Caught between stages: relational aggression emerging as a developmental advance in at-risk preschoolers.

Abstract. Eighty-two Head Start preschoolers were assessed with a peer rating measure of sociometric status, the Social Skills Rating System for Teachers (Gresham & Elliott, 1990), an Overt Aggression scale culled from items from the Aggressive Behavior subscale of the CBCL-TRF (Achenbach, 1997), and teacher ratings of relational aggression (Crick, Casas, & Mosher, 1997). In a regression model, overt aggression, age, social skills, and an interaction between social skills and overt aggression significantly predicted 27 percent of the variance in relational aggression. Being older, more overtly aggressive, and more socially skilled was predictive of increased relational aggression in this sample of Head Start students. The interaction indicated that at high and middle levels of social skills, relational and overt aggression significantly covaried, whereas at low levels of social skills this covariance was reduced to a trend. In other words, children with average or above average social skills were also significantly more aggressive--both relationally aggressive and overtly aggressive, which was not true for children with below average social skills. Interestingly, relational aggression appeared to be a developmental "advance" of sorts, developing commensurate with social skills, though not yet replacing the use of overt aggression. Of note for preschool educators, improvements in social skills may be associated with increases in indirect forms of aggression.

**********

From a developmental perspective, preschool aggression can be difficult to categorize. It is at once normative and potentially problematic. Reaching its peak in normative samples between the ages of 2 and 4, overt aggression exhibited at more stable and/or intense levels into the preschool period may launch children into a cycle of continued peer rejection and related adjustment concerns (Coie & Dodge, 1998; Ladd & Burgess, 1999; Nangle, Erdley, Carpenter, & Newman, 2002). Educators are becoming increasingly aware of the fact that aggression is not limited to overt forms, however, and interest in studying less direct forms of aggressive behavior is building. For instance, there is growing evidence that relational aggression, a verbally mediated form of indirect aggression focused on the undermining of relationships, is found as early as the preschool period (Crick, Casas, & Ku, 1999; Crick, Casas, & Mosher, 1997; Hart, Nelson, Robinson, Olsen, & McNeilly-Choque, 1998; McNeilly-Choque, Hart, Robinson, Nelson, & Olsen, 1996). Despite such evidence, researchers continue to focus on older children and so less is known about the early development of this form of aggressive behavior. In the absence of sufficient data, most have continued to assume that relational aggression follows a similar developmental trajectory as do physical forms.

That relational aggression is found at such young ages is somewhat surprising in light of its verbal and sophisticated nature. In their oft-cited hierarchical model charting the development of aggressive behavior, Bjorkqvist and his colleagues assert that indirect forms, such as relational aggression, emerge after more direct physical and verbal forms (Bjorkqvist, 1994; Bjorkqvist, Osterman, & Kaukiainen, 1992). Physical aggression is more prominent in the early years because young children have less developed social skills. Direct displays of verbal aggression become more prevalent as children develop their verbal skills. Last in this developmental progression, indirect verbal aggression (i.e., relational aggression) becomes more frequent not only as a response to enhanced social skills, but also with recognition of the "effect/danger" or cost/benefit ratio of aggressive behavior. That is, indirect transgressions result in a higher likelihood of injuring the intended party and a lower likelihood of putting oneself in danger. It is likely that girls progress through this hierarchy more quickly because of their relatively more advanced social and verbal skills (Bjorkqvist et al., 1992). Thus, according to this model, indirect aggression would develop at rates commensurate with that of social skills. As such, educators are likely to see more indirect aggression among older children and girls.

Before describing the empirical support for these hypotheses, the degree to which the terms indirect and relational refer to similar constructs needs to be considered. A detailed analysis of the differences in the dimensions of behavior underlying these terms is beyond the scope of the present discussion (see Underwood, Galen, & Paquette, 2001, for a review). Crick et al. (1999) contend that indirect is a broader category than relational aggression, but their actual operationalization is quite similar, with both researchers using similar items to describe both terms, such as: tells friends they will stop liking them unless friends do what they say; when mad at a person, ignores them; tries to keep certain people from being in their group during play time (Underwood et al., 2001). As suggested by Underwood et al. (2001), "Clearly the constructs of indirect and social and relational aggression overlap, and ... there is no consensus as to which construct makes the most sense" (p. 252).

