Aggressive and prosocial behaviors

Abstract

This study examined the social functioning of bistrategic youths (i.e., those who employ both aggressive and prosocial behavior) in order to further understand their social competence in peer social networks. Within our sample of 318 fifth-grade participants recruited from an urban school district in the northeastern US, bistrategic preadolescents were identified, along with aggressives, prosocials, typicals, and noncontrollers. Bistrategic youths were found to hold the highest individual status and group status among all five subtypes within the network. Bistrategic youths and other active controller youths (i.e., prosocials and aggressives) were more likely to associate with peers of the same subtype in social groups. In addition, bistrategics who associated with same subtype (i.e., homogenous) peers held elevated group status, whereas bistrategics who associated with non-homogenous peers held heightened individual status. Further, results illustrate how bistrategics selectively target highly popular peers as victims of social aggression. Findings extend previous research by demonstrating the social success of bistrategic youths based on their high individual and group status, and by illuminating their strategic employment of different forms of aggression towards peers of different popularity status.

Keywords

physical and social forms of aggression prosocial behavior bistrategic subtype peer groups popularity early adolescence

Extant research has demonstrated how aggressive behavior may serve to elevate a youth’s social status (Adler & Adler, 1998; Prinstein & Cillessen, 2003; Rodkin, Farmer, Pearl, & van Acker, 2000). Likewise, prosocial behavior has separately been shown to win friends and garner interpersonal success (Eisenberg, Morris, McDaniel, & Spinrad, 2009). Further, evidence has indicated that prosocial behavior may buffer the negative effects of aggression in adolescent friendships (McDonald, Wang, Menzer, Rubin, & Booth-LaForce, 2011). Bistrategic resource control (i.e., highly exhibiting both prosocial and coercive behavior strategies) has gained attention for its proposed optimal means of obtaining social dominance (Hawley, 1999). Bistrategic youths are viewed as highly popular and adept at resource control by fellow peers (Hawley, 2003; Olthof, Goossens, Vermande, Aleva, & van der Meulen, 2011), but information is needed on how these youths function within their social networks. As a youth’s social profile is not limited to individual perception, information on peer group functioning is helpful in order to illustrate a more comprehensive view of the social dynamics among bistrategic and other behavior subtype youths.

This article sought to examine the social functioning of bistrategic youths within their peer networks by analyzing their social status and peer group composition, as well as the recipients of the aggressive and prosocial behaviors that they employ. Specifically, we investigated bistrategic youths in comparison to four other behavior subtypes (i.e., noncontrollers, typicals, prosocials, and aggressives), which were all classified based on peer reported aggressive and prosocial behavior in this study.

Social resource control through aggression and affiliation

For a school-aged youth, social dominance is a desirable achievement as it relates to attaining important social resources (Hawley, 1999), including peer attention, status, and prominence (Adler & Adler, 1998; Farmer et al., 2002). Yet, social dominance is only tenable to a small proportion of those who seek it; according to the natural rules of social hierarchy, few individuals can occupy the upper rungs of the social ladder (Hawley, 1999). As with every resource that does not plentifully abound, competition is necessary to viably obtain it (Darwin, 1859/2009). The competition for social resources implicates the usage of social interaction strategies, including social behaviors (Hawley, 1999).

Prosocial behavior, the art of kindness, is recognized as a means of winning social favor by employing grace, goodwill, and support (Eisenberg et al., 2009). Prosocial behavior may go beyond simply being a good friend or likeable child, as affiliative behaviors can also be used strategically in the pursuit of goal attainment (Hawley, 2003; Pellegrini, 2008). However, while prosocial behavior offers a means of gaining resources under certain circumstances, this does not always translate to social prowess and true dominance (Pellegrini, 2008). Prosocial behavior may be used as a means of cooperation with others, but coercive methods may be necessary in order to climb the social hierarchy and emerge as socially dominant (Hawley, 2003).

In contrast, aggression is also a successful method of social resource control. While aggressive youths are often nominated by peers as being low on “liked” (i.e., peer preference) measures, they are also viewed as being popular and high status (Robertson et al., 2010; Xie, Cairns, & Cairns, 1999). In this way, aggression may be evolutionarily adaptive, as it promotes fitness and is advantageous for interindividual and intergroup competition (Vaughn & Santos, 2007). Additionally, aggressive youths who do maintain high status are more likely to remain both aggressive and popular over time (Shi & Xie, 2012), thus perpetuating their social success.

However, even for the socially skilled, aggressive behavior can elicit costs: punishment by authorities (Hawley, 2003), or revenge by targeted peers (Björkqvist, Österman, & Lagerspetz, 1994). Theoretically, bistrategic behavior usage offers the relational benefits associated with affiliative behavior as well as the dominance benefits associated with coercive behavior (Hawley, 2003). Further, bistrategic youths should be able to maximize the benefits of aggression while mediating costs with prosociality (Pellegrini, 2008). The conciliatory ability of prosocial behavior has been demonstrated as an important post-conflict recompense in non-human primates (Aureli, Cords, & van Schaik, 2002). Evidence has shown that preschoolers who display greater reconciliation behaviors are more likely to be socially dominant (Roseth et al., 2011). In addition, prosocial behavior has also been shown to moderate aggression’s negative impact on best friendship among adolescents (McDonald et al., 2011). Further, the use of prosocial behavior may allow youths’ concurrent aggressive behavior to be misperceived or less noticeable by authority figures (Hawley, 2003). Thus, bistrategic behavior usage should be indicative of a socially competent youth who is skilled at obtaining social resources.

Group status and individual status

Although previous research has demonstrated bistrategic youths’ success at obtaining perceived popularity and peer preference (Hawley, 2003), this offers only a partial view of their social position. Perceived popularity and preference are effective measures of a youth’s social capital (Cillessen & Mayeux, 2004), but there are other forms of social resources. During childhood and adolescence, peer interaction occurs both within and between peer groups (Rubin, Bukowski, & Parker, 2006). Groups vary in their own status level (Cairns, Cairns, Neckerman, Gest, & Gariépy, 1988), and the social reputation of a child is partially dependent upon the standing of the group to which he or she belongs (Adler & Adler, 1998). Further, this differentiation in status among groups allows for variation in groups’ access to social resources, such as peer attention (Farmer et al., 2002). Thus, the peer group itself is a means of obtaining status.

Further, the individual status a youth holds within his group can vary from that of a nuclear member, representing a vital role, to a peripheral member, representing a child on the fringe of the group (Cairns et al., 1988). High individual status within a group represents prominence, and allows nuclear members primary access to resources within the group; prominent members may be especially influential in group decision-making, or in setting behavioral norms (Adler & Adler, 1998; Latané, 1981; Shi & Xie, 2012). As bistrategic youths have already demonstrated their ability to obtain high popularity, preference, and resource control (Hawley, 2003; Olthof et al., 2011), we would expect these youths to achieve high network status, as well. It is thus proposed that bistrategics will be nuclear members within their own peer groups, and belong to high status groups within their peer network.

