Exploring Ethnic Variation in Preadolescent Aggressive Girls’ Social,Psychological,and Academic Functioning

Abstract

The authors examined whether the adjustment patterns of socially and overtly aggressive preadolescent girls, ages 9 to 11 years, from rural communities differed by ethnicity. Students were administered a series of questionnaires to assess the degree to which girls engaged in various forms of aggression and to assess aggressive girls’ social, psychological, and academic functioning. Four clusters of girls—nonaggressive, socially aggressive, and moderate and high mixed aggressive—were identified with cluster analysis based on cross-gendered peer nominations of social and overt aggression. Results indicated that ethnic variation exists in aggressive and nonaggressive girls’ attitudes toward school and satisfaction with their friendships, but no ethnic differences were found in aggressive girls’ social adjustment. Socially aggressive and nonaggressive African American girls endorsed greater feelings of disengagement from school than socially aggressive and nonaggressive European American girls. No ethnic differences were found in aggressive girls’ endorsement of internalizing symptoms. However, both nonaggressive and socially aggressive African American girls reported greater dissatisfaction with their interpersonal relationships than nonaggressive and socially aggressive European American girls.

Keywords

social aggression overt aggression girls ethnic differences adjustment

There is a growing body of research that suggests that the social and psychological adjustments of aggressive preadolescent girls are influenced by the form of aggression exhibited: social or overt aggression (Crick, 1995, 1997; Crick & Grotpeter, 1995; Crick, Ostrov, & Werner, 2006). Social aggression is socially manipulative and exclusionary verbal and nonverbal behaviors that are perceived as hurtful by victims and that intentionally damage the reputation, self-esteem, and interpersonal relationships of victims (Xie, Cairns, & Cairns, 2005; Xie, Farmer, & Cairns, 2003; Underwood, 2003). In contrast, overt aggression is characterized by physically and direct verbally aggressive behaviors. Studies have shown that in middle childhood or preadolescence girls who exhibit overt aggression are at greater risk for experiencing social and psychological adjustment difficulties than socially aggressive girls (Crick, 1995), perhaps due to their violation of gendered norms for anger expression (Crick, 1997). However, at earlier stages of development, overt aggression might be somewhat more acceptable for girls, as the skills needed to successfully manipulate and socially aggress against peers have yet to develop fully. Hawley (1999), for example, has outlined the developmental progression of skills needed for children to transition from overt to socially aggressive strategies to dominate peers. Continued reliance on overt means of aggression in preadolescence might be indicative of a failure to develop more sophisticated social skills and/or might set the stage for a disordered developmental trajectory.

By middle childhood, research suggests that children perceive social aggression as a female typical form of aggression and overt aggression as male typical (Crick, Bigebee, & Howes, 1996; Giles & Heyman, 2005). This in part might be due to children’s reliance on gender schemas to determine the appropriateness of their own behavior as well as to evaluate the behavior of their peers (Ostrov & Godleski, 2010). In their expanded theory of Crick and Dodge’s (1994) social information processing model, Ostrov and Godleski (2010) purported that children’s selection of aggressive responses are guided by children’s gender schemas, which might be influenced by peer (i.e., gender-segregated play) and adult socialization (e.g., teacher and parent) practices, their previous experiences with aggression (e.g., vicarious and direct exposure), or their own interpersonal goals.

Peer and adult gender socialization inadvertently shape children’s understanding of gender-appropriate behaviors and roles (Leaper, 2000). Therefore, it is possible that girls who are socialized to embody more traditional standards of femininity, which discourage the outward expression of anger, might be more likely to view social aggression as an acceptable form of anger expression than girls who are socialized to be androgynous, adopting both masculine and feminine characteristics (Brown & Gilligan, 1992). For androgynous girls, social aggression might serve as one method for aggressing against peers, but overt aggression might also be a viable option. This may be particularly true for African American girls who are more likely than European American girls to be socialized to be androgynous (Belgrave, 2009; Leaper, 2000; Lewis, 1975) and have been reported to use both socially and overtly aggressive strategies to harm peers (Putallaz et al., 2007).

All girls who engage in aggression are at some degree of risk for maladjustment. However, we suspect that for girls who have been socialized to adopt traditional standards of femininity, the use of overt aggression may lead to even more negative consequences than it would for those girls who have not been socialized to adopt those same standards. Unfortunately, much of the literature on the social and psychological adjustments of aggressive girls in the United States has focused on predominately European American populations, providing limited insight into how aggression might affect the lives of girls from different cultures and ethnic backgrounds, including African American girls. When studies have examined the adjustment patterns of aggressive girls from ethnically diverse populations, findings suggest that the relation between aggressive behavior and adjustment difficulties might be moderated by ethnicity as well as the form of aggression exhibited, with one exception (Putallaz et al., 2007).

Whereas some research suggests that socially and overtly aggressive African American girls are at less risk for maladjustment than socially and overtly aggressive European American girls (Kistner, Meltzer, Gatlin, & Risi, 1993; Kupersmidt & Coie, 1990; Xie et al., 2003), at least one study did not find evidence for ethnic differences in socially aggressive girls’ adjustment patterns (Putallaz et al., 2007). Given the inconsistent findings surrounding the relation between social aggression and overt aggression and the adjustment of preadolescent African American girls, the purpose of the present study was to examine ethnic variation in the adjustment patterns of preadolescent girls exhibiting social and overt forms of aggression. By understanding how different forms of aggression differentially influence the adjustment patterns of preadolescent girls from differing ethnicities, existing aggression interventions can be adapted to be more culturally relevant and to better meet the needs of aggressive African American girls.

Social Aggression and Adjustment

Girls who exhibit social aggression experience differing levels of severity of social adjustment difficulties than girls who aggress overtly (Crick, 1995). Studies examining aggression in girls in the latter half of elementary school have consistently found that social aggression is associated with elevated levels of perceived popularity (Lease, Kennedy, & Axelrod, 2002). Despite socially aggressive girls’ social centrality and the perception by peers that they often are prominent and prestigious individuals within the social network, socially aggressive girls are disliked more than their nonaggressive peers (Crick, 1995; Rys & Bear, 1997) and may experience significant levels of emotional and social distress as a result (Crick, 1995). Crick and colleagues have consistently demonstrated that socially aggressive girls experience significant internalizing problems and endorse elevated levels of loneliness, depression, and social anxiety (Crick, 1995, 1997; Crick et al., 2006; Crick & Grotpeter, 1995).

Although a number of studies have documented that socially aggressive girls are at risk for psychological maladjustment, there are a number of investigations that suggest the contrary; that is, some studies have reported that socially aggressive girls are highly competent and well adapted. Putallaz et al. (2007) did not find an association between social aggression and teacher ratings of depression or girls’ self-reported ratings of loneliness or depression. Moreover, Xie, Cairns, and Cairns (2002) reported that preadolescent girls who exhibited social aggression were able to form and maintain positive interpersonal relationships. Within a similar vein, Pulkkinen (1992) found that socially aggressive girls were achievement oriented, pursuing postsecondary education following graduation.

Overt Aggression and Adjustment

In contrast to social aggression, the adjustment outcomes for preadolescent girls who exhibit overt aggression are far less heterogeneous. Overtly aggressive girls are perceived by teachers and peers alike as more socially maladjusted than girls demonstrating social aggression. Investigations have shown that overtly aggressive girls are more likely to be rejected (i.e., disliked) by peers than nonaggressive girls and girls exhibiting social aggression (Crick, 1997; Putallaz et al., 2007). Outside of heightening interpersonal difficulties, overtly aggressive girls are at increased risk for experiencing serious psychological adjustment difficulties and academic difficulties. Girls who exhibit overt aggression endorse elevated levels of depression, anxiety, and low self-esteem (Crick, 1997; Prinstein, Boergers, & Vernberg, 2001) and are perceived by teachers as being delinquent, disruptive, and having greater difficulty with self-control (Crick, 1997; Farmer, Estell, Bishop, O’Neal, & Cairns, 2003). Overtly aggressive girls also are more likely to be disengaged from school and to display lower levels of achievement than their same-gendered peers (Crick, 1997; Crick et al., 2006).