In studies that employ each term, girls clearly show more relational aggression than boys, beginning at least by middle childhood (Bjorkqvist et al., 1992; Cairns, Cairns, Neckerman, Ferguson, & Gariepy, 1989; Crick, 1995, 1996; Lagerspetz, Bjorkqvist, & Peltonen, 1988; Rys & Bear, 1997). This pattern is in clear contrast to the established tendency for boys to be more physically aggressive at all ages (Bjorkqvist et al., 1992; Cairns et al., 1989; Coie & Dodge, 1998; Lagerspetz et al., 1988; Maccoby & Jacklin, 1980; Phillipsen, Bridges, McLemore, & Saponaro, 1999). Results for gender differences in relational aggression at the preschool level are slightly more mixed, with most authors finding that girls are more relationally aggressive than boys (Crick et al., 1997, 1999; McNeilly-Choque et al., 1996; Ostrov & Keating, 2004; Ostrov, Woods, Jansen, Casas, & Crick, 2004), one study finding males to be more relationally aggressive (McEvoy, Estrem, Rodriquez, & Olson, 2003), and another study finding no gender differences (Hart et al., 1998).

Less is known about age differences in relational aggression. In one of the few studies examining the impact of age, Crick, Bigbee, and Howes (1996) found that girls' relationally aggressive responses, when angered, were more common for 4th- and 5th-grade than for 3rd-grade girls. Three relational aggression studies have examined age as a variable in preschoolers, with one investigation finding no age effects (Crick et al., 1997) and one study finding that older preschoolers were more relationally aggressive (Hart et al., 1998). The third study found that a one-month period showing stability of relational aggression was equal in older and younger preschoolers (Crick et al., 1999), although this does not speak to differences in actual levels of this behavior for older versus younger preschoolers.

Replicating a study with elementary-age children (Crick & Grotpeter, 1995), Crick and her colleagues analyzed a combination of teacher ratings and peer nominations in a preschool sample and found evidence of both relational and overt aggression as independent factors (Crick et al., 1997). Teacher ratings indicated that relational aggression was higher in girls and overt aggression was higher in boys. After controlling for overt aggression, relational aggression was found to predict peer rejection and depressed affect in girls only. In contrast, relational aggression was predictive of peer acceptance for boys, indicating that this form of aggression was less negative for boys. For both boys and girls, however, relational aggression was negatively correlated with prosocial behavior.

Clearly, several variables are associated with relational aggression, such as prosocial behavior, peer rejection, depressed affect, gender, and age. It is important for preschool educators to better understand the early development of relational aggression, as there is reason to believe that it is at once associated with increased social skill yet predictive of some negative outcomes, particularly for girls. In the present study, Head Start preschoolers were assessed using a battery of measures, including peer sociometrics, as well as teacher ratings of social skills, internalizing behavior, overt aggression, and relational aggression. In this at-risk sample, skill and age differences in relational aggression were thought to be more detectable, given the increased variability in social skills and overt aggression in the Head Start population (Bryant, Burchinal, Lau, & Sparling, 1994; Duncan, Brooks-Gunn, & Klebanov, 1994; Lee, Brooks-Gunn, Schnur, & Liaw, 1990). The major goal was to determine what predictors from past research would emerge as significant for this sample. Given the association between social skills and the transition from overt aggression to indirect forms of aggression proposed by Bjorkqvist and colleagues, we hypothesized that social skills would predict relational aggression. With well-documented gender and age differences in prosocial behavior (see Fabes & Eisenberg, 1998, for a review), girls and older children were expected to exhibit higher levels of relational aggression. In sum, the specific hypotheses of the present study were that: 1) girls would be more socially skilled and relationally aggressive than boys, and 2) older children would be more socially skilled and relationally aggressive than younger children.

Method

Participants

Eighty-two preschool children enrolled in four local Head Start centers participated (n's=17, 20, 20, and 25). The participation rate was 100 percent in each of the four classrooms, due to a blanket consent achieved through the Parent Council to investigate the effects of a new curriculum. All measures reported here were collected at baseline, which was one month after the start of school to ensure that the teachers and the children all had time to get to know each other prior to providing ratings. Forty-two of the preschoolers were male and 40 were female. The sample was predominantly Caucasian (n=77), with two African American students, two Hispanic students, and one Native American student. The age of the preschoolers ranged from 36 to 62 months (M=48.4; SD=7.2).

Classroom Environment

Each Head Start classroom had a lead teacher and an assistant teacher. It is important to note that Head Start encourages parents to volunteer as aides in the classroom and this varies the student-teacher ratio day-by-day. In these classrooms, teachers reported that one parent volunteered approximately each day or every other day.

As is the case in many Head Start classrooms, the focus of the curriculum in these classrooms was on academic readiness, physical health, and social and emotional competence. Although not required to do so, all four classrooms were using activities in their lesson plans from the

Kindness Curriculum (Rice, 1995), which is an activity book provided to Maine Head Start teachers. These activities focus on social and emotional development, while also infusing academic readiness skills with work on fine and gross motor skills. Typical activities included multiple circle times; activity periods with "stations" (moving from one short activity to another); outdoor play in a small enclosed play yard that included stationary play equipment, sand box, tricycles, balls, etc.; mealtimes; and personal hygiene (toileting, handwashing, teethbrushing). The actual schedule varied according to morning versus afternoon classes, the weather, and the lesson plan of the day.