Birds of a feather

A related question concerns the type of peers with whom bistrategics affiliate, and how the overall behavior displayed by a peer group will influence a member’s own status. Homophily among behaviors has been suggested to be an important basis for friendships (McPherson, Smith-Lovin, & Cook, 2001), and evidence has demonstrated similarity among friend dyads in prosocial and aggressive based behavior subtypes (Hawley, Little, & Card, 2007). As most friendship dyads are embedded within peer groups (Cairns, Leung, Buchanan, & Cairns, 1995), it is reasonable to expect that homophily in behavior subtypes would be evident in groups, as well. We hypothesize that bistrategics and all other subtypes would be significantly more likely to belong to groups containing same subtype peers than cross subtype peers.

In regards to social resource control and intergroup interaction, befriending others of similar behavior subtypes can be advantageous. Behavior norms may promote consistency in the strategies in which group mates engage in the defense of group status (Nesdale, Milliner, Duffy, & Griffiths, 2009). However, depending on the type of behavior utilized, homogeneity in behavior subtypes in a group may limit the effectiveness of behavior strategy and serve less competitive advantage. It seems reasonable to expect that individuals in groups that are homogenous in behavior strategies may hold higher group status if the strategies enacted are socially competitive. Bistrategic youths have been evidenced as high status and highly effective at resource control (Hawley, 2003; Olthof et al., 2011), thus a group displaying overall bistrategic behavior should be advantageous in regards to intergroup competition. Relatedly, individuals in groups that exhibit ineffective behavior usage in regards to resource control (e.g. low aggressive or low prosocial behavior) would hold lower group status. Based on this, we hypothesize that bistrategic youths who belong to peer groups displaying bistrategic behavior will have higher competitive power, and will therefore hold high levels of group status.

However, homogeneity in socially adept (i.e., bistrategic) behavior within a group could heighten intragroup competition. Previous evidence has suggested that dyadic friendships of bistrategic youths are highly prone to both relational and overt aggression (Hawley et al., 2007) demonstrating that bistrategic youths display coercive strategies within their own relationships. To that extent, the effectiveness of resource attainment based on these strategies (Hawley, 2003; Olthof et al., 2011), could undermine each individual’s status within a collectively bistrategic group. Specifically, as social dominance is based on a power differential and the ability to obtain social resources more effectively than fellow peers (Hawley, 1999; Savin-Williams, 1979), peer groups that are largely bistrategic in nature should be composed of competitive, dominant individuals, subsequently heightening resource competition within the group. In this way, bistrategics would be more likely to emerge as a leader and hold individual prominence in groups that are not bistrategic in nature. We therefore expect that bistrategics should display higher individual status when belonging to non-homogenous groups.

Selection of behavior recipients

An additional set of questions concerns how bistrategics enact social behaviors, and who they select as recipients of their behavior. The utilization of specific behaviors as a means of gaining status implies not only that these behaviors may be used selectively, but that the targets of the behaviors may be chosen selectively as well. Those who are socially dominant should therefore be adept at targeting specific individuals; different behaviors have different costs and benefits (Pellegrini, 2008), and these costs and benefits may vary depending on the social power of the behavior target.

For example, aggressing against a high status peer demonstrates the aggressor is likely a worthy opponent, and competition with those who are adept at achieving social resources (e.g. status) may be necessary in order to obtain dominant positions in the social hierarchy (Hawley, 1999). However, targeting a low status victim and taking advantage of a power differential (Olweus, 1993) can help ensure victory and minimize the risk of revenge. Engaging in physical aggression offers the benefit of observed dominance, but risks opening the aggressor to punishment by authority figures, as well as revenge, defeat, or injury by peers (Hawley, 2007; Pellegrini, 2008). Social aggression, conversely, offers the benefit of inflicting damage as well as the reduced likelihood of detection (Xie, Cairns, & Cairns, 2005; Xie, Swift, Cairns, & Cairns, 2002). To that extent, directing social aggression towards high status peers may allow a youth to ascend his social hierarchy while buffering negative consequences. Indeed, evidence has shown that highly popular youths are more likely to be targets of reputational aggression (Prinstein & Cillessen, 2003). In light of this, we predict that bistrategic youths may target either low or high status peers, but that they will primarily use social aggression towards high status recipients as it may reduce the risk of retaliation.

Finally, prosociality is critical in the pursuit of successful social competition. Prosocial behavior usage is beneficial as it may establish friendships and demonstrates loyalty (Eisenberg et al., 2009), and prosocial youths are shown to be high on peer preference (Cillessen & Mayeux, 2004). Additionally, prosocial strategies may be used to help remediate damage caused by conflict and reconcile relations in both children (Pellegrini, 2008) and in non-human primates (de Waal, 1993). We predict that bistrategics will target high status peers as recipients of prosocial behavior for two reasons: first, because social relationships with these youths would be most desirable (Hawley, 1999), and second, because the reduction of antipathy with powerful peers would be necessarily protective post-conflict (Pellegrini, 2008).

The present study

To summarize, the goal of the present study was to more explicitly examine bistrategic youths’ social status and peer group composition, as well as to explore the utilization of behaviors within social dyads. First, we predicted that in comparison to other behavior subtypes, bistrategics would be more likely to be nuclear members of their own peer groups, and second, that these groups would be of higher status than peer groups to which other behavior subtypes belong. Third, we hypothesized that bistrategic youths would be significantly more likely to belong to peer groups comprised of other bistrategics than would other behavior subtypes, in order to effectively compete for social resources and social dominance within their network. Fourth, we hypothesized that bistrategic youths who belong to bistrategic peer groups (i.e., homogenous with their own behavior subtype) would hold higher group status than other subtypes, whereas bistrategic youths who belong to peer groups which exhibit a different behavior strategy profile (i.e., non-homogenous) would hold higher individual status than other subtype peers. Last, we hypothesized that in comparison to other active controller subtypes (i.e., prosocial and aggressive youths) bistrategics would selectively target high status recipients when using social aggression in order to be more covert, and they should primarily direct prosocial behavior towards high status peers, as well.

Method

Participants

Children in Grade 5 were recruited from six elementary schools in an urban school district in the northeastern United States. Written informed consent from a parent or guardian was obtained before any child participated. A total of 318 children, resulting in a 61% participation rate, took part in the study (49% males, M age = 11.00, SD = 0.44). The families of the participating children were largely of a lower middle to middle class socioeconomic status, and 61.5% were eligible for free or reduced-price lunch, based on family income totaling 1.35 times the poverty line and 1.85 times the poverty line, respectively, as per federal requirements (Department of Agriculture, 2005). Ethnic representation was 48% African-American, 33% Caucasian, 18% Hispanic, and 2% other ethnicity (including Asian and multi-ethnicity). Participants and non-participants did not differ on race, χ² (3, 492) = 4.84, p = .183, or lunch status, χ² (2, 488) = 2.63, p = .269. Participants and non-participants did differ on gender, χ² (1, 522) = 6.66, p = .012.

