Andreou, E. (2006). Social preference, perceived popularity, and social intelligence rela- tions to overt and relational aggression. School Psychology International, 27, 339–351.

Archer, J., & Coyne, S.M. (2005). An integrated review of indirect, relational, and social ag- gression. Personality and Social Psychology Review, 9, 212–230.

Card, N.A., Stucky, B.D., Sawalani, G.M., & Little, T.D. (2008). Direct and indirect aggression during childhood and adolescence: A meta-analytic review of gender differences, in- tercorrelations, and relations to maladjustment. Child Development, 79, 1185–1229.

Crain, M.M., Finch, C.L., & Foster, S.L. (2005). The relevance of the social information pro- cessing model for understanding relational aggression in girls. Merrill-Palmer Quarter- ly, 51, 213–249.

Crick, N.R., & Grotpeter, J.K. (1995). Relational aggression, gender, and socio-psychological adjustment. Child Development, 66, 710–722.

Delič, L., Novak, P., Kovačič, J., & Avsec, A. (2011). Self-reported emotional and social intel- ligence and empathy as distinctive predictors of narcissism. Psychological Topics, 3, 477–488.

Golmaryami, F.N., & Barry, C.T. (2010). The associations of self-reported and peer- reported relational aggression with narcissism and self-esteem among adolescents in a residential setting. Journal of Clinical Child and Adolescent Psychology, 39, 128–133.

Hawley, P.H., Little, T.D., & Card, N.A. (2008). The myth of the alpha male: A new look at dominance related beliefs and behaviors among adolescent males and females. International Journal of Behavioral Development, 32, 76–88.

Hayes, A.F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford.

Kaukiainen, A., Björkqvist, K., Lagerspetz, K., Österman, K., Salmivalli, C., Rothberg, S., & Ahlbom, A. (1999). The relationships between social intelligence, empathy, and three types of aggression. Aggressive Behavior, 25, 81–89.

Lachar, D., & Gruber, C.P. (1995). Personality inventory for youth (PIY) manual: Technical guide. Los Angeles, CA: Western Psychological Services.

Li, J., Fraser, M.W., & Wike, T.L. (2013). Promoting social competence and preventing childhood aggression: A framework for applying social information processing theory in intervention research. Aggression and Violent Behavior, 18, 357–364.

Malove, S.C. (2014). Using relational aggression theory to treat adolescent girls victimized by social aggression. Clinical Social Work Journal, 42, 1–12.

Marsee, M.A., Barry, C.T., Childs, K.K., Frick, P.J., Kimonis, E.R., Munoz, L.C., ... Lau, K.S.L. (2011). Assessing the forms and functions of aggression in youth using self-report: Confirmatory factor analysis and evaluation of factorial invariance across sample and gender. Psychological Assessment, 23, 792–804.

Marsee, M. A., Kimonis, E. R., & Frick, P. J. (2004). Peer conflict scale. Unpublished measure.

Mayeux, L., & Cillessen, A.H.N. (2008). It's not just being popular, it's knowing it too: The role of self-perceptions of status in the associations between peer status and aggres- sion. Social Development, 17, 871–888.

McQuade, J.D., Achufusi, A.K., Shoulberg, E.K., & Murray-Close, D. (2014). Biased self- perceptions of social competence and engagement in physical and relational aggres- sion; The moderating role of peer status and sex. Aggressive Behavior, 40, 512–525.

Mikami, A.Y., Lee, S.S., Hinshaw, S.P., & Mullin, B.C. (2008). Relationships between social information processing and aggression among adolescent girls with and without ADHD. Journal of Youth and Adolescence, 37, 761–771.

Salmivalli, C., & Kaukiainen, A. (2004). ‘Female aggression’ revisited: Variable and person- centered approaches to studying gender differences in different types of aggression. Aggressive Behavior, 30, 158–163.

Salmivalli, C., Kaukiainen, A., & Lagerspetz, K. (2000). Aggression and sociometric status among peers: Do gender and type of aggression matter? Scandinavian Journal of Psychology, 41, 17–24.

Silvera, D.H., Martinussen, M., & Dahl, T.I. (2001). The tromsø social intelligence scale, a self-report measure of social intelligence. Scandinavian Journal of Psychology, 42, 313–319.

Tomada, G., & Schneider, B.H. (1997). Relational aggression, gender, and peer acceptance: Invariance across cultures, stability over time, and concordance among informants. Developmental Psychology, 33, 601–609.