The teachers and staff employed time out for aggressive behavior or the repeated failure to obey a command. Each classroom had a designated time-out chair. Children were given immediate time outs for aggression and were given warnings (e.g., "If you do not share the truck, you will get a time out"), followed by a time out if the command was not followed once a warning was given. Time outs were brief, roughly two or three minutes long. Although some classrooms had timers to time the time outs, these tended to become misplaced and were not used. Overt aggression is a low base-rate behavior and does not occur very frequently. In a pilot study for this investigation, baseline aggressive behavior was found to occur between 0.2 to 2.3 percent of the time, while generally negative behavior (e.g., crying, noncompliance, etc.) occurred between 13.3 and 16.5 percent of the time in a similar Maine Head Start classroom, although not one of the same ones used in the present study (Carpenter & Nangle, 2002). Teachers in these classrooms did not have a specific plan for managing relational aggression, 1 although they recognized that it did occur.

Measures

Lead teachers completed three behavior rating scales for each child in their classroom. Additionally, each child met one-on-one with graduate or advanced undergraduate research assistants to complete peer acceptance ratings. Please see Table I for means and standard deviations of the measures described below.

Peer Ratings of Acceptance. A sociometric rating scale was administered individually to each child to provide a general measure of peer acceptance. The children rated how much they liked to play with each of their classmates, through pictures shown in random order, by assigning a happy, neutral, or sad face to the picture of each child (Asher, Singleton, Tinsley, & Hymel, 1979). Adequate reliability and validity have been demonstrated for peer ratings in preschoolers (e.g., Asher et al., 1979; Boivin & Begin, 1986; Hymel, 1983; Olson & Lifgren, 1988; Poteat, Ironsmith, & Bullock, 1986). Average ratings, with a possible range of 1 to 3, were determined for each child by creating a mean rating based on each child's ratings by each of his or her classmates. These average ratings provided a peer report of each child's global social acceptance or likeability.

Social Skills Rating System-Teacher Form. The Social Skills Rating System-Teacher Form (SSRS-T), Preschool Level (Gresham & Elliott, 1990), was included to provide a comprehensive assessment of social skills typically displayed in the classroom. The SSRS-T is composed of two main scales: Social Skills and Problem Behaviors (subdivided into Internalizing and Externalizing scales). Only the Social Skills scale (30 items) and the Internalizing scale (4 items) were utilized in the present study. Teachers were asked to rate on a 3-point scale the degree to which the child, for example, "Makes friends easily," "Cooperates with peers without prompting," and "Invites others to join in activities." The average standard score is reported in Table 1. The Externalizing scale or the Problem Behaviors scale was not used due to the particular focus on overt aggression in the present study, rather than on general problem behaviors. The psychometric properties of the SSRS-T indicate adequate reliability and validity (e.g., Byrne, Bawden, DeWolfe, & Beattie, 1998; Gresham & Elliott, 1990; Lyon, Albertus, Birkinbine, & Naibi, 1996; Treuting & Elliott, 1997). Cronbach's alpha for our sample was .61 on the Internalizing scale and .95 on the Social Skills scale.

Aggressive Behavior Subscale of the Child Behavior Checklist Caregiver-Teacher Report Form for Ages 2-5. Teachers were asked to complete the 23-item Aggressive Behavior subscale of the Child Behavior Checklist Caregiver-Teacher Report Form for Ages 2-5 (C-TRF; Achenbach, 1997) for each child in their classroom. Teachers are asked to rate on a 3-point scale the degree to which the child is, for example, "Defiant." For the purposes of the present study, an Overt Aggression scale was created by selecting only those items measuring overt forms of aggressive behavior. Specifically, these 9 items were: Destroys own things, Destroys others' things, Cruelty/bullying, Gets into many fights, Hits, Hurts animals/people, Physically attacks people, Screams a lot, and Teases a lot. The average sum is reported in Table 1. Cronbach's alpha for the Overt Aggression scale was .84.