Procedure

In the spring semester of 2005, trained research assistants and the principal investigator administered a battery of questionnaires to participants during a 45-minute session in each classroom. During administration, a lead research assistant read the instructions aloud, while additional assistants provided mobile assistance to students. Before beginning the survey, participants were assured of confidentiality. A semi-structured individual interview (audio-recorded) was conducted with the primary investigator and each participant in school. Questions about peer group, friendship, perception of popularity, and interpersonal conflicts were asked during the interview. Each interview lasted 25 to 45 minutes, and was later transcribed verbatim.

Measures

Peer reported aggressive and prosocial behavior

Participants were asked to nominate, from free recall, three peers in their class who best fit descriptor items of prosocial and aggressive behavior. Participants were asked to nominate students who were prosocial (e.g. “This person is always willing to help other kids,” “This person is usually friendly to others,” r = .65). Participants were also asked to nominate peers displaying social aggression (“This person gossips and says things about others,” “This person is good at causing people to get mad at each other,” r = .65), and physical aggression (“This person starts fights. This person pushes other kids or hits them,” “This person bullies others. This person is always hurting or picking on others,” r = .74). Proportion scores for each behavior were created by dividing the number of nominations for items by the total number of potential nominators in the class.

Following previous methodology (Hawley et al., 2007; Olthof et al., 2011), strategy use subtypes were determined by dividing peer nominations of aggressive and prosocial behavior usage into thirds in order to classify participants into groups.

Social network measures

The Social Cognitive Map (SCM) procedure (e.g. Cairns, & Cairns, 1994) was employed to identify peer groups in each classroom. During individual interviews, participants were asked to nominate peer groups in their class: “Are there any students in your class who hang around together a lot?” The SCM procedure was then used to aggregate group nominations from all participants in a classroom that met nomination requirements (i.e., groups must be nominated two or more times). Following previous studies (e.g. Cairns, Perrin, & Cairns, 1985), no class list was provided, and participants nominated groups from free recall. In previous studies, groups identified by the SCM procedure have been validated by direct observations of social interactional patterns (e.g. Cairns et al., 1985; Gest, Farmer, Cairns, & Xie, 2003). A total of 96 groups were identified (Xie & Shi, 2009), with group size ranging from 2 to 12 participants (M = 4.98, SD = 1.95). Thirty participants were isolates who were not identified as belonging to any peer social groups.

Individual status within a group, representing how prominent an individual is within his peer group, was measured by Individual Nomination into a peer group, that is to say, the number of times an individual was nominated into a specific peer group (e.g. Cairns et al., 1985), which was then standardized within group to control for group differences in nominations. Groups that had identical values for nominations among members did not produce z scores. Zeroes, the mean score of a standardized distribution, were added as the z scores for the participants in these groups. This substitution was made on the basis that missing scores for these individuals should not occur, and that individual members who received equal nominations into a given group should hold the same level of individual status. The assignment of zero, the average score of a standardized distribution, was made in order to demonstrate that group members did not hold high (represented by a greater amount of nominations) or low (fewer nominations) individual status within their group. The 30 isolates did not have any score for individual status within a group because they did not belong to any group. Subsequently, these isolates were excluded from all analyses that included individual status as the dependent variable.

Group status was determined by a group nomination score calculated as the average of the Individual Nomination scores of the two most central (i.e., highest individual status) members. As peers freely nominated participants, some participants were nominated into multiple groups (Xie & Shi, 2009). The calculation of individual and group status was based on membership in primary group. For isolates, a score of zero was assigned for their group nomination, because they were not identified as members of any peer group. The scores for group nomination were subsequently standardized by class to assess group status.

Peer nominations of perceived popularity and preference

Participants were asked to nominate up to three peers of any gender who fit the descriptor “popular” (“Some kids are very popular with their peers. That is, many classmates like to play with them or do things with them.”). Perceived popularity was assessed based on the proportion score on this item, which was calculated by dividing the number of nominations received for a child by the total possible number of nominators in the class (Lease, Musgrove, & Axelrod, 2002; Parkhurst & Hopmeyer, 1998). Participants were also asked to nominate three peers who they “like most” and three peers who they “like least,” and peer preference was calculated as the difference between the proportion scores on each child’s “Like most” and “Like least” items (Cillessen & Mayeux, 2004; Lease et al., 2002).

Behavior dyads

In order to examine the usage of behavior strategies we conducted analyses on dyad nominations for social aggression, physical aggression, and prosocial behavior. Peer reports of aggressor-victim dyads were used for both physical and social aggression. Towards the end of the individual interview, each participant was asked to identify, from free recall, victims of physical aggression (“Some students get hit or pushed by other students. Can you think of anyone in your class whom this happens to a lot?”). The interviewer wrote down the first name and the initial of the last name of each victim nominated on the left column of an answer sheet. Then for each victim, the participant was asked to nominate the aggressors (“Who hit or pushed X the most?”). The interviewer then wrote down each aggressor’s name on the right column of the sheet and drew a line between the names of the aggressor and the victim. Each participant was allowed to nominate unlimited victims and aggressors.

A similar procedure was used for social aggression. Each participant was asked to first identify the aggressors (“Some students get other people to turn against another student. Can you think of anyone in your class who does this a lot?”). The interviewer wrote down the first name and first initial of the last name of each aggressor on the left column of an answer sheet. For each aggressor, the participant was asked to nominate the victims (“Whom does X get people to turn against?”). The interviewer then wrote down each victim’s name on the right column of the sheet and drew a line between the names of the aggressor and the victim.

Recipients’ reports of prosocial behavior dyads were measured by asking participants to nominate peers who had acted prosocially towards them as part of a questionnaire (“This person has given you help when needed” and “This person stood up for you”). Participants were allowed to nominate an unlimited number of peers.

Nominations were coded into dyads. For each perpetrator of physical and social aggression, victims were identified, and for each prosocial provider, recipients were identified. An average score on perceived popularity was then created across recipients of (a) prosocial behavior, (b) physical aggression, and (c) social aggression separately in order to measure the mean popularity of the recipients for each behavior type.

Results

Classification of behavior subtypes

Participants were classified into subtypes based on high (top third), middle, or low (bottom third) behavior usage, as determined by peer report. The subtypes were formed as follows: bistrategic controllers (n = 34) scored in the top third on both behavior dimensions; aggressive controllers (n = 104) scored in the top third on aggressive behavior but average or low on prosocial behavior; prosocial controllers (n = 73) scored in the top third on prosocial behavior but average or low on aggressive behavior; non-controllers scored in the lower third on both dimensions (n = 26); all others were typicals (n = 81). Full distribution by gender can be seen in Table 1.¹

Table 1.

Proportion of participants in each subtype by gender.

	Noncontroller	Typical	Prosocial	Aggressive	Bistrategic	Total
Gender
Girl	9	32	53	44	25	163
Boy	17	49	20	60	9	155
Total	26 (8.2%)	81 (25.5%)	73 (22.9%)	104 (32.7%)	34 (10.7%)	318

Configural frequency analysis was used to examine the distributions of gender within each subtype by assessing whether the observed frequency of each cell occurred more often or less often than expected by chance. Analyses were performed using the BASIC program developed by von Eye (1990), controlling for overall error rate with a Bonferroni-adjusted alpha of .05 / (2 × 5) = .005. Analyses revealed that greater numbers of prosocials were female (f _e = 37.418, f _o = 53, t = 4.150, p < .001) than expected by chance, and greater numbers of bistrategics were female (f _e = 17.428, f _o = 25, t = 2.745, p < .001) than expected by chance. For this reason, all analyses on subtypes’ social functioning and behavior dyads were run with gender as a control variable.