Ethnicity, Aggression, and Adjustment

In comparison with European American girls, African American girls are more likely to exhibit social and overt aggression (Centers for Disease Control and Prevention [CDC], 2006; Putallaz et al., 2007) and, as a result, might suffer significant negative social and emotional consequences. For example, African American girls are perceived by peers as exhibiting greater levels of overt and social aggression than European American girls (Putallaz et al., 2007). This in part might be due to racial bias influenced by the media’s stereotypical portrayal of African American females as being angry and hostile (Collins, 2004; West, 1995).

Despite the growing literature on the prevalence of aggression among African American girls, the literature on the social and psychological adjustments of aggressive African American girls, particularly socially aggressive African American girls, is scant and inconsistent. To date, only a handful of studies have explored the adjustment patterns of African American girls who exhibit social and overt aggression, and findings on whether aggressive African American girls’ risk for adjustment difficulties parallels that of European American girls have been mixed (Blake, Roberts, & Lease, 2002; Kistner et al., 1993). Whereas some research has shown that both socially and overtly aggressive African American girls experience serious social and psychological adjustment difficulties, other research suggests that socially and overtly aggressive African American girls are well liked and exhibit few emotional problems.

In a preliminary investigation examining the sequelae of aggression for preadolescent girls, Blake et al. (2002) found that neither socially or overtly aggressive African American girls reported experiencing symptoms of depression, loneliness, or decrements in self-esteem. Socially and overtly aggressive African American girls also did not express difficulty relating to peers and were not rejected by peers. Moreover, Xie et al. (2003) found that neither socially or overtly aggressive African American girls differed from nonaggressive African American girls with respect to teacher-rated popularity. In contrast to teacher reports of social standing, socially and overtly aggressive African American girls exhibited higher levels of peer-reported network centrality than nonaggressive girls. Results from the Xie et al. (2003) study suggest that peers perceive socially and overtly aggressive African American girls to be more socially prominent and influential members of the peer group than nonaggressive African American girls.

Alternatively, at least two studies have failed to document the positive adjustment of aggressive African American girls, with one study yielding mixed results regarding their adjustment patterns. In a longitudinal study of African American youth, Miller-Johnson, Moore, Underwood, and Coie (2005) found that overt aggression in elementary school was predictive of externalizing behavior and arrest outcomes in preadolescence and early adulthood for African American girls but not internalizing difficulties. Farmer et al. (2003) also found inconsistent results with respect to aggressive African American girls’ social and psychological adjustments. Although high-status African American girls who exhibited either social or overt aggression were not perceived by peers as evidencing internalizing difficulties, teachers viewed African American girls who exhibited overt aggression more negatively than they viewed socially aggressive and nonaggressive girls. That is, teachers rated overtly aggressive African American girls as having greater attention problems, more difficulties with behavioral inhibition (i.e., hyperactivity), and less involvement in extracurricular activities. Overall, these findings suggest that whereas overtly aggressive African American girl’s use of aggression does not affect their internal psychological functioning, it does appear to negatively affect their adjustment to school and to indicate difficulty with behavior regulation.

The extant literature suggests that the linkage between aggressive behavior and adjustment for girls is likely moderated by the form of aggression exhibited as well as ethnicity (Kistner et al., 1993; Kupersmidt & Coie, 1990; Xie et al., 2003). For African American girls, social aggression might not have the same negative impact on social and emotional functioning as it does for European American girls. However, the relation between adjustment and overt aggression for African American girls is less clear. Whereas some research suggests that overtly aggressive African American girls are psychologically and socially well adjusted, other studies depict African American girls who exhibit overt aggression as being just as socially unskilled, delinquent, and defiant as overtly aggressive European American girls. The reason underlying this difference is unclear and warrants further investigation.

Possible Explanations for Ethnic Differences in Aggressive Girls’ Adjustment

Ethnic differences in the relation between aggression and girls’ social, psychological, and academic functioning might be attributed to ethnic differences in girls’ adoption of gender norms. Some scholars have speculated that African American girls are more likely to be socialized to be androgynous (Leaper, 2000; Lewis, 1975) and, therefore, embody both masculine and feminine characteristics than European American girls (Buckley & Carter, 2005; Harris, 1996). As a result, aggressive African American girls might be less likely to perceive overt aggression as a gender atypical form of aggression, and social aggression as gender typical, and might elect to engage in both forms of aggression when choosing to aggress as a way to dominate or harm peers (Belgrave, 2009). Moreover, overtly aggressive African American girls might not experience negative psychological and social consequences for their engagement in overt aggression if their same-race peers do not view their displays of overt aggression as deviating from gender norms. Unfortunately, the paucity of research examining aggression in African American females inhibits further conjecture and warrants further study.

Current Study

In an effort to contribute to the scant literature on aggression in African American girls, the purpose of this study was to explore how different forms of aggression differentially affect the psychological, social, and academic adjustments of African American girls, in comparison with European American girls. Specifically, we were interested in examining whether girls displaying different forms of aggression (e.g., social vs. overt) differed with respect to their social adjustment (i.e., perceived popularity and social preference as rated by peers). We were also interested in examining whether socially and overtly aggressive girls’ ratings of self-reported loneliness, school engagement, and experience of internalizing symptoms differed and if this relationship was moderated by ethnicity. Thus, we had two research questions of interest. First, does the form of aggression girls’ use differentially influence their social, psychological, and school adjustments? Second, does the pattern of associations between form of aggression and social, psychological, and school adjustments differ by ethnicity?

Whereas a number of studies have used teachers’ ratings to assess girls’ aggressive behavior and social competence (Crick, 1996), we elected to use peer nominations to assess girls’ use of aggression. Peer nominations of aggression are moderately correlated with teacher ratings of aggression (Crick, 1996). However, peer nominations provide collective information about the salience of specific behaviors to group dynamics and children’s social relationships. Moreover, peers provide a unique perspective into the social world of children that adult observers might not be able to detect or be privy to (McEvoy, Estrem, Rodriguez, & Olson, 2003). Peer nominations are also not as subject to gender bias compared with teacher ratings of aggression (Archer & Coyne, 2005).

Although it is conceivable that girls could exhibit mixed (i.e., social and overt) aggression, research on aggression in girls has focused on girls exhibiting either social or overt forms of aggression. This, in turn, has influenced many researchers to perceive females as aggressing in one of two ways: overtly or socially. However, an increasing number of studies have found that some girls use mixed aggressive strategies to aggress against others, suggesting that multiple types of aggressive girls might exist, with differential impacts on social and psychological adjustments (Crick et al., 2006; Talbott, Celinska, Simpson, & Coe, 2002; Xie, Swift, Cairns, & Cairns, 2002). Thus, we used cluster analyses to identify patterns of aggressive behavior among girls that are found to occur naturally in the school environment.

