Siblings -friends or foes?

Alison Pike, Tina Kretschmer and Judith F. Dunn on what the research says about achieving a harmonious household
The vast majority of us have at least one brother or sister, yet sibling relationships have received scant attention in the UK. While many parents claim to have a second child as a ‘companion’ for their first child, the reality is that many brothers and sisters spend much of their time locked in conflict. At the extreme, it is in fact the case that children are more likely to become the victims of abuse by a sibling than by any other family member. On the flip side, many brothers and sisters are, at least at times, the best of friends. This article describes a study designed to uncover features of families as well as individual children that foster warm, intimate relationships between siblings versus hostile, conflict-ridden brother–sister interactions.

For most of us, our relationships with brothers and sisters are the longest lasting of our lives. Siblings come before friendships or romantic relationships, and usually outlive our parents. The theme of sibling rivalry might be the most prevalent in biblical and epic writings, and also in Sigmund Freud’s psychoanalytic theory of the Oedipus complex.

However, although rivalry is one salient characteristic of most sibling relationships, there are many others. Sibling relationships are usually characterised by both positive and negative interactions, and provide a safe training ground for interactions outside the home. Conflicts, quarrels and aggression are more prevalent among some sibling pairs; others are high in intimacy, emotional and social support; yet others involve an ambivalent combination of the two (Brody, 1998; Dunn, 2002).

Whatever the relationship, it is often highly charged; exchanges between siblings are marked by both greater warmth and greater conflict than exchanges with either parents or friends (Dunn et al., 1996).

Siblings also spend a considerable amount of time with each other, more so than with their parents (Larson & Richards, 1994), and know each other well. The intimate knowledge that siblings have of one another as well as the emotional intensity of the relationship means that siblings have the potential to significantly impact on one another’s development and well-being.

Until about three decades ago, the importance of siblings in terms of individual children’s psychological development was neglected by researchers, despite the fact that in both the UK and the US around 80 per cent of children still grow up with at least one brother or sister. Given that some brothers and sisters get on well and others badly, it’s not surprising that psychologists are often asked by our hairdressers/taxi drivers/party guests: ‘My kids fight all the time, what can I do about it?’ or ‘What’s the ideal spacing between children so they get along?’ So for practical as well as theoretical reasons, we, along with other sibling researchers, have focused on what factors predict which brothers and sisters will get along well, and which badly.

Structural factors such as age spacing and gender composition of siblings have been examined in the US by Buhrmester and Furman (1990) and found to affect the quality of relationships between brothers and sisters in childhood and adolescence. Sibling pairs in which children were less than four years apart in age, for example, were more intimate, but also more competitive. In addition, the interactions of same-sex sibling pairs in childhood often involve more aggression and dominance than interactions of opposite-sex sibling pairs, although this does not preclude warmth and closeness. Interestingly, all children in Buhrmester and Furman’s study reported greater intimacy and companionship with sisters than with brothers, regardless of gender composition or age difference. Brothers’ and sisters’ temperaments are also important factors that impact on how well they get along with one another. Brody (1998) summarised that especially active and emotionally intense children experience elevated levels of conflict in relationships with their siblings. Other relationships within the family are also associated with sibling relationship quality. Studies have compared the quality of children’s sibling relationships and patterns of other familial relationships and found impressive overlap. In other words, children and adolescents who report positive relationships with their parents also describe their relationships with their siblings as good (see Brody, 1998, for a review). This pattern fits well into the theoretical concepts of attachment and social learning, according to which children’s social development is largely determined by their relationships to their parents or other primary caregivers.

However, all this research used samples from the US. There has been little if any psychological research of sibling relationships in the UK since Judy Dunn and colleagues’ intensive study of a small group of Cambridge families in the early 1980s (e.g. Dunn et al., 1982). In addition, previous research on young children’s sibling relationships has relied on parental reports of sibling relationship quality. We began the Sisters and Brothers Study with the aim of describing sibling relationships among typical brothers and sisters aged four to eight years old, living in southern England. A book and several journal articles stemming from the project have been published, and form the basis of this article (see Atzaba-Poria & Pike, 2008; Coldwell et al., 2006, 2008; Kretschmer& Pike, 2008; Pike et al., 2005, 2006).

The Sisters and Brothers Study
We asked the children themselves about their relationship with their brother or sister, via a puppet interview (Ablow & Measelle, 1993). During the interview, two identical puppets made opposing statements about their sibling (e.g. ‘I like my brother’, ‘I don’t like my brother’) and then asked the child about their sibling (e.g. ‘How about your brother?’). The researcher’s face was covered by the puppets, to encourage the child to interact directly with ‘Iggy’ and ‘Ziggy’.

Children often simply repeated the item back to the puppets, or repeated the item providing some explanation:
Iggy: I don’t let my sister play in my room.
Ziggy: I do let my sister play in my room. How about you?
David: I have to let her play in my room because we share the room but I don’t let her play on my bed.

Iggy: I do get cross when my sister plays with my toys.
Ziggy: I don’t get cross when my sister plays with my toys. How about you?
Andrew: I do get cross when my sister plays with my toys cos I’ve got them up in a high place and she knocks them down.

Sometimes children amplified statements made by the puppets:

Iggy: My brother hates me.
Ziggy: My brother doesn’t hate me. How about your brother?
Jess: My brother doesn’t hate me, I think he loves me actually.