Social status within the network

The social functioning of participants was examined in three aspects: network centrality, perceived popularity, and peer preference. Network centrality was indexed by two measures: individual status in group and group status. In order to test our hypotheses that bistrategics would be nuclear members of their peer groups, and that these groups would be high-status in comparison to other groups in the network, we performed analyses of individual status and group status by participant subtype. The two-way ANOVA performed on individual status controlling for gender returned a significant effect by subtype, F(4, 284) = 4.21, η_p ² = .056, p = .002. In order to test our hypothesis that bistrategics would hold higher individual status than all other subtypes, direct comparisons were made by individual ANOVAs comparing bistrategics with each alternate subtype. Note that participants identified as isolates, who had no score for individual status due to no group membership, were not included in these analyses. Results showed that bistrategics held significantly higher individual status than noncontrollers, F(1, 54) = 13.98, η_p ² = .209, p < .001, typicals, F(1, 106) = 11.04, η_p ² = .095, p = .001, and aggressives, F(1, 128) = 4.45, η_p ² = .034, p = .037; comparisons between bistrategics and prosocials approached significance, F(1, 95) = 3.32, η_p ² = .034, p = .072, with bistrategics holding higher individual status.

Analyses of group status were performed on the entire sample, including isolates who had been assigned the lowest score for group nominations. The two-way ANOVA performed on group status controlling for gender revealed a significant effect by subtype, F(4, 314) = 5.76, η_p ² = .069, p < .001. Comparisons between bistrategics with each of the other subtypes revealed that bistrategics belonged to higher status groups than noncontrollers, F(1, 59) = 10.46, η_p ² = .153, p = .002, typicals, F(1, 114) = 6.92, η_p ² = .058, p = .010, and prosocials, F(1, 106) = 8.37, η_p ² = .074, p = .005; bistrategics were not different from aggressives, F(1, 137) = 0.26, p = .611.

In order to confirm previous findings (Hawley, 2003), we examined bistrategics’ social status in two other measures: perceived popularity and peer preference. An ANOVA performed on peer nominated perceived popularity controlling for gender returned a significant effect by subtype, F(4, 314) = 13.07, η_p ² = .145, p < .001. Bistrategics held higher perceived popularity than noncontrollers, F(1, 59) = 32.91, η_p ² = .362, p < .001, typicals, F(1, 114) = 38.52, η_p ² = .254, p < .001, prosocials, F(1, 106) = 4.40, η_p ² = .040, p = .038, and aggressives, F(1, 137) = 5.81, η_p ² = .041, p = .017. Lastly, an ANOVA performed on peer preference revealed significant differences among subtypes, F(4, 314) = 12.37, η_p ² = .138, p < .001, with bistrategics holding higher preference than typicals, F(1, 114) = 6.32, η_p ² = .053, p = .013, and aggressives, F(1, 137) = 15.58, η_p ² = .103, p < .001. Comparisons between bistrategics and noncontrollers tended towards significance, F(1, 59) = 3.29, η_p ² = .054, p = .075, with bistrategics holding higher preference; bistrategics were no different than prosocials on preference, F(1, 106) = 0.43, p = .514. Full results may be seen in Table 2.

Table 2.

Mean differences in peer perceptions of social functioning across subtypes.

	Noncontroller		Typical		Prosocial		Aggressive		Bistrategic
Peer nominations	M	(SD)	M	(SD)	M	(SD)	M	(SD)	M	(SD)	F
Individual status	−.38	(.81)	−.16	(.84)	.09	(.82)	.08	(.83)	.41	(.73)	4.21**
Group status	−.40	(1.07)	−.04	(.95)	−.23	(1.04)	.25	(.86)	.34	(.67)	5.76**
Popularity	.02	(.04)	.08	(.13)	.19	(.23)	.19	(.21)	.30	(.24)	13.07**
Peer preference	.03	(.15)	.02	(.19)	.16	(.19)	−.07	(.25)	.13	(.26)	12.37**

Note. ** p < .01. N = 288 for Individual Status, N = 318 for Group status, Peer-rated popularity, and Peer preference. Individual status = individual status within peer group. Group status = group status within social network. Individual status and group status are standardized scores. Popularity and peer preference measured by a proportion score of nominations by peers in class.

Peer group homophily

In order to understand similarity between a participant’s subtype and the subtype of peers in their group, we examined the average profile of behavior strategy use of each participant’s group mates.² We first calculated the mean levels of aggressive and prosocial behaviors across a participant’s group mates, and then classified mean group behavior in an identical fashion as the classification methodology used for individual participants (i.e., groups whose mean level of prosocial behavior fell in the highest third for the sample, and whose mean level of aggressive behavior fell in the highest third for the sample, were classified as bistrategic groups).

Configural frequency analysis was used to examine the distributions of subtypes with same subtype peers using the BASIC program (von Eye, 1990), controlling for overall error rate with a Bonferroni-adjusted alpha of .05 / (5 × 5) = .002. Analyses revealed that there were greater numbers of prosocials associating with peers displaying prosocial behavior (f _e = 13.125, f _o = 22, t = 3.110, p < .001) than expected, that there were greater numbers of aggressives associating with peers displaying aggressive behavior (f _e = 29.333, f _o = 49, t = 5.327, p < .001) than expected, and that there were greater numbers of bistrategics associating in peer groups displaying bistrategic behavior (f _e = 4.240, f _o = 12, t = 4.283, p < .001) than expected. These findings demonstrate that all active controller types were more likely to associate with other same subtype peers. ³ Full results are presented in Table 3.

Table 3.

Distribution of average group mate behavior subtypes by participant subtype.

Subtype	Non-controller group mate subtype	Typical group mate subtype	Prosocial group mate subtype	Aggressive group mate subtype	Bistrategic group mate subtype	Total
Non-controller	1	11	2	6	2	22
Typical	5	32	14	18	5	74
Prosocial	1	20	22^*	10	10	63
Aggressive	3	21	15	49^*	8	96
Bistrategic	2	7	7	5	12^*	33
Total	12	91	60	88	37	288

Note. Configural Frequency Analyses revealed that the match between prosocial, aggressive, and bistrategic youths and their group mates’ average behavior subtype exceeded the expected counts, *ps < .05. N = 288. Values in bold font represent distribution of participants in each subtype with same group mate subtypes.

Homogeneity of peer affiliation and social status

In order to examine how homogeneity of behavior subtypes within a group could affect a participant’s individual status and group status, we classified group mate behaviors into homogenous or non-homogenous in accordance with participants’ own subtype; those participants whose behavior subtype matched the subtype of their group mates were coded as affiliating with homogenous peers, and those participants whose behavior subtype was discrepant from the subtype of their group mates were coded as affiliating with non-homogenous peers. Noncontrollers were removed from the following analyses due to low nomination into peer groups, and subsequently low sample representation here.