Method

Participants

Participants in the current study were drawn from a larger study on children’s peer relations. Peer- and self-report data were collected from 473 students from 26 fourth- and fifth-grade classrooms, located in six rural elementary schools in five counties in the southeastern United States. The consent rate for child participation in the larger study (parental consent and child assent were required for participation) was 88% across classrooms. Given that the focus of this study was girls, the subsample selected for this study consisted of 244 elementary school girls, with participants ranging in age from 9 to 11 years. Free and reduced lunch eligibility averaged 70% across schools. According to school records, 58% of the sample identified themselves as “White (European American)” students, 40% as “African American/Black,” and 2% as being from other racial/ethnic backgrounds (e.g., “Asian,” “Hispanic/Latino,” “American Indian/Alaska Native,” “Native Hawaiian/Pacific Islander,” and “other”). Students from “other” racial/ethnic backgrounds were excluded from subsequent analyses. Thus, the final sample consisted of 240 girls (n = 98 African American; n = 142 European American).

School records indicated that three of the schools from which students were sampled were majority–African American (School 1: 98%, School 2: 97%, and School 3: 82% African American), two were majority–European American (School 1: 74% and School 2: 87% European American), and one was racially mixed (40% African American and 55% European American). The racial/ethnic composition of the classrooms within each school closely approximated the schools’ overall composition and the demographic composition of counties. Overall, 82.5% of girls were in classrooms in which their racial/ethnic group was the numerical majority. Socioeconomic data were not collected on participating students. However, U.S. census data revealed that the counties in which participating schools were located were equivalent on wealth and educational status (U.S. Census Bureau, 2006). The mean median household income was $28,574 and the mean percentages of high school and college graduates (e.g., BS or higher) were 66.6% and 11%, respectively, across counties.

Procedure

Child participants from the larger study were asked to complete a series of behavioral, sociometric, and perceived popularity nominations using a modified version of the Revised Class Play (Lease & Blake, 2005; Masten, Morison, & Pellegrini, 1985). The instructions were as follows:

Pretend you are selecting roles for a class play. Please choose the student who best fits the role description listed below. We would like for you to nominate up to three children who fit each description as listed below. First, find the person you’d like to nominate on the list of names we gave you. Then write down the number beside their name by the role that best describes that person.

Student questionnaires were group administered in two 1-hour sessions, with written items read aloud to students by a member of the research team. A second data collection proctor was available to assist students who had difficulty completing questionnaire items. All students, both participating and nonparticipating, were given a token gift for their involvement in the study.

Measures

Social status nominations

To obtain an index of children’s social status within their peer group, social preference (i.e., like-most and like-least; Coie, Dodge, & Coppotelli, 1982), and perceived popularity (i.e., most-popular and least-popular; Lease et al., 2002) nominations were collected from children. Participants viewed a cross-gender roster with the names of all their participating classmates. Male and female students nominated three peers as liked-most, three peers as liked-least, three peers who are most-popular, and three peers who are least-popular. Numbers of most-liked, least-liked, most-popular, and least-popular received by each participant were summed and standardized, within classroom and gender, to a mean (M) of 0 and a standard deviation (SD) of 1. Social preference scores were calculated by subtracting standardized like-least scores from standardized like-most scores. A parallel approach was also adopted to calculate perceived popular scores, in which standardized least-popular scores were subtracted from standardized most-popular scores.

Peer nominations of aggression

Peer nominations of aggression, frequently used in studies of aggression, were collected in order to identify children who were perceived as aggressive by peers (Crick et al., 2006). Each child participating in the larger study was requested to nominate up to three peers, male or female, on a series of items that described physical (“tries to get what he or she wants by hitting, shoving, pushing, or threatening others”), verbal (“says mean things to people, calls names, and teases others in a mean way”), nonverbal social (“this person makes mean faces at someone when they are upset with them; this person acts bored and sighs often when someone they don’t like is talking”), and verbal social aggression (“this person tries to keep certain people from being in their group during activities; some children tell others that they will stop liking them unless the friends do what they say”). Children were instructed that they could nominate a child for more than one item. Nominations for each item were summed and standardized within gender and classroom to an M of 0 and an SD of 1, for consistency with the peer-nominated social status items.

Given the lack of prior research on social nonverbal aggression items, exploratory factor analyses with maximum likelihood extraction and varimax rotation were analyzed to determine the underlying dimensionality of peer nominations of aggression items. The exploratory factor analysis yielded two factors: an overt aggression factor and a social aggression factor. The two physical and verbal aggression items were found to load highly on the overt aggression factor: physical (λ = .89) and verbal aggression (λ = .66). Verbal and nonverbal social aggression items also loaded moderately to highly on the social aggression factor, with factor loadings ranging from .56 to .80. Based on exploratory factor analyses, the overt aggression variable was created from the mean of physical and direct verbal aggression items, and the social aggression variable was derived from the mean of relational and nonverbal aggression items. Internal consistency analyses conducted with the current sample suggest that the overt (α = .88) and social aggression (α = .92) variables evidence adequate reliability.

Self-reported adjustment

The depression scale and school and clinical maladjustment composites scales from the Behavior Assessment System for Children–Self-Report of Personality-Child (BASC-SRP; Reynolds & Kamphaus, 1992) were used to measure children’s self-reported school and psychological adjustments. The BASC-SRP is a 152-item true-false scale that yields clinical and adaptive measures of adjustment. Composite and scale scores are reported in T scores with an M of 50 and an SD of 10. The BASC-SRP was normed on an ethnically diverse sample of school-age children from typically developing and clinical populations. The technical manual reports that the general normative sample closely represented the ethnic diversity of the U.S. population following weighting procedures for 1990 (Reynolds & Kamphaus, 1992).

The school and clinical maladjustment composites are composed of clinical scales that assess school relatedness and disengagement (i.e., attitude toward teachers and attitude toward school scales) and symptoms of psychological maladjustment (e.g., social stress, anxiety, locus of control, etc.). The depression scale provides a measure of whether a child evidences depressive symptomatology. The school and clinical maladjustment composites and the depression scale are reported to demonstrate adequate reliability in the BASC manual (Reynolds & Kamphaus, 1992). The technical manual indicates that the reliabilities for the school maladjustment composite (α_{general normative} = .85 and α_clinical = .87), clinical maladjustment composite (α_{general normative} = .95 and α_clinical = .95), and depression scale (α_{general normative} = .88 and α_clinical = .89) were adequate. For the current sample, the coefficient αs for the clinical and school maladjustment composites and depression scale were .94, .64, and .92, respectively. A more comprehensive description of the psychometric properties of individual scales and composites as well as the factor analytic structure of the BASC-SRP is available for review in the technical manual (Reynolds & Kamphaus, 1992).

The Peer Network and Dyadic Loneliness Scale (PNDLS; Hoza, Bukowski, & Beery, 2000) was used to assess the degree of loneliness that children experience within their friendships. A 4-point Likert-type scale (never true to really true) was used to measure the degree to which children experienced satisfaction in their dyadic and network relationships with peers. Higher ratings on the PNDLS reflect greater levels of dissatisfaction and loneliness in dyadic relationships. Although the PNDLS yields two scales, only the dyadic scale was used in this study to assess children’s perception of the quality of their close relationships with peers. The PDNLS is a research-based instrument developed with a sample of predominately European American children from the Northeastern United States. The PNDLS is reported to demonstrate adequate reliability (dyadic loneliness: α = .84) and concurrent validity with well-established measures of loneliness and friendship quality (Hoza et al., 2000). Internal consistency analyses also indicate that the PNDLS was a reliable measure for the current sample (dyadic loneliness: α = .77).

Results

To examine whether the adjustment of aggressive girls was moderated by form of aggression and ethnicity, two sets of analyses were conducted. First, a cluster analysis with two peer-rated aggression input variables was conducted to identify types of aggressive and nonaggressive girls. We used cluster analyses as opposed to cutoff scores to form aggressive groups in order to identify naturally occurring groups of aggressive girls that are found to exist in classroom settings and to preserve statistical power, which is often lost when artificial cutoffs are used for group formation (Giang & Graham, 2008). Second, clusters of aggressive and nonaggressive girls obtained from the cluster analysis were compared with a series of 4 × 2 (Cluster × Ethnicity) univariate analyses of variance (ANOVAs) to determine if adjustment outcomes differed by form of aggression and ethnicity.

