Crossing social contexts: Relational aggression between siblings and friends during early and middle childhood

Abstract

Children often learn and apply behaviors in sibling interactions before they routinely use such behaviors in peer interactions. As part of a longitudinal study of relational aggression, 46 children were videotaped at home in semi-structured free play situations with siblings and with friends at age 4 and again at age 8. Observations revealed that sibling dyads exhibited a higher rate of relationally aggressive behaviors than did friend dyads at age 4, but by age 8, sibling and friend dyads showed similar rates of relationally aggressive behavior. Overall, this shift occurred because relational aggression between siblings decreased whereas relational aggression between friends increased. However, this pattern varied to some degree with the sex of the target child and whether the sibling was older or younger than the target. These results suggest significant implications for the development and transmission of children's social behaviors across contexts, highlighting the importance of intervening early in childhood and in the sibling context.

Introduction

Over the last 10 years, it has been well documented that children and adolescents who engage in or are victims of high levels of relational aggression (e.g., spreading rumors, excluding others, intentionally ignoring, and threatening to terminate a friendship) are at high risk for social and psychological adjustment problems (e.g., Crick, 1997, Crick and Grotpeter, 1995, Rys and Bear, 1997, Werner and Crick, 1999). However, surprisingly little is known about the roots of relational aggression. The extent to which its use is normative in early childhood, the emergence of individual differences in its use, or the ways in which individual children's use of relational aggression may vary across interaction partners and change over time is as yet unknown. A more complete understanding of these issues could help in the early identification of children most at risk for problems with relational aggression and the development of interventions and parenting strategies to reduce that risk.

Several recent studies have documented the existence and significance of relational aggression among peers in early childhood (e.g., Nelson et al., 2005, Ostrov and Keating, 2004, Ostrov et al., 2004). However, peer interactions are not the only social context in which preschoolers might use relational aggression. In fact, preschoolers with siblings often spend at least as much time with their brothers and sisters as they do with peers outside the family (McHale & Crouter, 1996). There are several reasons to believe that sibling relationships may be an important context in which to observe the early development of relationally aggressive behaviors. Research suggests that social cognitive skills are often observable earlier in sibling interactions than in interactions with peers (e.g., Dunn & Dale, 1984), lending credence to the idea that sibling relationships might be one of the first social contexts in which children exhibit relational aggression. Several features of sibling relationships during early childhood also suggest that children may be more likely to use relational aggression in interactions with brothers and sisters than with peers and friends. During early childhood, siblings spend more time together, have a longer shared history, share a wider range of contexts and experiences, and demonstrate a higher level of intimacy than do friends (for a review, see DeHart, 1999). These relationship qualities provide siblings with a wide range of information that can later be used as ammunition to hurt or embarrass each other. In early childhood, siblings are also more likely than friends or peers to share many highly valued relationships that can potentially be vulnerable to attack, including relationships with parents, other relatives, friends, babysitters, and neighbors (O'Brien, 1999). Competition for attention from these influential people provides another explanation for why siblings might try to damage their siblings' relationships with others. In addition, because children display higher rates of conflict and physical aggression with siblings than they do with peers (DeHart, 1999), it is reasonable to expect that they may also display higher rates of relational aggression in sibling relationships than in peer relationships.

Several researchers have already demonstrated that the sibling relationship is a fruitful social context in which to observe relational aggression (Lockwood, 2002, O'Brien, 1999, Updegraff et al., 2005). Both Lockwood (2002) and O'Brien (1999) demonstrated that brothers and sisters reported engaging in rates of relational aggression that were not unlike rates reported for peer contexts. Further, their work highlights the potential influence of the sex of siblings on rates of relational aggression. Lockwood demonstrated that although research on relational aggression between peers has demonstrated that females engage in more relational aggression than do males, no such difference exists when brothers and sisters are compared. Updegraff and colleagues (2005) demonstrated that sibling intimacy and relationship negativity were highly related to adolescents' reports of their relational aggression with their siblings.

Although these findings suggest that sibling relationships are a highly conducive context for developing relational aggression, this context may become less significant in relational aggression as children age. During middle childhood peer relationships dramatically increase in salience and influence, and friendships become increasingly intimate and important (for a review, see Bernd, 2004). At the same time, sibling relationships become less intimate, less exclusive, more egalitarian, less asymmetrical, and less intense (for a review, see Buhrmester & Furman, 1990). As a result of these two processes, the differences between sibling and friend relationships diminish in important ways as children grow older-in part because the increasing symmetry in sibling relationships makes them resemble peer relationships, and in part because the increasing intimacy of friendships makes them resemble sibling relationships (DeHart et al., 2002).