ANOVAs performed on individual status by subtype for homogenous peer affiliations, controlling for gender, revealed no significant differences between subtypes, F(3, 102) = 0.89, p = .448. ANOVAs performed on individual status by subtype for non-homogenous peer affiliations, controlling for gender, revealed significant differences between subtypes, F(3, 128) = 4.25, η_p ² = .089, p = .003. When affiliating with non-homogenous peers, bistrategics held higher individual status within group than typicals, F(1, 52) = 11.40, η_p ² = .183, p = .001, prosocials, F(1, 50) = 5.76, η_p ² = .105, p = .020, and aggressives, F(1, 59) = 8.31, η_p ² = .125, p = .006. Full results are presented in Table 4.

Table 4.

Mean differences in individual and group status by homogeneity of peer affiliations across participant subtype.

	Typical		Prosocial		Aggressive		Bistrategic
	M	(SD)	M	(SD)	M	(SD)	M	(SD)	F(η_p ²)
Individual status Homogenous peer affiliations	−.14	(.81)	.16	(.79)	.13	(.84)	.20	(.75)	.89 (.027)
Non-homogenous peer affiliation	−.17	(.93)	.08	(.93)	.03	(.85)	.72	(.72)	4.02^* (.089)
Group status Homogenous peer affiliations	−.48	(1.05)	−.19	(1.08)	.38	(.70)	.53	(.43)	7.98^** (.177)
Non-homogenous peer affiliations	.04	(.99)	−.20	(.95)	.17	(.83)	−.21	(.99)	1.43 (.029)

Note. *p < .05; **p < .01. Ns for subtypes in individual status range from 11 to 47. Ns for subtypes in group status range from 12 to 49. Due to low N, noncontrollers not included in analyses. Homogenous peer affiliations = group mate mean behavior subtype matching participant behavior subtype. Non-homogenous peer affiliations = group mate mean behavior subtype different from participant behavior subtype. Individual status = individual status within peer group. Group status = group status within social network. Individual status and group status are standardized scores.

ANOVAs performed on group status by subtype for homogenous peer affiliations, controlling for gender, revealed significant differences between subtypes, F(3, 111) = 6.75, η_p ² = .177, p < .001. When affiliating with homogenous peers, bistrategic youths’ groups were of higher status than those of typical youths, F(1, 43) = 10.14, η_p ² = .195, p = .003, and prosocial youths, F(1, 33) = 4.76, η_p ² = .129, p = .037. There was no difference in group status between bistrategics and aggressives, F(1, 60) = 0.48, p = .491. ANOVAs performed on group status by subtype for non-homogenous peer affiliations, controlling for gender, revealed no significant differences between subtypes, F(3, 142) = 1.29, p = .277. Full results are presented in Table 4.

Popularity of behavioral recipients by subtype

In order to test our hypotheses that bistrategics would selectively choose the recipients of their prosocial and aggressive behaviors, we first identified the participants in each subtype who were nominated as an instigator in each type of behavioral dyad: prosocial behavior, physical aggression, and social aggression. The recipients’ popularity for each behavior was calculated as the mean score across all recipients’ popularity for each instigator. We should note that not every participant in each subtype was nominated as an instigator in these behavioral dyads. The analyses for each type of behavior only included those participants who were nominated as instigators for that behavior.

Two sets of analyses were conducted. In the first set, we examined subtype differences in their recipients’ social status when specific behaviors were used: prosocial, social aggression, and physical aggression. For each type of behavior, a two-way ANOVA was performed, examining differences on recipient popularity across the instigator’s subtype, controlling for gender. For these analyses, only prosocial, aggressive, and bistrategic subtypes were included as instigators. This decision was made because prosocial, aggressive, and bistrategic controllers are “active” behavior controllers; noncontrollers and typicals are, by nature, those who do not use a high level of either prosocial or aggressive behavior (Hawley, 2003), and may further fail to selectively target others. Additionally, the noncontrollers and typicals within our sample were infrequently nominated as instigators for any of the three types of behavior.

Analyses performed on recipients of prosocial behavior revealed that differences between instigator subtypes on recipients’ popularity approached significance, F(2, 135) = 3.31, η_p ² = .048, p = .056. Direct comparisons showed that recipients of bistrategic youths’ use of prosocial behavior tended to have higher popularity than recipients of prosocial youths, F(1, 80) = 3.47, η_p ² = .043, p = .066. Bistrategic youth and aggressive youth did not differ on the popularity of their prosocial behavior recipients, F(1, 82) = 0.25, p = .875. Analyses performed on victims of physical aggression revealed no significant differences between subtypes on their victims’ popularity, F(2, 95) = 0.17, p = .846. Analyses performed on victims of social aggression revealed significant differences between subtypes, F(2, 56) = 5.41, η_p ² = .17, p = .015. No differences on victims’ popularity were found between prosocial and bistrategic youths, F(1, 20) = 0.09, p = .768. However, social aggression victims of bistrategic youths had higher popularity than the victims of aggressive youths, F(1, 47) = 6.21, η_p ² = .121, p = .016. Full results are presented in Table 5.

Table 5.

Means differences in recipient popularity by behavior form across participant subtype.

	Prosocial subtype instigators		Aggressive subtype instigators		Bistrategic subtype instigators
Behavior type	M (SD)	N	M (SD)	N	M (SD)	N	F
Prosocial behavior recipients	.16 (.16)	54	.22 (.19)	56	.21 (.18)	28	3.31^†
Physical aggression recipients	.10 (.18)	5	.10 (.14)	57	.11 (.17)	16	0.17
Social aggression recipients	.14 (.15)	10	.14 (.16)	36	.23 (.16)	12	5.41^**

Note. †p < .06; **p < .01. Instigators = those using prosocial behavior, physical aggression, or social aggression. Recipients = targets of prosocial behavior, physical aggression, or social aggression. Popularity measured by proportion score of nominations of popularity by peers in class.

In the second set of analyses, we compared the popularity of victims across both physical and social forms of aggression by the same instigators in order to test whether bistrategics selectively targeted high status peers as victims of social aggression. For each of the three subtypes, we identified those who were nominated as using both social and physical aggression towards others. A repeated measures ANOVA, with an interaction term between subtype and aggression form, was performed. A significant subtype by aggression form interaction was found, F(2, 31) = 7.36, p = .002. As portrayed in Figure 1, when these subtypes used both forms of aggression, the disparity in mean popularity between victims of physical and social forms of aggression was much greater for bistrategics than that of prosocials and aggressives, demonstrating that bistrategics selectively target highly popular victims when using social aggression.

Figure 1.

Victim popularity: differential targeting using physical and social aggression by the same instigator varies across subtypes.