Cluster Analysis

A two-step cluster analytic procedure was performed using the Ward and K means procedures. Initial cluster solutions were identified with the Ward method and reassigned with the K means procedure, an iterative partitioning procedure that controls for poor initial clustering solutions (Gore, 2000; Huberty, DiStefano, & Kamphaus, 1997). Individuals were assigned to clusters so that the squared Euclidean distance was minimized for both procedures (DiStefano, Kamphaus, Horne, & Winsor, 2003). Three, four-, and five-cluster solutions were examined and evaluated for final selection based on the following four criteria: (a) pseudo-F value, with larger values suggesting better fitting models (SAS, 1999); (b) the interpretability of the cluster mean values; (c) reliability of cluster solutions; and (d) researcher judgment.

The three-cluster solution yielded the largest pseudo-F value, 444.89, but examination of cluster means indicated that the three clusters were grouped by level of aggression rather than form of aggression. The pseudo-F values for the four-cluster and five-cluster solutions were 415.5 and 399.86, respectively. Examination of cluster means for the four- and five-cluster solutions indicated that the four-cluster solution was most consistent with the extant literature on female aggression. The five-cluster solution closely resembled the four-cluster solution but with two clusters of nonaggressive girls.

The reliability of each cluster solution was tested with a split-half procedure that divided the sample into halves and reanalyzed each sample half three times with the Ward and K means procedures. The frequency of times that the specified cluster solution reemerged was tallied and used as an indicator of cluster reliability. Whereas the three- and four-cluster solutions were replicated four out of the six trials, the five-cluster solution was replicated only once.

The four-cluster solution was judged the most appropriate solution based on theoretical and substantive reasons, cluster means, and cluster reliability. Cluster means were used to interpret and label the four-cluster solution: Cluster 1 was labeled as “nonaggressive,” due to below-average (i.e., cluster mean < 0) social and overt aggression peer-nomination scores; Cluster 2 was labeled as “socially aggressive” due to elevated peer nominations of social aggression (cluster mean >.25) in comparison with overt aggression (cluster mean = −.09); Cluster 3 was labeled as moderate mixed aggression based on above-average nominations of overt aggression (cluster mean >1) and average nominations of social aggression (cluster mean >.5); and Cluster 4 was labeled as “high mixed aggressive” based on above-average (i.e., cluster mean >1) peer nominations of social and overt aggression.

To further test the distinctiveness of the clusters, two one-way ANOVAs with cluster as the independent variable and social and overt aggression as the dependent variables were conducted. Main effects for cluster on social aggression, F(3, 236) = 273.08, p < .001, and overt aggression, F(3, 236) = 272.95, p < .001, were significant. Subsequent pairwise comparisons, which were significant at the .01 level, indicated that the clusters were appropriately labeled. With respect to social aggression, girls who were in the high mixed aggressive cluster displayed the highest level of social aggression in comparison with girls in all other clusters. Specifically, girls in the nonaggressive cluster displayed the lowest level of social aggression relative to girls in the social aggression cluster, moderate mixed aggressive cluster, and high mixed aggressive cluster. Girls in the social aggression cluster exhibited lower levels of social aggression than girls in the moderate and high mixed aggressive clusters and girls in the moderate aggressive cluster displayed lower levels of social aggression than girls in the high mixed aggression cluster. A similar pattern emerged for overt aggression, with girls in the high mixed aggressive cluster exhibiting the greatest levels of overt aggression than girls in all other clusters and girls in the nonaggressive cluster displaying the lowest level of overt aggression relative to girls in other clusters. Table 1 presents the mean aggression variables for each cluster by ethnicity.

Table 1.

Sample Size and Mean of Aggression Variables for Clusters by Ethnicity

		Total		EA		AA
Aggression Clusters	N (EA/AA)	Social	Overt	Social	Overt	Social	Overt
Cluster 1: Nonaggressive group	127 (91/36)	−0.38_a	−0.59_a	−0.42	−0.60	−0.28	−0.57
Cluster 2: Socially aggressive group	55 (32/23)	0.47_b	−0.21_b	0.43	−0.26	0.52	−0.13
Cluster 3: Moderate mixed aggressive group	40 (12/28)	0.93_c	0.71_c	1.17	0.75	0.83	0.69
Cluster 4: High mixed aggressive group	18 (7/11)	2.13_d	1.72_d	2.26	1.46	2.05	1.89
Total	240 (142/98)

Note: EA = European American; AA = African American. All aggression variables are standardized within gender and classroom to a mean of 0 and standard deviation of 1. Within-column means with the same subscript do not differ significantly at p < .01.

Given prior research, which suggests that African American girls are perceived as more aggressive than their same-gendered peers of differing ethnicities, a chi-square analysis was conducted to assess whether race/ethnicity deviated from expected across the four aggressive groups. A significant chi-square for race/ethnicity, χ²(3) = 25.36, p = .000, was found. An examination of the standardized residuals from cross-tab analyses indicated that African American girls were underrepresented in the nonaggressive group but were overrepresented in the moderate mixed-aggression group relative to European American girls. A contingency table is presented in Table 2.

Table 2.

Contingency Table of Cluster Membership by Ethnicity

Ethnicity	Nonaggressive	Social	Moderate Mixed	High Mixed	Total
EA	91 (1.8)	32 (−0.1)	12 (−2.4)	7 (−1.1)	142
AA	36 (−2.2)	23 (0.1)	28 (2.9)	11 (1.3)	98
Total	127	55	40	18	240

Note: EA = European American; AA = African American. Standardized residuals have a mean of 0 and standard deviation is 1.0; standardized residuals are listed in parentheses.

Social Adjustment

Descriptive statistics for study variables by cluster are presented in Table 3. To determine if adjustment outcomes differed by form of aggression and ethnicity, a series of ANOVAs were conducted (4 × 2: Cluster × Ethnicity) for each outcome variable. The interaction effects for social preference and perceived popularity were nonsignificant. No main effects for ethnicity were found. However, significant main effects for cluster were found for the social preference, F(3, 232) = 9.16, p = .001, and perceived popularity, F(3, 232) = 4.03, p = .008, variables. Subsequent pairwise comparisons were significant at the .01 level and revealed that girls who exhibited moderate and high levels of mixed aggression were liked less by their peers than nonaggressive girls, d = .60 and d = .20, respectively. Clusters with girls exhibiting moderate and high levels of mixed aggression were also liked less than girls who engaged only in social aggression, d = .53 and d = .18, respectively. However, no differences were found in likeability between socially aggressive and nonaggressive girls. With respect to popularity, socially aggressive and high mixed aggressive girls were perceived as more popular than nonaggressive girls, d = −.53 and d = −.31, respectively. No differences in popularity were found between socially aggressive or moderately mixed aggressive girls or between moderately and high mixed aggressive girls.

Table 3.