Several features of friendships during middle childhood make them a likely context for the increased use of relational aggression. As children age, they share a wider range of experiences with their friends (school, home, neighborhood, and extracurricular activities), and friendships tend to last longer than they did during younger years (Bernd, 2004, Sullivan, 1953). These changes make friends privy to an ever-widening range of information that can be used to hurt or embarrass each other. Whereas sibling relationships might have consumed most of preschoolers' social time and much of their experiences, in middle childhood friendships increasingly take over many of the niches previously reserved for siblings. During middle childhood, children's social groups also begin expanding in size (Sullivan, 1953), likely providing additional influential relationships for children to utilize during such relationally aggressive behaviors as secret-sharing or exclusion from the group.

These features suggest that relational aggression may become an increasingly useful tool with which to hurt friends as children enter middle childhood. In addition, general increases in social–cognitive skills during middle childhood would be expected to produce an overall increase in children's use of relational aggression and their sophistication in using it. Indeed, many studies have demonstrated high frequencies of relational aggression in peer relationships during this developmental period (e.g., Cillessen and Mayeux, 2004, Crick et al., 2002, Giles and Heyman, 2005, Henington et al., 1998, Roecker Phelps, 2001, Rose et al., 2004, Rys and Bear, 1997, Sumrall et al., 2000, Tomada and Schneider, 1997, Xie et al., 2003).

Although several studies have examined relational aggression in sibling relationships (Lockwood, 2002, O'Brien, 1999, Stauffacher and DeHart, 2005, Updegraff et al., 2005), as yet the extent to which relational aggression is transferred between sibling and peer contexts has not been examined, in part because previous studies have not included data from both sibling and peer interactions. Several studies have suggested that physical aggression is learned and practiced in the sibling context and then later applied to the peer context (Duncan, 1999, MacKinnon-Lewis et al., 1997). It is reasonable to expect that a similar process might operate for relational aggression. This study is the first of its kind in its aim to examine the transmission of relationally aggressive social behaviors between the sibling and peer context. Although sibling relationships are a likely place to observe the earliest forms of relational aggression, children may gradually begin applying these behaviors to friend relationships as friendships become more intimate and stable during middle childhood.

Previous studies have found that amount and type of relational aggression vary, depending on the sex and age composition of sibling dyads (O'Brien, 1999, Stauffacher and DeHart, 2005). In O'Brien's study, younger sisters were more likely than younger brothers to report using relational aggression toward their older siblings, and older siblings were more likely to report using relational aggression toward their younger sisters than toward their younger brothers. In our own data on preschoolers (Stauffacher & DeHart, 2005), we did not find child sex–age composition differences in overall rates of relational aggression, but we did observe differences in the form of relational aggression used and in children's likelihood of receiving relational aggression from friends and siblings. Based on these findings, we anticipated that the specific sex composition of dyads would make a difference in the rates at which children use relational aggression with their siblings. However, the physical aggression literature contains mixed results when considering specific sibling sex constellations (Abramovich, Corter, Peplar, & Stanhope, 1986). As yet, no study has examined the role that sibling sex constellation plays in the likelihood that aggressive behaviors will cross social contexts. Given the paucity of research on the effect of sibling sex and age constellations on relational aggression, specific hypotheses regarding individual sibling gender–age compositions are premature. However, given the strong rationale for transmission of social behaviors between the sibling and friend contexts, it was expected that if sibling sex and age constellation plays a role in predicting children's relational aggression in the sibling context, a similar effect would be observed in the friend context.

Our primary goals were to examine: (1) how the rates at which children use relational aggression with siblings and with friends change as they move from early to middle childhood, and (2) how the age and sex composition of sibling pairs is related to the rate of children's relationally aggressive behavior both inside and outside of the sibling relationship. This information is important in furthering our understanding of how relational aggression within the sibling relationship facilitates the development of similar interactions in friend relationships. We anticipated that children's overall use of relational aggression would increase from early to middle childhood, but that the difference in levels of relational aggression between the sibling and friend contexts would decrease. We further expected that the age and sex composition of the sibling dyad would make a difference in target children's use of relational aggression with both siblings and friends.