Discussion

By classifying participants based on peer nominations of aggressive and prosocial behavior (Hawley, 2003), we found five unique behavior subtypes: bistrategics, aggressives, prosocials, typicals, and noncontrollers. Findings largely confirmed our hypotheses. Results demonstrated the social success of bistrategic youths based on their high group status within their social network, as well as their high individual status within their peer groups. In addition, this study discovered a tendency for bistrategics and other active controller subtypes to associate with peers of a similar behavior subtype. Further, we determined that bistrategics who associated with same subtype peers held elevated group status, whereas bistrategics who associated with non-homogenous peers held heightened individual status. Finally, analyses of aggressor-victim dyads for physical and social aggression illustrate how bistrategics are selective in choosing their targets depending on the form of aggression; when confronting a high status victim, bistrategics tended to use social aggression rather than physical aggression.

This study replicated previous findings (Hawley, 2003) and reaffirmed bistrategic youths’ ability to achieve both popularity and peer preference. Prosocial and bistrategic youths shared similar levels of peer preference in this study, which indicates that the relational benefits associated with prosociality may outweigh the negative impact of aggression in regards to how bistrategic youths are perceived by fellow peers. In addition, bistrategic youths may themselves be representative of a social resource based on their high status, further propelling their social magnetism and associated preference among peers (Hawley, Little, & Card, 2007). To that extent, this study demonstrates that bistrategic youths exhibit a degree of social skillfulness and subsequently enjoy interpersonal benefits.⁴

As predicted, bistrategics held high levels of both individual and group status. This finding expands on existing knowledge of the social prowess of bistrategics (Hawley, 2003) and offers a view of bistrategics’ place in their own social hierarchies. While bistrategic youths were similar to aggressive youths on group status, and only trended towards being significantly higher than prosocial youths on individual status, holding high levels of status on both dimensions demonstrates their ability to achieve ultimate social dominance, as they are able to reap the benefits of both forms of social resources. First, bistrategic youths demonstrate their ability to capture peer attention in their broad social network (Adler & Adler, 1998; Farmer et al., 2002) by being members of high status groups. Second, bistrategic youths are then able to control resources, as well as demonstrate leadership and influence, within their own peer group (Cairns et al., 1988; Latané, 1981). This finding identifies bistrategics as not only socially competent, but also as socially dominant “key players” in their social network.

We had predicted that bistrategic youths would associate in peer groups that were bistrategic in nature. This tendency to associate with same subtype peers was true not only for bistrategic youths, but for aggressive and prosocial youths as well. This tendency to associate with peers of a similar behavior profile is consistent with previous findings on homophily in friendships and groups (Cairns, Xie, & Leung, 1998; Hawley et al., 2007; Kandel, 1978). It is important to note previous research has determined evidence for homophily in popularity among friends (Marks, Cillessen, & Crick, 2012; Witvliet et al., 2010), which may supersede the effect of behavioral homophily (Dijkstra, Berger, & Lindenberg, 2011). To that extent, the homophily among behavior usage within groups may be a result of the similar status level shared by friends.

What is also illuminating in this study is the effect of associating with homogenous or non-homogenous peer groups on a bistrategic youth’s own status. On one hand, a bistrategic who affiliates with peers employing overall bistrategic behavior gains elevated group status. On the other, a bistrategic who affiliates with peers employing a non-homogenous strategy (e.g. prosocial behavior) is more likely to emerge as a leader in his group. In this way, bistrategic youths demonstrate their ability to maximize access to resources (Hawley, 2003) either by being prominent and influential within his or her group (Adler & Adler, 1998; Latané, 1981; Shi & Xie, 2012) as a high status group member, or else by capturing attention as a member of a high status group (Farmer et al., 2002).

Finally, findings from the dyad analyses illustrate how bistrategics achieve social dominance through the selection of their behavior targets. It was hypothesized that bistrategics will strategically select targets who likely will maximize the benefits of the behaviors employed. Findings showed that victims of bistrategics’ social aggression held higher status than the victims of prosocials or aggressives. This difference could be due to the fact that bistrategics in our sample were high status themselves; it is likely that bistrategics were afforded the opportunity of regular social interaction with these victims (McPherson et al., 2001). To test this possibility, we further conducted within-subject analyses; examining the difference in popularity between targets of social and physical aggression, we found that bistrategics indeed were selective in their choice of targets, as their physical aggression victims were not similarly high status. These findings illustrate bistrategics’ proficiency in interaction for two reasons. First, by targeting high status peers to aggress against, they are choosing a worthy competitor, as it serves the bistrategic’s interests for these peers to be “knocked down” in their social hierarchy (Prinstein & Cillessen, 2003). Second, by using social aggression, bistrategics are aggressing in a concealed manner, and are less likely to suffer retribution from the target of the aggression (Xie et al., 2005, 2002).

Bistrategics were not found to differ from aggressives and prosocials in their selection of recipients when utilizing prosocial behavior or physical aggression. As was previously noted, using prosocial behavior may help obtain social preference (Cillessen & Mayeux, 2004) as well as repair damage post-conflict (Pellegrini, 2008). In addition, selecting lower status peers as targets of physical aggression is socially advantageous; targeting weaker victims of aggression (Olweus, 1993; Veenstra et al., 2007) offers the benefit of reduced retaliation. In this respect, all three behavior subtypes demonstrated social aptitude to some extent in the employment of physical aggression and prosocial behaviors. Additionally, noncontroller and typical youths were not included in dyad analyses based on their low nomination by peers into these behavior dyads. This is notable for two reasons. First, this reinforces the classification of noncontroller and typical youths as being low to moderate on aggressive and prosocial behaviors. Second, failing to be nominated into these dyads by peers implies that these youths fail to be noticed by peers, and subsequently have limited impact in their social network (Hawley, 2003).

Limitations and future directions

Several limitations should be addressed. First, the present study has a correlational design. Future studies should examine bistrategics longitudinally in order to determine whether they preserve their high status, or maintain their subtype over time. Previous work has shown that popularity predicts future aggression (Cillessen & Mayeux, 2004), and that youths who are popular and aggressive are likely to continue to be both over time (Shi & Xie, 2012). This may imply that bistrategics are likely to both maintain their prominence as well as continue to aggress throughout their tenure in their social context. Second, while the patterns of participant distribution in the five behavior subtypes were similar to those in Hawley’s (2003) study as well as Olthof and colleagues’ (2011) study (i.e., more aggressive boys, more prosocial girls), the actual proportions were discrepant. This difference could be due to our classification based on behaviors as opposed to behavior strategies, in that behaviors may be more susceptible to gender stereotypes or social sanctions (Underwood, Galen, & Paquette, 2001), whereas strategies, as perceived by self and peers, may supersede social norms in the interest of achieving a specific goal. Further, proportions may vary as a result of type of reporter upon which classifications are based; within a single sample, distributions of participants within subtypes have been shown to change based on whether classifications are based on peer or self report (Olthof et al., 2011). Third, our study had a 61% participation rate, which could limit the generalizability of our findings. However, it should be noted that findings from our sample were similar to those of other samples with higher participant rates (Hawley, 2003; Hawley et al., 2007; Olthof et al., 2011). The limited sample size also prevented us from examining gender differences in behavioral and social profiles of the subtypes. Analyses were performed controlling for gender in order to account for the over-representations of boys in the aggressive subtype and of girls in the prosocial subtype. It is important for future studies to determine possible gender differences in the social functions of different subtypes, as this has not been systematically investigated in previous studies. Finally, as many of our variables were formed on the basis of peer nominations, it is possible that some of our findings (i.e., homophily among group mates) may be confounded by reputational influence (Hawley et al., 2007). However, previous research has evidenced similar findings to ours on homophily based on teacher-rated variables (Xie et al., 1999), which we believe underscores the reliability of our peer-nominated measures.