Teacher Ratings of Relational Aggression. The relational aggression factor of the Preschool Social Behavior Scale-Teacher Form (PSBS-T; Crick et al., 1997) was completed by teachers. The six-item relational aggression factor taps into the construct of relational aggression with such items as, "Tells others not to play with or be a peer's friend," rated on a one to five Likert scale (1=never or almost never true, 2=not true most of the time, 3=sometimes true, 4=true most of the time, 5=always true or almost always true). The average sum is reported in Table 1. The items were derived by face-validity standards and a factor analysis indicated that relational aggression emerged as a separate factor independent of the overt aggression, prosocial behavior, and depressed affect factors. A Cronbach's alpha of .96 on the relational aggression factor indicates high internal consistency of the subscale (Crick et al., 1997). In our sample, a Cronbach's alpha of .93 was obtained.

Results

In order to conserve power, the analyses were conducted in two steps. A step-wise multiple regression analysis was conducted with relational aggression as the outcome variable. The zero-centered predictors were: Overt Aggression, Age, Gender, SSRS Social Skills scale, SSRS Internalizing scale, and Peer Acceptance. This regression accounted for 22 percent of the variance in relational aggression, F (3, 73) = 4.53, p < .001. The significant predictors from this analysis, namely Age (13 =.09), Overt Aggression ([beta]=.45), and SSRS Social Skills scale ([beta]=.05), were retained for a second multiple regression analysis that included both these predictors and all two-way interactions between these variables. The predictors were entered first into the model (i.e., Age, Overt Aggression, and SSRS Social Skills scale). Next, all two-way interactions between these variables were entered in a stepwise fashion. The model explained 27 percent of the variance in relational aggression, F (4, 74) = 6.71, p < .001 (see Table 3 for regression coefficients). Age entered the model first and in the final model, the significance of this predictor was borderline, t = 1.72, p = .09. Overt Aggression was also a significant predictor, t = 3.93,p < .001. The SSRS Social Skills subscale was a significant predictor as well, t = 1.95, p = .05. With respect to main effects, it appears that being older, more overtly aggressive, and more socially skilled all predict higher levels of relational aggression in Head Start preschoolers.

One interaction term was also significant, Overt Aggression by Social Skills, t = 2.24, p < .05. The interaction between social skills and overt aggression was disordinal when examined with overt aggression at the cross point, meaning that the cross point of overt aggression on social skills was within the meaningful range of scores for social skills, rendering the interaction interpretable (Aiken & West, 1991). However, when examining the interaction from the other direction, the cross point of social skills was outside the meaningful range of overt aggression scores (i.e., was ordinal), and these results were therefore not interpreted. In order to determine the nature of the interaction of Social Skills by Overt Aggression, the simple slopes were tested at one standard deviation above the mean, the mean, and one standard deviation below the mean, in a regression of relational aggression on overt aggression at these values for social skills (Aiken & West, 1991). Zero-centered scores were employed for these analyses.

Interpretation was restricted to the regression of overt aggression on relational aggression at fixed values of social skills, presented in Figure 1. At high social skill levels, overt aggression significantly decreased with relational aggression scores, t (1, 73) = 4.03, p < .001. At middle levels, relational aggression scores also significantly decreased as overt aggression decreased, t (1, 73) = 3.99, p < .001. At low levels, a trend for relational aggression scores to decrease as overt aggression decreased was found, t (1, 73) = 1.86, p < .07.

[FIGURE 1 OMITTED]

This interaction signifies that at high and middle levels of social skills, overt and relational aggression covary; whereas at low levels of social skills, this covariance drops to the level of a trend. In other words, children with above average or average social skills have significantly higher levels of both relational and overt aggression than do those children with below average social skills. The trend suggests that at low social skill levels, overt and relational aggression are somewhat less likely to covary, meaning a child might have somewhat higher relational aggression and lower overt aggression, or vice versa. However, examination of the plotted scores further reveals that children with low overt aggression tend to have low levels of relational aggression regardless of their overall social skill level (i.e., above average, average, or below average).

Discussion

This study sought to add to our understanding of the early development of relational aggression. Past research has implicated several variables as potential predictors, all measured in the current study: overt aggression, social skills, age, gender, peer acceptance, and internalizing behavior. Age, overt aggression, and social skills all emerged as significant predictors with an interaction between overt aggression and social skills. In looking only at the main effects, relational aggression in this sample appeared to be a function of being older, overtly aggressive, and socially skilled.

The interaction between social skills and overt aggression indicates that at high and middle levels of social skills, overt and relational aggression covary. At lower levels of social skills, there is a trend for this covariance only. In real-world terms, a preschool teacher needs to be aware that an older child in the classroom is at risk for displaying relational aggression if he or she has average to above average social skills in combination with physically aggressive behavior. For children with below average social skills, the relationship between these two forms of aggression is not as strong. Therefore, it cannot be as easily predicted when a child with below average social skills and physically aggressive behavior will display relational aggression. In looking at the plotted scores, however, it is clear that children with low overt aggression scores also have low relational aggression scores, regardless of social skill level. This relationship does not hold at higher overt aggression score levels. It appears that overt aggression may be the "gateway" to relational aggression for children at all social skill ability levels.