Descriptive Statistics for Study Variables by Cluster

		Nonaggressive (n = 127)		Social (n = 55)		Moderate Mixed (n = 40)		High Mixed (n = 18)
	Ethnicity	M	SD	M	SD	M	SD	M	SD
Social preference	EA	0.18	0.96	0.00	1.01	−0.36	0.91	−1.57	1.40
	AA	0.20	1.03	0.28	0.83	−0.40	0.77	−0.35	1.01
	Total	0.19	0.97	0.12	0.94	−0.39	0.81	−0.82	1.29
Perceived popularity	EA	−0.23	1.02	0.16	0.94	−0.01	0.85	0.13	1.48
	AA	−0.15	0.98	0.47	0.76	0.15	0.88	0.65	1.22
	Total	−0.21	1.00	0.29	0.87	0.10	0.86	0.45	1.31
Depression	EA	51.36	11.98	49.30	10.71	50.42	12.85	60.29	9.48
	AA	54.24	10.71	55.07	10.21	53.91	10.76	55.91	11.76
	Total	52.17	11.66	51.71	10.80	52.86	11.37	57.61	10.86
Clinical maladjustment	EA	50.04	10.69	49.51	10.23	49.67	11.87	59.43	2.99
	AA	52.20	10.40	52.89	9.80	54.66	8.41	51.91	11.08
	Total	50.65	10.62	50.93	10.10	53.16	9.70	54.83	9.46
Dyadic dissatisfaction	EA	1.55	0.67	1.37	0.50	1.92	1.04	2.16	0.85
	AA	1.98	0.76	1.83	0.76	2.01	0.79	1.73	0.70
	Total	1.68	0.72	1.56	0.66	1.98	0.86	1.90	0.77
School maladjustment	EA	47.41	8.02	49.59	8.96	51.92	9.70	64.29	8.73
	AA	53.64	9.92	55.34	10.57	58.56	12.86	54.45	11.90
	Total	49.18	9.01	52.00	9.99	56.57	12.27	58.28	11.60

Note: EA = European American; AA = African American. Perceived popularity and social preference variables are standardized within gender and classroom to a mean (M) of 0 and standard deviation (SD) of 1. Dyadic dissatisfaction is scored on a 4-point Likert-type scale (ranging from 1 = never true to 4 = really true); Behavior Assessment System for Children–Self-Report of Personality-Child (BASC-SRP) depression, clinical maladjustment, and school maladjustment are reported in T scores with a mean of 50 and standard deviation of 10.

Psychological Adjustment

A significant interaction effect was found for girls’ self-reported dissatisfaction with their dyadic relationships with peers, F(3, 232) = 3.77, p = .001. Simple effect analyses revealed that socially aggressive African American girls, t(53) = −2.74, p = .008, d = −.72, endorsed greater feelings of loneliness and dissatisfaction in their friendships than socially aggressive European American girls. A similar finding was found for nonaggressive girls as well, with nonaggressive African American girls experiencing greater dissatisfaction in their friendships than nonaggressive European American girls, t(125) = 3.77, p = .001, d = −.60. Interaction and main effects for depression and clinical maladjustment were not significant.

School Adjustment

A significant interaction effect for the school maladjustment composite was found, F(3, 232) = 3.62, p = .014, indicating ethnic differences in aggressive and nonaggressive girls’ perceptions of school. Simple effect analyses revealed that nonaggressive African American girls reported greater levels of school disengagement than nonaggressive European American girls, t(125) = −3.68, p = .000, d = −.65. African American girls who engaged in social aggression also endorsed having more negative attitudes toward school and their teachers relative to socially aggressive European American girls, t(53) = −2.17, p = .034, d = −.59. A marginally significant relation between ethnicity and high levels of mixed aggression was also found, t(15.54) = 2.01, p = .061, d = .94. European American girls who exhibited high levels of mixed aggression reported greater disengagement from schools than high mixed aggressive African American girls.

Discussion

The purpose of the current research was to examine variation in the adjustment patterns of girls who exhibit social and overt forms of aggression and to explore whether social and overt aggression differentially influences the adjustment of African American and European American girls from rural communities. Ultimately, our goal was to advance knowledge of aggressive African American girls and to assist in the development of culturally centered aggression interventions for preadolescent girls. Intervening with girls at this developmental stage might serve to prevent an escalation of externalizing behavior problems and other problem outcomes reported for aggressive adolescent African American females.

Our findings suggest that preadolescent girls aggress in multiple ways, using social aggression as well as a mixture of social and overt aggression to harm peers. Consistent with prior research, we found that African American girls were perceived as more aggressive than European American girls (CDC, 2006; Putallaz et al., 2007). Specifically, African American girls were perceived by peers as more likely to exhibit moderate levels of mixed aggression than European American girls; yet no ethnic differences were found for social aggression. Our findings suggest that African American girls are more likely to be bi-strategic aggressors, using both socially and overtly aggressive strategies to aggress against peers than European American girls—a finding that has been noted, but not fully explored, in other studies as well (Putallaz et al., 2007; Talbott et al., 2002; Xie, Cairns, & Cairns, 2002). We suspect that African American girls may be more likely to engage in both social and overt aggression due to gender socialization practices within the African American community, which encourage androgyny in females (Belgrave, 2009); however, additional research is necessary to support this conjecture.

We found some support that the form of aggression exhibited differentially influences girls’ social adjustment but not girls’ psychological adjustment. However, this relation did not vary by ethnicity. Whereas nonaggressive and socially aggressive girls were liked more than girls who displayed moderate and high levels of mixed aggression, socially aggressive girls did not differ from nonaggressive girls with respect to social preference or peer likeability. A different pattern of findings emerged when aggressive girls’ levels of popularity was examined. Girls who were socially aggressive and who displayed high levels of mixed aggression were perceived as more popular than nonaggressive girls. Interestingly, the popularity of socially aggressive and moderately mixed aggressive girls and the popularity of moderately mixed aggressive girls and girls who exhibited high levels of mixed aggression did not differ in our study.

Our findings seem to contradict the work of prior research, which suggests that social aggression is predictive of social and psychological maladjustments, but instead lend support to the adaptive value hypothesis of social aggression (Pulkkinen, 1992; Xie, Cairns, & Cairns, 2002). Recent findings suggest that social centrality and elevated social status may buffer the negative effects (i.e., psychological maladjustment) of social aggression (Rose & Swenson, 2009). Thus, socially aggressive girls might fail to experience psychological maladjustment because of their greater social centrality and Machiavellian nature (Hawley, 2003). The fact that socially aggressive girls in our study were neither unpopular nor rejected by peers implies that socially aggressive girls evidence some level of social skill. Xie et al. (2005) reported that in order for socially aggressive girls to be successful in their aggressive bidding they must be part of an aggressive network, that is, girls cannot be socially aggressive in the absence of social relationships. Therefore, it is not surprising that socially aggressive girls in this study were more popular than nonaggressive peers and as well liked, since skill in relationship building is necessary for orchestrating manipulation ranks.

However, it is important to note that the adaptive value of social aggression might not translate across cultures, as socially aggressive African American girls reported greater feelings of loneliness within their friendships than socially aggressive European American girls. Both socially aggressive and nonaggressive African American girls indicated experiencing greater loneliness and dissatisfaction in their friendships than European American girls who were socially aggressive or who did not engage in aggression. Although nonaggressive African American girls reported greater levels of dissatisfaction and loneliness in their friendships in comparison with European American girls, the magnitude in the difference between African American and European American girls’ ratings for dyadic loneliness was far greater for socially aggressive African American girls than nonaggressive African American girls. Socially aggressive African American girls might feel less cared about and supported in their close relationships with peers because they embody characteristics that are not traditionally valued in the African American community and, therefore, by their same-race peers.

Kochman (1981) proposed that intense emotional expression is normative and consequently is perceived as a marker of sincerity and passion by African Americans. If sincere emotional expression is the typical method of communication in the African American community in which African Americans communicate their feelings openly and directly to others, then it could be argued that African American children would be less trusting of same-race peers who fail to express their feelings openly and directly but, instead, use more covert means such as gossiping and social exclusion to express anger or to wield power. Thus, it is plausible that socially aggressive African American girls’ dissatisfaction with their close relationships with peers might be explained by their failure to adhere to cultural norms surrounding emotional expression.