Findings from our study demonstrate important implications for intervention strategies among youths in the attempt to diminish aggressive behavior. As bistrategic youths in our sample and others (Hawley, 2003; Olthof et al., 2011) are high status, it is possible that traditional intervention strategies may exempt these youths, as they do not fit the “typical” profile of a socially maladaptive aggressive youth. In addition, future studies should continue to investigate the social perception and social goals of bistrategic youths. Previous findings have shown that bistrategic youths desire social dominance more so than other behavior subtypes, with the exception of prosocial youths (Olthof et al., 2011), and that they believe themselves to be effective at resource control (Hawley, 2003). Future investigation may explore specific facets of social cognition, including behavior attributions, goals, and perception of others’ social standing.

Conclusions

Overall, the findings in this study extend previous work on bistrategics’ social competence (Hawley, 2003; Hawley et al., 2007), as well as identify status and peer group implications for all behavior subtypes. Although previous findings on bistrategic youths demonstrated their popularity, preference, and friendships, this study illustrates specific resource attainment in regards to group and individual status, as well as identifies possible mechanisms for displaying dominance via their group membership and selection of behavior recipients. Further, these findings reinforce the reality that youths who display aversive behaviors are not necessarily socially incompetent; instead, those who moderate aggressive behavior with prosociality are evidently socially skilled and highly successful (Hawley, 2003; Hawley et al., 2007). To that extent, this study suggests that not only are bistrategic youths noteworthy within their social contexts, but these individuals are deserving of continued attention by researchers, as well.

Footnotes

Funding

This research was supported by grants from the National Science Foundation [NSF 0339070] and William T. Grant Foundation awarded to the second author.

*This article was accepted under the editorship of previous Editor-in-Chief, Marcel van Aken.

Notes

References

Adler

P. A.

Adler

(1998). Peer power: Preadolescent culture and identity. Piscataway, NJ: Rutgers University Press.

Aureli

Cords

van Schaik

C. P.

(2002). Conflict resolution following aggression in gregarious animals: A predictive framework. Animal Behaviour, 64, 325–343.

Björkqvist

Österman

Lagerspetz

K. J.

(1994). Sex differences in covert aggression among adults. Aggressive Behavior, 20, 27–33. doi:10.1002/1098-2337(1994)20:1<27:: AID-AB2480200105>3.0.CO;2-Q

Cairns

R. B.

Cairns

B. D.

(1994). Lifelines and risks: Pathways of youth in our time. New York, NY: Cambridge University Press.

Cairns

R. B.

Cairns

B. D.

Neckerman

H. J.

Gest

S. D.

Gariépy

(1988). Social networks and aggressive behavior: Peer support or peer rejection? Developmental Psychology, 24, 815–823. doi:10.1037/0012-1649.24.6.815

Cairns

R. B.

Leung

M. C.

Buchanan

Cairns

B. D.

(1995). Friendships and social networks in childhood and adolescence: Fluidity, reliability, and interrelations. Child Development, 66, 1330–1345. doi: 10.1111/j.1467-8624.1995.tb00938.x

Cairns

R. B.

Perrin

J. E.

Cairns

B. D.

(1985). Social structure and social cognition in early adolescence: Affiliative patterns. The Journal of Early Adolescence, 5, 339–355. doi:10.1177/0272431685053007

Cairns

R. B.

Xie

Leung

M. C.

(1998). The popularity of friendship and the neglect of social networks: Toward a new balance. New Directions for Child Development, 80, 25–53.

Cillessen

A. N.

Mayeux

(2004). Sociometric status and peer group behavior: Previous findings and current directions. In Kupersmidt

J. B.

Dodge

K. A.

(Eds.), Children’s peer relations: From development to intervention (pp. 3–20). Washington, DC: American Psychological Association. doi:10.1037/10653-001

10.

Darwin

(2009). The origin of species (1859). In Gentile

B. F.

Miller

B. O.

(Eds.), Foundations of psychological thought: A history of psychology (pp. 320–336). Thousand Oaks, CA: SAGE.

11.

de Waal

F. M.

(1993). Reconciliation among primates: A review of empirical evidence and unresolved issues. In Mason

W. A.

Mendoza

S. P.

(Eds.), Primate social conflict (pp. 111–144). Albany: State University of New York Press.

12.

Department of Agriculture (2005). Food and nutrition service: Child nutrition programs—income eligibility guidelines. Federal Register, 70(52), pp. 13160–13163.

13.

Dijkstra

Berger

Lindenberg

(2011). Do physical and relational aggression explain adolescents’ friendship selection? The competing roles of network characteristics, gender, and social status. Aggressive Behavior, 37, 417–429. doi:10.1002/ab.20402

14.

Eisenberg

Morris

McDaniel

Spinrad

T. L.

(2009). Moral cognitions and prosocial responding in adolescence. In Lerner

R. M.

Steinberg

(Eds.), Handbook of adolescent psychology. Vol 1: Individual bases of adolescent development (3rd ed., pp. 229–265). Hoboken, NJ: John Wiley & Sons Inc.

15.

Farmer

T. W.

Leung

Pearl

Rodkin

P. C.

Cadwallader

T. W.

Van Acker

(2002). Deviant or diverse peer groups? The peer affiliations of aggressive elementary students. Journal of Educational Psychology, 94, 611–620. doi:10.1037/0022-0663.94.3.611

16.

Gest

S. D.

Farmer

T. W.

Cairns

B. D.

Xie

(2003). Identifying children’s peer social networks in school classrooms: Links between peer reports and observed interactions. Social Development, 12, 513–529. doi:10.1111/1467-9507.00246

17.

Hawley

P. H.

(1999). The ontogenesis of social dominance: A strategy-based evolutionary perspective. Developmental Review, 19, 97–132. doi:10.1006/drev.1998.0470

18.

Hawley

P. H.

(2003). Prosocial and coercive configurations of resource control in early adolescence: A case for the well-adapted Machiavellian. Merrill-Palmer Quarterly: Journal of Developmental Psychology, 49, 279–309. doi:10.1353/mpq.2003.0013

19.

Hawley

P. H.

(2007). Social dominance in childhood and adolescence: Why social competence and aggression may go hand in hand. In Hawley

P. H.

Little

T. D.

Rodkin

P. C.

(Eds.), Aggression and adaptation: The bright side to bad behavior (pp. 1–29). Mahwah, NJ: Lawrence Erlbaum Associates.

20.