We originally hypothesized that 1) girls would be more socially skilled and relationally aggressive than boys, and 2) older children would be more socially skilled and relationally aggressive than younger children. As expected, older children were more relationally aggressive than younger children. It was rather surprising that girls did not emerge to be more relationally aggressive than boys. Past research with preschoolers has revealed gender differences either in expression (e.g., Crick et al., 1997, 1999; McNeilly-Choque et al., 1996) or association with other variables (e.g., Crick et al., 1997; Hart et al., 1998). Perhaps, commensurate with a developmental theory of aggressive strategies, these gender differences are not yet evident in an at-risk sample of preschoolers. Such speculation is consistent with the finding that higher levels of relational aggression are found in preschoolers from higher SES families (McNeilly-Choque et al., 1996). Stabilization of relational aggression patterns may be required before significant differences in the expression and implication of relational aggression for boys versus girls are found. In reviewing the literature in the this area, Underwood et al. (2001) noted that due to mixed results on gender differences in relational aggression, observational studies are necessary for different developmental periods.

Some speculation as to why there might be a connection between social skills and relational aggression is warranted, as their positive association seems somewhat counterintuitive. As suggested in the hierarchical model, relational forms of aggression would appear to require more advanced verbal and social skills (Bjorkqvist et al., 1992). For instance, a recent investigation by Hawley (2003) found that preschoolers who were rated by teachers as higher in relational aggression also were classified as using prosocial means to control their social resources and as higher in moral maturity. Further, the effective manipulation of peer network constituents would require some degree of social group integration. Not surprising, given its more abstract nature, indirect aggression has been found to relate positively to peer-rated social intelligence in older samples (Kaukiainen et al., 1999). Relational aggression also may be positively related to social skills because it is rewarded during its initial emergence in the preschool years. Given the emphasis on "using your words" in most preschool classrooms, teachers are likely to view any verbal assertion, even relational aggression, as preferable to physical aggression. This tendency is probably exacerbated in a low-income sample with elevated levels of externalizing behavior (e.g., Dodge, Pettit, & Bates, 1994). Whereas hitting another child earns an immediate time out, a child saying, "I'm not going to be your friend!" may not even result in teacher recognition. Further, the indirect nature of relational aggression makes it difficult to detect by adults. As such, although past research has found teacher ratings of relational aggression to be more reliable than behavioral observations in preschoolers (McNeilly-Choque et al., 1996), reliance on teacher ratings may have been a limitation of the current study. Future research should attempt to replicate these findings with more diverse measurements of behavior, perhaps through the use of an analog behavioral measure.

Although relational aggression can be considered to be a developmental "advance" of sorts, our regression results demonstrate that relational aggression also has negative implications for developing preschoolers. Overt aggression was a significant predictor of relational aggression and interacted with social skills to indicate that relational aggression and overt aggression covary at high and medium levels of social skills. To reiterate, this means that for children with average and above average social skill abilities, if they have high relational aggression, then they also have high overt aggression. If they have low relational aggression, then they also have low overt aggression. This finding is interesting because it appears that at the preschool level, relational aggression is not yet "replacing" overt aggression. At low levels of social skills, this relationship between the two forms of aggression is not as strong (i.e., at the level of a trend). Overall, as social skills scores decrease, relational aggression scores decrease, regardless of the level of overt aggression, although those with the lowest overt aggression scores appear to also have the lowest relational aggression scores. Those with high overt aggression scores, on the other hand, have varying levels of relational aggression. Therefore, although being low in overt aggression appears to be related to low relational aggression, high overt aggression does not guarantee high levels of relational aggression, at least for those with below average social skill levels. There are some preschoolers who may not have "advanced" to this more sophisticated form of aggression, and higher social skill levels may actually be one mark of this advancement. However, the expression of relational aggression is related to variables as yet unmeasured in this literature, evidenced by the fact that our model captures a commendable, yet modest, 27 percent of the variance in relational aggression.

There are a number of implications for early childhood educators. Once preschoolers who tend to be physically and/or verbally aggressive start becoming more prosocial, these seeming advances could be coupled with gains in the skills needed to become relationally aggressive. Teachers should be sure to encourage the prosocial behavior through praise and rewards, in order to reinforce these positive behaviors while instituting consequences, such as time out, for overt or relational aggression (Webster-Stratton, 1992). Teachers may choose curriculums that focus on positive social skills, such as sharing, taking turns, cooperating with others during play, etc., and devise a behavioral plan to reward behavior that is consistent with this curriculum. Teachers may be less likely to enforce consequences for relational aggression because it is less intrusive than overt aggression, and because the impression may be that the effects are lost on less sophisticated peers. While that may be true, it is critical that children's negative behavior patterns are corrected early before children are on the trajectory to peer rejection, which is difficult to deviate from and leads to multiple negative outcomes, such as school drop-out, juvenile delinquency, psychiatric impairment, and even suicide (Dodge, 1989).