Relative to socially aggressive European American girls, socially aggressive African American girls reported greater levels of school disengagement. This trend was observed for nonaggressive African American girls as well. It is not clear why nonaggressive and socially aggressive African American girls reported more negative attitudes toward their teachers and school than nonaggressive and socially aggressive European American girls. Much of the research on academic achievement and engagement of African American youth has focused on African American males, with limited attention directed toward African American females. This could be attributed to African American females’ greater levels of engagement and achievement relative to African American males and also to the perception of African American girls as demonstrating lower risk for behavioral problems (Blake, Butler, Lewis, & Darensbourg, 2011; Graham, Taylor, & Hudley, 1998). However, when African American girls’ are directly compared with European American girls, research suggests that African American girls are more likely to disidentify with school (Griffin, 2002). In alignment with this finding, our results suggest that African American girls may be at risk for school disengagement regardless of their display of aggressive behavior and that more research is needed to explore why this trend might occur.

We did not find evidence to support previous findings that mixed aggression is associated with greater maladjustment than social aggression in girls (Crick et al., 2006). Although moderately mixed aggressive girls were less liked by their peers than nonaggressive and socially aggressive girls, girls who engaged in mixed aggression at moderate levels did not significantly differ from socially aggressive girls with respect to popularity or in their self-reported ratings of internalizing symptoms. Moreover, girls who exhibited high levels of mixed aggression did not report experiencing emotional problems along these domains, although this cluster was quite small. These findings imply that when coupled with social aggression, moderate to high levels of overt aggression may not be as detrimental to girls’ psychological adjustment—at least in the context of relatively economically disadvantaged, rural schools—as indicated in prior studies. A marginally significant effect for ethnic differences in high mixed aggressive girls’ school adjustment was found but due to the small sample size of this group we encourage caution in the interpretation of this finding.

Study Limitations and Future Directions

Although this study has notable strengths, it has some limitations that warrant discussion. One limitation is the shared method variance of using a single rater—peers—for measures of both aggression and social status as well as our sole use of self-report ratings to assess girls’ psychological and school adjustments. Peer nominations have been identified as a reliable measure of child behavior that reduces gender bias when used to assess aggression (Archer & Coyne, 2005; Merrell, 2003). However, like all procedures, peer nominations are not without measurement flaws and, therefore, are subject to both racial and reputational bias (Jackson, Barth, Powell, & Lochman, 2006). Thus, it is possible that African American girls’ elevated aggression reflects peer racial bias rather than their actual engagement in aggressive acts (Collins, 2004; West, 1995). Research suggests that when African American children are in classrooms in which their racial/ethnic group is the numerical minority, they are more likely to receive more negative sociometric ratings and peer nominations (e.g., liked-least, aggression, etc.) than European American children (Jackson et al., 2006). However, given that most African American girls were in classrooms in which their racial/ethnic group was the numerical majority in this study, it is less plausible that their elevated aggression ratings were due to racial bias.

Alternatively, girls’ aggression ratings in general might be explained by reputational bias since the sample was drawn from rural communities. Relative to urban settings, rural communities tend to be smaller and have more closely connected social networks (in which students and their families are well known and the same students are placed in the same classes year after year; Beggs, Haines, & Hurlbert, 1996). This increased familiarity may allow children to draw from their past perceptions of how children have behaved in previous situations when using peer nominations to rate their classmates’ behavior. On the other hand, since children from rural communities are more familiar with their peers they may have a better understanding of how children behave normally. In future studies, investigators may wish to consider corroborating peer nomination items of aggression with alternative measures of aggression to reduce rater bias (e.g., observational data and self-report ratings). Children are deemed to be the best raters of their internal psychological functioning, but teachers may provide a better assessment of children’s academic functioning in school (Frick, Barry, & Kamphaus, 2010). It is possible that our findings surrounding school maladjustment may be due to rater effects. Therefore, we recommend that future studies use teacher ratings to assess the academic functioning and school engagement of aggressive girls.

A second limitation of the present study was our inability to control for aggression severity. Girls who exhibited mixed aggression were perceived by peers as exhibiting higher levels of aggression than socially aggressive girls. Based on these findings, it is difficult to attribute differences in aggressive girls’ adjustment patterns solely to the form of aggression they displayed: social versus a mixture of social and overt aggression. However, it is noteworthy that there were relatively few differences in the adjustment patterns of socially aggressive girls who exhibited low levels of aggression in comparison with girls who displayed moderate levels of mixed aggression.

Our findings contradict the work of prior studies on aggression in girls in many ways. Specifically, our inability to identify psychological maladjustment in aggressive girls may be attributed to the insensitivity of our psychological measure to detect psychological maladjustment in subclinical populations. Alternatively, it is possible that community context might explain our discrepant findings. Research suggests that children residing in rural communities are at less risk for experiencing aggression and externalizing behavior problems than children residing in urban communities (Hope & Bierman, 1998). Although socially aggressive and mixed aggressive girls exhibited elevated levels of aggression relative to their peers, perhaps their level of aggression did not meet a threshold significant enough to cause psychological maladjustment. Replication of this study with a larger sample and with aggressive girls from both rural and urban communities will be important to support this hypothesis.

Despite these limitations, this study provides important information with respect to the social and psychological adjustments of aggressive African American and European American girls. The present study differs from prior research on female-based aggression in that we examined the social and psychological adjustments of African American and European American preadolescent girls from low-income and fairly racially homogeneous schools in rural communities. The comparison of African American and European American girls from similar communities provides information that is more pertinent to the role of ethnicity than to the role of social class or economic status in girls’ aggressive behavior and adjustment. Furthermore, the sampling of children from schools in which their ethnic group serves as the numerical majority provides us with greater insight into the influence of cultural norms on aggressive behavior.

Few studies have explored the effects of social aggression or the mixture of social and overt aggression on girls’ school engagement. It is not surprising that European American girls who exhibited high levels of mixed aggression displayed negative attitudes toward school and their teachers; however, it is interesting that similar elevations were not observed in the cluster of high mixed aggressive African American girls and yet were noted for socially aggressive African American girls. These findings imply that African American girls who aggress only socially might have differing types of adjustment patterns and skill sets than European American girls. Therefore, more research exploring the relationship between social and mixed aggression and school adjustment is needed.

One avenue for increasing aggressive girls’ school engagement, while simultaneously reducing their aggressive behavior, is to foster the development of positive student-teacher relationships. Whereas positive and supportive relationships with teachers are predictive of positive social and emotional behavior for all African American children (Baker, 2006), Meehan, Hughes, and Cavell (2003) found that aggressive African American children may be more receptive to teacher warmth and student support than aggressive European American youth. Although aggression interventions have called for the greater involvement of teachers in modeling and reinforcing intervention strategies to foster skill generalization (Olweus, 1995), few interventions stress the importance of positive student-teacher relationships in reducing aggression in youth. Therefore, we recommend that clinicians supplement existing interventions directed toward reducing aggression among African American girls with teacher training that focuses on fostering quality student-teacher relationships.

An additional strength of our study is our use of person-centered analysis (e.g., cluster analysis) as opposed to variable-centered analysis to identify naturally occurring groups of aggressive girls (von Eye & Bogat, 2006). In prior studies exploring aggression in girls, it has been assumed that girls are only capable of aggressing in one of two ways, either using verbal social aggression or physical aggression. Unfortunately, this assumption has limited our understanding surrounding the manner in which girls aggress. Our use of cluster analysis allowed us to examine the various ways in which girls aggress naturally in the school environment. Our findings suggest that African American girls are more likely than European American girls to be bi-strategic aggressors. Although interventions targeting aggression in girls attempt to address girls’ use of social aggression, our findings suggest that clinicians should focus intervention efforts on the function of African American girls’ aggression rather than the form of aggression displayed. By focusing on the reason bi-strategic aggressive African American females engage in aggression, we believe that clinicians may be more successful in reducing aggression in this population.