Hawley

P. H.

Little

T. D.

Card

N. A.

(2007). The allure of a mean friend: Relationship quality and processes of aggressive adolescents with prosocial skills. International Journal of Behavioral Development, 31, 170–180. doi:10.1177/0165025407074630

21.

Kandel

D. B.

(1978). Homophily, selection, and socialization in adolescent friendships. American Journal of Sociology, 85, 427–436. doi:10.1086/226792

22.

Latané

(1981). The psychology of social impact. American Psychologist, 36, 343–356.

23.

Lease

Musgrove

K. T.

Axelrod

J. L.

(2002). Dimensions of social status in preadolescent peer groups: Likability, perceived popularity, and social dominance. Social Development, 11, 508–533. doi:10.1111/1467-9507.00213

24.

Marks

P. E. L.

Cillessen

A. H. N.

Crick

N. R.

(2012), Popularity contagion among adolescents. Social Development, 21, 501–521. doi:10.1111/j.1467-9507.2011.00647.x

25.

McDonald

K. L.

Wang

Menzer

M. M.

Rubin

K. H.

Booth- LaForce

(2011). Prosocial behavior moderates the effects of aggression on young adolescents’ friendships. International Journal of Developmental Science, 5, 127–137. doi:10.3233/DEV-2011-10066

26.

McPherson

Smith-Lovin

Cook

J. M.

(2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415–444. doi:10.1146/annurev.soc.27.1.415

27.

Nesdale

Milliner

Duffy

Griffiths

J. A.

(2009). Group membership, group norms, empathy, and young children’s intentions to aggress. Aggressive Behavior, 35, 244–258. doi:10.1002/ab.20303

28.

Olthof

Goossens

F. A.

Vermande

M. M.

Aleva

E. A.

van der Meulen

(2011). Bullying as strategic behavior: Relations with desired and acquired dominance in the peer group. Journal of School Psychology, 49, 339–359. doi:10.1016/j.jsp.2011.03.003

29.

Olweus

(1993). Bully/victim problems among schoolchildren: Long-term consequences and an effective intervention program. In Hodgins

(Ed.), Mental disorder and crime (pp. 317–349). Thousand Oaks, CA: SAGE.

30.

Parkhurst

J. T.

Hopmeyer

(1998). Sociometric popularity and peer-perceived popularity: Two distinct dimensions of peer status. The Journal of Early Adolescence, 18, 125–144. doi:10.1177/0272431698018002001

31.

Pellegrini

A. D.

(2008). The roles of aggressive and affiliative behaviors in resource control: A behavioral ecological perspective. Developmental Review, 28, 461–487. doi:10.1016/j.dr.2008.03.001

32.

Prinstein

M. J.

Cillessen

A. N.

(2003). Forms and functions of adolescent peer aggression associated with high levels of peer status. Merrill-Palmer Quarterly: Journal of Developmental Psychology, 49, 310–342. doi:10.1353/mpq.2003.0015

33.

Robertson

D. L.

Farmer

T. W.

Fraser

M. W.

Day

S. H.

Duncan

Crowther

Dadisman

K. A.

(2010). Interpersonal competence configurations and peer relations in early elementary classrooms: Perceived popular and unpopular aggressive subtypes. International Journal of Behavioral Development, 34, 73–87. doi:10.1177/0165025409345074

34.

Rodkin

P. C.

Farmer

T. W.

Pearl

Van Acker

(2000). Heterogeneity of popular boys: Antisocial and prosocial configurations. Developmental Psychology, 36, 14–24. doi:10.1037/0012-1649.36.1.14

35.

Roseth

C. J.

Pellegrini

A. D.

Dupuis

D. N.

Bohn

C. M.

Hickey

M. C.

Hilk

C. L.

Peshkam

(2011). Preschoolers’ bistrategic resource control, reconciliation, and peer regard. Social Development, 20(1), 185–211. doi:10.1111/j.1467-9507.2010.00579.x

36.

Rubin

K. H.

Bukowski

W. M.

Parker

J. G.

(2006). Peer interactions, relationships, and groups. In Eisenberg

Damon

Lerner

R. M.

(Eds.), Handbook of child psychology: Vol. 3, Social, emotional, and personality development (6th ed., pp. 571–645). Hoboken, NJ: John Wiley & Sons Inc.

37.

Savin-Williams

R. C.

(1979). Dominance hierarchies in groups of early adolescents. Child development, 50, 923–935. doi:10.2307/1129316

38.

Shi

Xie

(2012). Socialization of physical and social aggression in early adolescents’ peer groups: High-status peers, individual status, and gender. Social Development, 21, 170–194. doi:10.1111/j.1467-9507.2011.00621.x

39.

Underwood

M. K.

Galen

B. R.

Paquette

J. A.

(2001). Top ten challenges for understanding gender and aggression in children: Why can’t we all just get along? Social Development, 10, 248–266. doi:10.1111/1467-9507.00162

40.

Vaughn

B. E.

Santos

(2007). An evolutionary/ecological account of aggressive behavior and trait aggression in human children and adolescents. In Hawley

P. H.

Little

T. D.

Rodkin

P. C.

(Eds.), Aggression and adaptation: The bright side to bad behavior (pp. 31–63). Mahwah, NJ: Lawrence Erlbaum Associates.

41.

Veenstra

Lindenberg

Zijlstra

B. H.

De Winter

A. F.

Verhulst

F. C.

Ormel

(2007). The dyadic nature of bullying and victimization: Testing a dual-perspective theory. Child Development, 78, 1843–1854. doi:10.1111/j.1467-8624.2007.01102.x

42.

von Eye

(1990). Introduction to configural frequency analysis: The search for types and antitypes in cross-classifications. New York, NY: Cambridge University Press.

43.

Witvliet

Olthof

Hoeksma

J. B.

Goossens

F. A.

Smits

M. S.

Koot

H. M.

(2010). Peer group affiliation of children: The role of perceived popularity, likeability, and behavioral similarity in bullying. Social Development, 19, 285–303. doi: 10.1111/j.1467-9507.2009.00544.x

44.

Xie

Cairns

B. D.

Cairns

R. B.

(2005). The development of aggressive behaviors among girls: Measurement issues, social functions, and differential trajectories. In Pepler

D. J.

Madsen

K. C.

Webster

Levene

K. S.

(Eds.), The development and treatment of girlhood aggression (pp. 105–136). Mahwah, NJ: Lawrence Erlbaum Associates.

45.

Xie

Cairns

R. B.

Cairns

B. D.

(1999). Social networks and social configuration in inner-city schools: Aggression, popularity, and implications for students with EBD. Journal of Emotional and Behavioral Disorders, 7, 147–156. doi:10.1086/226792

46.

Xie

Shi

(2009). Gender similarities and differences in preadolescent peer groups: Group structure and ethnic diversity. Merrill-Palmer Quarterly: Journal of Developmental Psychology, 55, 157–183. doi:10.1353/mpq.0.0021

47.

Xie

Swift

D. J.

Cairns

R. B.

(2002). Aggressive behaviors in social interaction and developmental adaptation: A narrative analysis of interpersonal conflicts during early adolescence. Social Development, 11, 205–224. doi:10.1111/1467-9507.00195