One explanation for the findings of the present study is the use of a Head Start sample. Head Start typically is composed of preschoolers from low-income families. Although we did not directly measure cognitive abilities, including verbal skills, past research has clearly documented lower ability levels in at-risk preschoolers (e.g., Bryant et al., 1994; Duncan et al., 1994). Due to the fact that the use of relational aggression requires higher levels of verbal ability (e.g., Bjorkqvist, 1994; Bjorkqvist et al., 1992), Head Start preschoolers may exhibit relational aggression at a later age than preschoolers with higher socioeconomic status. In fact, preschoolers with higher socioeconomic status have been found to display higher levels of relational aggression (Bonica, Arnold, Fisher, Zeljo, & Yershova, 2003; McNeilly-Choque et al., 1996). Past research also has suggested that due to lower verbal abilities of preschoolers from low-income homes, these youngsters may be more impulsive in reaction to social problems (Mize & Cox, 1989), rather than employing more sophisticated verbal techniques consistent with relational aggression.

We hypothesize that it may be verbal skills that directly account for the findings of the present study. It could be that those preschoolers who are more verbal are precisely those who are able to exhibit relational aggression. Indeed, Bonica et al. (2003) found a positive relationship between relational aggression and verbal ability in preschoolers. We further hypothesize that these children are also more likely to have advanced social skills in comparison to age-mates. Those children expressing the highest rates of relational aggression would also have the highest rates of overt aggression, as found by our relationship between overt and relational aggression. We do, however, conceptualize overt aggression and relational aggression as distinct and independent constructs. These constructs may covary during certain developmental periods and in combination with certain variables, such as verbal skill level.

In addition to a somewhat small sample size made up of predominantly white Head Start preschoolers, this study was limited by a reliance on teacher reports of relational aggression. It would have been enriched by the addition of peer report, parent report, and/or observational data of relational aggression. Moreover, the results might be different if the measures had been given more than one month after the start of the school year. Future research should extend the present findings to preschoolers from a higher socioeconomic status and with a broader range of assessment tools for relational aggression. In addition, future research should particularly examine the role of verbal ability in the relationships among relational aggression, overt aggression, age, and social skills, ideally employing a longitudinal research design.

References

Achenbach, T. M. (1997). Guide for the caregiver-teacher report form for ages 2-5. Burlington, VT: University of Vermont Department of Psychiatry.

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.

Asher, S. R., Singleton, L. C., Tinsley, B. R., & Hymel, S. (1979). A reliable sociometric measure for preschool children. Developmental Psychology, 15, 443-444.

Bjorkqvist, K. (1994). Sex differences in physical, verbal, and indirect aggression: A review of recent research. Sex Roles, 30, 177-188.

Bjorkqvist, K., Osterman, K., & Kaukiainen, A. (1992). The development of direct and indirect aggressive strategies in males and females. In K. Bjorkqvist & P. Niemela (Eds.), Of mice and women: Aspects of female aggression (pp. 51-64). San Diego, CA: Academic Press.

Boivin, M., & Begin, G. (1986). Temporal reliability and validity of three sociometric status assessments with young children. Canadian Journal of Behavioral Science, 18, 167-172.

Bonica, C., Arnold, D. H., Fisher, P. H., Zeljo, A., & Yeshova, K. (2003). Relational aggression, relational victimization, and language development in preschoolers. Social Development, 12, 551-562.

Bryant, D. M., Burchinal, M., Lau, L. B., & Sparling, J. J. (1994). Family and classroom correlates of Head Start children's developmental outcomes. Early Childhood Research Quarterly, 9, 289-309.

Byrne, J. M., Bawden, H. N., DeWolfe, N. A., & Beattie, T. L. (1998). Clinical assessment of psychopharmacological treatment of preschoolers with ADHD. Journal of Clinical and Experimental Neuropsychology, 20, 613-627.

Cairns, R. B., Cairns, B. D., Neckerman, H. J., Ferguson, L. L., & Gariepy, J.-L. (1989). Growth and aggression: 1. Childhood to early adolescence. Developmental Psychology, 25, 320-330.

Carpenter, E. M., & Nangle, D.W. (2002). Effects of brief verbal instructions on aggression: Replication with Head Start preschoolers. Child and Family Behavior Therapy, 24, 21-38.