Footnotes

Portions of this research were based on the dissertation of the first author.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by grants awarded to the first author by the Georgia Association of School Psychologists and the University of Georgia Dean of Graduate School Social Sciences Award, and fellowships sponsored by the Elizabeth Munsterberg Koppitz Child Psychology Fellowship and Southern Regional Educational Board.

References

Archer

Coyne

S. M.

(2005). An integrated review of indirect, social, and social aggression. Personality and Social Psychology Review, 9, 212-230. doi:10.1207/s15327957pspr0903_210.1207/s15327957pspr0903_2

Baker

J. A.

(2006). Contributions of teacher-child relationships to positive school adjustment during elementary school. School Psychology Quarterly, 44, 211-229. doi:10.1016/j.jsp.2006.02.00210.1016/j.jsp.2006.02.002

Beggs

J. J.

Haines

V. A.

Hurlbert

J. S.

(1996). Revisiting the rural-urban contrast: Personal networks in nonmetropolitan and metropolitan settings. Rural Sociology, 61, 306-325.

Belgrave

F. Z.

(2009). African American girls: Reframing perceptions and changing experiences. New York, NY: Springer.

Blake

J. J.

Butler

B. A.

Lewis

C. L.

Darensbourg

(2011). Unmasking the inequitable discipline experiences of urban Black girls: Implications for urban stakeholders. Urban Review, 43, 90-106.

Blake

J. J.

Roberts

Lease

A. M.

(2002, March). Influences of aggression on peer perceptions and girls’ self-concept. Paper presented at the Conference of Human Development, Charleston, North Carolina.

Brown

L. M.

Gilligan

(1992). Meeting at the crossroads: Women’s psychology and girl development. Cambridge, MA: Harvard University Press.

Buckley

T. R.

Carter

R. T.

(2005). Black adolescent girls: Do gender role and racial identity impact their self-esteem? Sex Roles, 53, 647-661. doi:10.1007/s11199-005-7731-610.1007/s11199-005-7731-6

Centers for Disease Control and Prevention. (2006, June 9). Youth risk behavioral surveillance—Unites States 2005. Morbidity and Mortality Weekly Report: Surveillance Summaries (Vol. 55, No. S-55). Retrieved from http://www.cdc.gov/mmwr/indss_2006.html

10.

Coie

J. D.

Dodge

K. A.

Coppotelli

(1982). Dimensions and types of social status: A cross-age perspective. Developmental Psychology, 18, 557-570. doi:10.1037/0012-1649.18.4.55710.1037/0012-1649.18.4.557

11.

Collins

P. H.

(2004). Black feminist thought: Knowledge, consciousness, and the politics of empowerment (2nd ed.). New York, NY: Routledge.

12.

Crick

N. R.

(1995). Relational aggression: The role of intent attributions, feelings of distress, and provocation type. Development and Psychopathology, 7, 313-322. doi:10.1017/S095457940000652010.1017/S0954579400006520

13.

Crick

N. R.

(1996). The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children’s future social adjustment. Child Development, 67, 2317-2327. doi:10.1111/j.1467-8624.1996.tb01859.x10.1111/j.1467-8624.1996.tb01859.x

14.

Crick

N. R.

(1997). Engagement in gender normative versus nonnormative forms of aggression: Links to social-psychological adjustment. Developmental Psychology, 33, 610-617. doi:10.1037/0012-1649.33.4.61010.1037/0012-1649.33.4.610

15.

Crick

N. R.

Bigbee

M. A.

Howes

(1996). Gender differences in children’s normative beliefs about aggression: How do I hurt thee? Let me count the ways. Child Development, 67, 1003-1014. doi:10.1111/j.1467-8624.1996.tb01779.x10.1111/j.1467-8624.1996.tb01779.x

16.

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information processing mechanism in children’s social adjustment. Psychological Bulletin, 115, 74-101. doi:10.1037/0033-2909.115.1.7410.1037/0033-2909.115.1.74

17.

Crick

N. R.

Grotpeter

J. K.

(1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710-722. doi:10.1111/j.1467-8624.1995.tb00900.x10.1111/j.1467-8624.1995.tb00900.x

18.

Crick

N. R.

Ostrov

J. M.

Werner

N. E.

(2006). A longitudinal study of relational aggression, physical aggression, and children’s social-psychological adjustment. Journal of Abnormal Child Psychology, 34, 131-142. doi:10.1007/s10802-005-9009-410.1007/s10802-005-9009-4

19.

DiStefano

Kamphaus

R. W.

Horne

A. M.

Winsor

A. P.

(2003). Behavioral adjustment in the U.S. elementary school: Cross-validation of a person-oriented typology of risk. Journal of Psychoeducational Assessment, 21, 338-357. doi:10.1177/07342829030210040310.1177/073428290302100403

20.

Farmer

T. W.

Estell

D. B.

Bishop

J. L.

O’Neal

K. K.

Cairns

B. D.

(2003). Rejected bullies or popular leaders? The social relations of aggressive subtypes of rural African American early adolescents. Developmental Psychology, 39, 992-1004. doi:10.1037/0012-1649.39.6.99210.1037/0012-1649.39.6.992

21.

Frick

P. J.

Barry

C. T.

Kamphaus

R. W.

(2010). Clinical assessment of child and adolescent personality and behavior. New York, NY: Springer.

22.

Giang

M. T.

Graham

(2008). Using latent class analysis to identify aggressors and victims of peer harassment. Aggressive Behavior, 34, 203-213. doi:10.1002/ab.2023310.1002/ab.20233

23.

Giles

J. W.

Heyman

G. D.

(2005). Young children’s beliefs about the relationship between gender and aggressive behavior. Child Development, 73, 1143-1150. doi:10.1111/j.1467-8624.2005.00833.x10.1111/j.1467-8624.2005.00833.x

24.

Gore

P. A.

(2000). Cluster analysis. In Tinsley

H. E. A.

Brown

S. D.

(Eds.), Handbook of applied multivariate statistics and mathematical modeling (pp. 298-318). San Diego, CA: Academic Press.

25.

Graham

Taylor

A. Z.

Hudley

(1998). Exploring achievement values among ethnic minority early adolescents. Journal of Educational Psychology, 90, 606-620. doi:10.1037/0022-0663.90.4.60610.1037/0022-0663.90.4.606

26.

Griffin

B. W.

(2002). Academic disidentification, race and high school dropouts. High School Journal, 85, 71-81. doi:10.1353/hsj.2002.000810.1353/hsj.2002.0008

27.

Harris

A. C.

(1996). African American and Anglo-American gender identities: An empirical study. Journal of Black Psychology, 22, 182-194. doi:10.1177/0095798496022200410.1177/00957984960222004

28.

Hawley

P. H.

(1999). The ontogenesis of social dominance: A strategy-based evolutionary perspective. Developmental Review, 19, 97-132. doi:10.1006/drev.1998.047010.1006/drev.1998.0470

29.

Hawley

P. H.

(2003). Prosocial and coercive configurations of resource control in early adolescence: A case for the well-adapted Machiavellian. Merrill-Palmer Quarterly, 49, 279-309. doi:10.1353/mpq.2003.001310.1353/mpq.2003.0013

30.

Hope

T. L.

Bierman

K. L.

(1998). Patterns of home and school behavior problems in rural and urban settings. Journal of School Psychology, 36, 45-58. doi:10.1016/S0022-4405(97)00049-610.1016/S0022-4405(97)00049-6

31.