Coie, J. D., & Dodge, K. A. (1998). Aggression and antisocial behavior. In W. A. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp. 779-862). New York: Wiley.

Crick, N. R. (1995). Relational aggression: The role of intent attributions, feelings of distress, and provocation type. Development and Psychopathology, 7, 313-322.

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.

Crick, N. R., Bigbee, M. A., & Howes, C. (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.

Crick, N. R., Casas, J. F., & Ku, H.-C. (1999). Relational and physical forms of peer victimization in preschool. Developmental Psychology, 35, 376-385.

Crick, N. R., Casas, J. F., & Mosher, M. (1997). Relational and overt aggression in preschool. Developmental Psychology, 33, 579-588.

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

Crick, N. R., Werner, N. E., Casas, J. F., O'Brien, K. M., Nelson, D. A., Grotpeter, J. K., & Markon, K. (1999). Childhood aggression and gender: A new look at an old problem. In D. Bernstein (Ed.), Nebraska symposium on motivation: Vol. 45. Gender and motivation (pp. 75-141). Lincoln, NE: University of Nebraska Press.

Dodge, K. A. (1989). Problems in social relationships. In E. J. Mash & R. A. Barkley (Eds.), The treatment of childhood disorders (pp. 222-244). New York: Guilford Press.

Dodge, K. A., Pettit, G. S., & Bates, J. E. (1994). Socialization mediators of the relation between socioeconomic status and child conduct problems. Child Development, 65, 649-665.

Duncan, G. J., Brooks-Gunn, J., & Klebanov, P. K. (1994). Economic deprivation and early childhood development. Child Development, 65, 296-318.

Fabes, R. A., & Eisenberg, N. (1998). Prosocial development. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional and personality development (pp. 701-778). New York: Wiley.

Gresham, F. M., & Elliott, S. N. (1990). Social Skills Rating System: Manual. Circle Pines, MN: American Guidance Service.

Hart, C. H., Nelson, D. A., Robinson, C. C., Olsen, S. F., & McNeilly-Choque, M. K. (1998). Overt and relational aggression in Russian nursery-school-age children: Parenting style and marital linkages. Developmental Psychology, 34, 687-697.

Hawley, P. H. (2003). Strategies of control, aggression and morality in preschoolers: An evolutionary perspective. Journal of Experimental Child Psychology, 85, 213-235.

Hymel, S. (1983). Preschool children's peer relations: Issues in sociometric assessment. Merrill-Palmer Quarterly, 29, 237-260.

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

Ladd, G. W., & Burgess, K. B. (1999). Charting the relationship trajectories of aggressive, withdrawn, and aggressive/withdrawn children during early grade school. Child Development, 70, 910-929.

Lagerspetz, K. M. J., Bjorkqvist, K., & Peltonen, T. (1988). Is indirect aggression typical of females? Gender differences in aggressiveness in 11- to 12-year-old children. Aggressive Behavior, 14, 403-414.

Lee, V. E., Brooks-Gunn, J., Schnur, E., & Liaw, F. (1990). Are Head Start effects sustained? A longitudinal follow-up comparison of disadvantaged children attending Head Start, no preschool, and other preschool programs. Child Development, 61, 495-507.

Lyon, M. A., Albertus, C., Birkinbine, J., & Naibi, J. (1996). A validity study of the Social Skills Rating System-Teacher version with disabled and nondisabled preschool children. Perceptual and Motor Skills, 83, 307-316.

Maccoby, E. E., & Jacklin, C. N. (1980). Sex differences in aggression: A rejoinder and reprise. Child Development, 51, 964-980.

McEvoy, M. A., Estrem, T. L., Rodriguez, M. C., & Olson, M. L. (2003). Assessing relational and physical aggression among preschool children. Topics in Early Childhood Special Education, 23, 53-63.

McNeilly-Choque, M. K., Hart, C. H., Robinson, C. C., Nelson, L. J., & Olsen, S. F. (1996). Overt and relational aggression on the playground: Correspondence among different informants. Journal of Research in Childhood Education, 11, 47-67.

Mize, J., & Cox, R.A. (1989). Social knowledge and social competence: Number and quality of strategies as predictors of peer behavior. The Journal of Genetic Psychology, 151, 117-127.

Nangle, D. W., Erdley, C. A., Carpenter, E. M., & Newman, J. E. (2002). Social skills training as a treatment for aggressive children and adolescents: A developmental-clinical integration. Aggression and Violent Behavior, 7, 169-199.

Olson, S. L., & Lifgren, K. (1988). Concurrent and longitudinal correlates of preschool peer sociometrics: Comparing rating scale and nomination measures. Journal of Applied Developmental Psychology, 9, 409-420.