Hoza

Bukowski

W. M.

Beery

(2000). Assessing peer network and dyadic loneliness. Journal of Clinical Child Psychology, 29, 119-128. doi:10.1207/S15374424jccp2901_1210.1207/S15374424jccp2901_12

32.

Huberty

C. J.

DiStefano

Kamphaus

R. W.

(1997). Behavioral clustering of school children. Multivariate Behavioral Research, 32, 105-134. doi:10.1207/s15327906mbr3202_210.1207/s15327906mbr3202_2

33.

Jackson

M. F.

Barth

J. M.

Powell

Lochman

J. E.

(2006). Classroom contextual effects of race on children’s peer nominations. Child Development, 66, 1325-1337. doi:10.1111/j.1467-8624.2006.00937.x10.1111/j.1467-8624.2006.00937.x

34.

Kistner

Meltzer

Gatlin

Risi

(1993). Classroom racial proportions and children’s peer relations: Race and gender effects. Journal of Educational Psychology, 85, 446-452. doi:10.1037/0022-0663.85.3.44610.1037/0022-0663.85.3.446

35.

Kochman

(1981). Styles in conflict. Chicago, IL: University of Chicago Press.

36.

Kupersmidt

J. B.

Coie

J. D.

(1990). Preadolescent peer status, aggression, and school adjustment as predictors of externalizing problems in adolescence. Child Development, 61, 1350-1362. doi:10.1111/j.1467-8624.1990.tb02866.x10.1111/j.1467-8624.1990.tb02866.x

37.

Leaper

(2000). The social construction and socialization of gender during development. In Miller

P. H.

Scholnick

E. K.

(Eds.), Toward a feminist developmental psychology (pp. 127-152). New York, NY: Routledge.

38.

Lease

A. M.

Blake

J. J.

(2005). A comparison of majority-race children with and without a minority-race friend. Social Development, 14, 20-41. doi:10.1111/j.1467-9507.2005.00289.x10.1111/j.1467-9507.2005.00289.x

39.

Lease

A. M.

Kennedy

C. A.

Axelrod

J. L.

(2002). Children’s social constructions of popularity. Social Development, 11, 87-109. doi:10.1111/1467-9507.0018810.1111/1467-9507.00188

40.

Lewis

D. K.

(1975). The Black family: Socialization and sex roles. Phylon, 36, 221-237. doi:10.2307/27438810.2307/274388

41.

Masten

A. S.

Morison

Pellegrini

(1985). A revised class play method of peer assessment. Developmental Psychology, 21, 523-533.

42.

McEvoy

M. A.

Estrem

T. L.

Rodriguez

M. C.

Olson

M. L.

(2003). Assessing relational and physical aggression among preschool children: Intermethod agreement. Topics in Early Childhood Special Education, 23, 53-63. doi:10.1177/0271121403023002010110.1177/02711214030230020101

43.

Meehan

B. T.

Hughes

J. N.

Cavell

T. A.

(2003). Teacher-student relationships as compensatory resources for aggressive children. Child Development, 74, 1145-1157. doi:10.1111/1467-8624.0059810.1111/1467-8624.00598

44.

Merrell

K. W.

(2003). Assessment of social skills and peer relations. In Merrell

K. W.

(Ed.), Behavioral, social, and emotional assessment of children and adolescents (2nd ed., pp. 311-339). New York, NY: Routledge/Taylor & Francis.

45.

Miller-Johnson

Moore

B. L.

Underwood

M. K.

Coie

J. D.

(2005). African-American girls and physical aggression: Does stability of childhood aggression predict later negative outcomes? In Pepler

D. J.

Madsen

K. C.

(Eds.), Development and treatment of girlhood aggression (pp. 75-95). Mahwah, NJ: Lawrence Erlbaum.

46.

Olweus

(1995). Bullying or peer abuse at school: Facts and intervention. Current Directions in Psychological Science, 4, 196-200. doi:10.1111/1467-8721.ep1077264010.1111/1467-8721.ep10772640

47.

Ostrov

J. M.

Godleski

S. A.

(2010). Toward an integrated gender-linked model of aggression subtypes in early and middle childhood. Psychological Review, 117, 233-242. doi:10.1037/a001807010.1037/a0018070

48.

Prinstein

M. J.

Boergers

Vernberg

E. M.

(2001). Overt and relational aggression in adolescents: Social-psychological adjustment of aggressors and victims. Journal of Clinical Child Psychology, 30, 479-491. doi:10.1207/S15374424JCCP3004_0510.1207/S15374424JCCP3004_05

49.

Pulkkinen

(1992). The path to adulthood for aggressively inclined girls. In Björkqvist

K. N.

(Ed.), Of mice and women: Aspects of female aggression (pp. 113-121). San Diego, CA: Academic Press.

50.

Putallaz

Grimes

C. L.

Foster

K. J.

Kupersmidt

J. B.

Coie

J. D.

Dearing

(2007). Overt and relational aggression and victimization: Multiple perspectives within a school setting. Journal of School Psychology, 45, 523-547. doi:10.1016/j.jsp.2007.05.00310.1016/j.jsp.2007.05.003

51.

Reynolds

C. R.

Kamphaus

R. W.

(1992). Behavior Assessment System for Children. Circle Pines, MN: American Guidance Service.

52.

Rose

A. J.

Swenson

L. P.

(2009). Do perceived popular adolescents who aggress against others experience emotional adjustment problems themselves? Developmental Psychology, 45, 868-872. doi:10.1037/a001540810.1037/a0015408

53.

Rys

G. S.

Bear

G. G.

(1997). Relational aggression and peer relations: Gender and developmental issues. Merrill-Palmer Quarterly, 43, 87-106.

54.

SAS. (1999). SAS Online Doc (Version 8). Cary, NC: SAS Institute. Retrieved from http://v8doc.sas.com/sashtml/

55.

Talbott

Celinska

Simpson

Coe

M. G.

(2002). “Somebody else making somebody else fight”: Aggression and the social context among urban adolescent girls. Exceptionality, 10, 203-220. doi:10.1207/S15327035EX1003_410.1207/S15327035EX1003_4

56.

Underwood

M. K.

(2003). Social aggression among girls. New York, NY: Guilford Press.

57.

U.S. Census Bureau. (2006). American fact finder. Retrieved January 10, 2007, from http://www.factfinder.census.gov

58.

West

C. M.

(1995). Mammy, sapphire, and jezebel: Historical images of Black women and their implications for psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 32, 458-466. doi:10.1037/0033-3204.32.3.45810.1037/0033-3204.32.3.458

59.

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.

(Eds.), Development and treatment of girlhood aggression (pp. 105-136). Mahwah, NJ: Lawrence Erlbaum.

60.

Xie

Cairns

R. B.

Cairns

B. D.

(2002). The development of social aggression and physical aggression: A narrative analysis of interpersonal conflicts. Aggressive Behavior, 28, 341-355. doi:10.1002/ab.8000810.1002/ab.80008

61.

Xie

Farmer

T. W.

Cairns

B. D.

(2003). Different forms of aggression among inner-city African-American children: Gender, configurations, and school social networks. Journal of School Psychology, 41, 355-375. doi:10.1016/S0022-4405(03)00086-410.1016/S0022-4405(03)00086-4

62.

Xie

Swift

D. J.

Cairns

B. D.

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:10.1111/1467-9507.0019510.1111/1467-9507.00195

63.

von Eye

Bogat

G. A.

(2006). Person-oriented and variable-oriented research: Concepts, results, and development. Merrill-Palmer Quarterly, 52, 390-420. doi:10.1353/mpq.2006.003210.1353/mpq.2006.0032