Ostrov, J. M., & Keating, C. F. (2004). Gender differences in preschool aggression during free play and structured interactions: An observational study. Social Development, 13, 255-277.

Ostrov, J. M., Woods, K. E., Jansen, E. A., Casas, J. F., & Crick, N. R. (2004). An observational study of delivered and received aggression, gender, and social-psychological adjustment in preschool: "This white crayon doesn't work ..." Early Childhood Research Quarterly, 19, 355-371.

Phillipsen, L. C., Bridges, S. K., McLemore, T. G., & Saponaro, L.A. (1999). Perceptions of social behavior and peer acceptance in kindergarten. Journal of Research in Childhood Education, 14, 68-77.

Poteat, G. M., Ironsmith, M., & Bullock, J. (1986). The classification of preschool children's sociometric status. Early Childhood Research Quarterly, 1, 349-360.

Rice, J. A. (1995). The Kindness Curriculum: Introducing young children to loving values. St. Paul, MN: Redleaf Press.

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

Treuting, M. V., & Elliott, S. N. (1997). Social behavior ratings of at-risk preschool children: Comparisons with typical preschool children by parents and teachers. Canadian Journal of School Psychology, 13, 68-84.

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.

Webster-Stratton, C. (1992). The incredible years: A trouble-shooting guide for parents of children aged 3-8. Toronto, Canada: Umbrella Press.

All correspondence regarding this manuscript and requests for reprints should be addressed to: Erika M. Carpenter, Ph.D., Harbor Regional Center, 21231 Hawthorne Blvd., Torrance, CA 90503. E-mail: [e-mail]. Telephone: [phone]. This research was partially funded by Penquis Community Action Program of Bangor, Maine. A very special thanks to Ellen Martzial, L.C.S.W., and her staff at Cleveland Street, Brewer, Dexter, and Old Town Head Start Centers in Maine for their dedication to both research and children.

Erika M. Carpenter

Harbor Regional Center, Torrance, California

Douglas W. Nangle

University of Maine 

Table 1
Table of Means

                           Males          Females        Total

Relational Aggression    7.71 (3.47)    7.43 (3.05)    7.57 (3.25)
                         n=42           n=40           n=82

Overt Aggression Scale   1.95 (2.91)    1.11 (2.28)    1.54 (2.64)
                         n=41           n=38           n=79

SSRS Social Skills       97.46 (16.74)  92.61 (17.43)  95.13 (17.14)
                         n=41           n=38           n=79

Peer Acceptance          2.37 (.24)     2.45 (.25)     2.41 (.25)
                         n=40           n=37           n=77

SSRS Internalizing       0.98 (1.27)    1.13 (1.40)    1.05 (1.33)
                         n=41           n=38           n=79

Age                      48.67 (6.86)   48.67 (7.70)   48.67 (7.23)
                         n=45           n=42           n=87

Note. Means are presented followed by standard deviations in
parentheses. The appropriate sample size is noted below the mean and
standard deviation.

Table 2
Total Sample Correlation Matrix

                          Age       Gender    Relational
                                              Aggression
Age                       --
Gender                    .00         --
Relational                .28 **     -.05         --
Aggression
Overt                    -.09        -.16        .27 (*)
Aggression
SSRS Social               .50 ***     .10        .22 (*)
Skills
Peer Ratings              .25 *       .15       -.10
of Acceptance
SSRS                      .05         .06       -.02
Internalizing

                           Overt      SSRS-Social   Peer Ratings
                         Aggression     Skills      of Acceptance
Age
Gender
Relational
Aggression
Overt                      --
Aggression
SSRS Social               -.45 **        --
Skills
Peer Ratings              -.26 *        .34 *           --
of Acceptance
SSRS                       .16         -.29 **        -.10
Internalizing

Note. Pairwise sample sizes range from 77 to 82.

* p < .05. ** p < .01. *** p < .001.

Table 3
Hierarchical Multiple Regression Predicting Relational Aggression
Scores

Dependent Variable           [beta]    SE    Standardized
Step/Variables                                  [beta]

Block 1: Main Effects
  Age                         .081    .052       .196
  Overt Aggression            .531    .135       .425
  SSRS Social Skills          .044    .023       .230
Block 2: Interaction Terms
  Social Skills by Overt
    Aggression                .015    .007       .235

Dependent Variable           t Test          Significance
Step/Variables

Block 1: Main Effects
  Age                         1.72               .09
  Overt Aggression            3.93               .001
  SSRS Social Skills          1.95               .05
Block 2: Interaction Terms
  Social Skills by Overt
    Aggression                2.24               .03