At other times children qualified the statements:

Iggy: My brother and I argue.
Ziggy: My brother and I don’t argue. How about you and your brother?
Sophie: Me and my brother squabble over little things, not really big things.

Gaining meaningful information from very young children does pose significant challenges to researchers and practitioners alike. We were thus extremely pleased with the success of the puppet interviews. This child-friendly method yielded quantifiable information that ‘agreed’ with reports from the parents to a similar degree as has been reported for older adolescents (e.g. Hetherington et al., 1999). We think it was equally important, however, that the children’s reports did not simply duplicate the reports provided by the parents. The moderate levels of agreement indicate that there is also a great deal of distinctiveness to children’s perceptions.

Next, we moved on to look at whether it was possible to predict the quality of sibling relationships from demographic factors. First, we found no differences in sibling relationship quality between children living with single mothers and children from two-parent families. Such findings run counter to the stereotypical view, and further challenge definitions of families that focus on structural components. This implies that families are best conceptualised by their constituent relationships, rather than household membership.

The next demographic factor considered was the gender constellation of the sibling pairs. Thus, we compared boy–boy, boy–girl, girl–boy, and girl–girl sibling pairs in terms of their sibling relationship quality. According to parent reports there were no significant differences according to the gender of the sibling pairs. When looking at the children’s reports via the puppet interviews, however, a difference did emerge from the older siblings’ perspective. Specifically, older sisters with younger sisters reported the most warmth, and older brothers with younger sisters reported the lowest levels of warmth.

The final demographic factor considered was the age gap between siblings. This ranged in magnitude from zero (there were two set of fraternal twins n our sample) to 52 months. A modest but consistent pattern emerged, indicating that a smaller age gap enhances sibling relationships. However, we found that once older sibling age was accounted for, age gap no longer predicted sibling relationship quality. In other words, regardless of the age gap between siblings, those sibling pairs containing an older sibling of seven or eight enjoyed less positive sibling relationships than did those pairs containing an older sibling of five or six.

It is worth highlighting that the ‘significant’ links with demographic factors were all quite small in magnitude. Our findings thus confirm previous research indicating that these structural factors do not play a key role in family life.

Next, we turned to the individual children’s characteristics. The four aspects of child temperament assessed were emotionality (specifically a tendency to display negative emotions such as anger, fear, or upset), activity, sociability and shyness. As expected, highly emotional children were involved in less positive and more negative sibling relationships. Only one additional association reached significance, that between the older siblings’ sociability and sibling relationship positivity (as reported by the parents). More sociable older siblings were involved in more positive sibling relationships.

Finally, we looked to contextual factors that we hypothesised would work as protective or risk factors for how well brothers and sisters get on with one another. Four distinct aspects of the family context were examined: household organisation, marital satisfaction (in the case of two-parent families), socio-economic status, and household crowding. We expected that less stressful family circumstances would be supportive of better-quality sibling relationships. Thus, we hypothesised that families living in more organised homes (specifically homes with regular routines and low levels of background noise), where parents were more satisfied with their marriages, of higher socio-economic status, and lower levels of crowding would also contain siblings with high-quality relationships.

Household organisation emerged as an most important contextual factors for sibling harmony. More organised families fostered more positive and less negative sibling relationships. Among the two-parent families, three of the four correlations with marital satisfaction also yielded significant findings. That is, better-quality parental relationships appear to ‘spill-over’ into more positive relationships between their children, suggesting that children may learn positive relationship skills from their parents, or that happily married parents are better able to structure their children’s environment and respond in a manner conducive to more amicable relations between their children.

It was equally important to find that no significant links emerged with socio-economic status, or household crowding. This mirrored the earlier demographic findings – these more objective, structural aspects of families wane in importance when compared to more subjective, detailed aspects of the everyday lives of families.

Links that emerged between contextual factors and the sibling relationship highlighted the fact that sibling relationships do not occur in a vacuum, but are inextricably tied up with their surroundings. In particular, we were struck by the links with the marital relationship as well as household organisation. Unlike parenting, these factors do not directly relate to the children involved, yet these associations were moderate in magnitude. These findings remind us that children are not immune from tension between parents, and that a more chaotic home environment can put a strain on all family members, perhaps facilitating sibling conflict. A holistic approach to family functioning is thus warranted (Minuchin, 2002). The family system is also permeable to contextual factors within and outside the family.

We are forced to conclude that there is no easy answer as to why some siblings get along well while others become locked in conflict. However, we can say that enhancing sibling relationships has very little to do with family structure, age spacing, or even the sex constellation of sibling pairs. Instead, our findings indicate that children’s individual characters and the family environment are important factors in determining whether brothers and sisters become friends or foes.

Future directions
The sample of our own study included working- and middle-class families; however, the population from which the sample was drawn is primarily Caucasian. Given Britain’s ethnic and cultural diversity, future research including ethnic minority groups and families across different cultures has the potential to increase the generalisability of current findings or to uncover cultural variations reflecting differing family processes. In addition, further research using selected samples will illuminate the family dynamics that emerge in the case of physically or mentally ill children.

Recent advances in statistics are also beginning to have a profound effect on sibling research. Traditional analytic techniques made it difficult to contemplate incorporating information from more than two children per family. Multilevel modelling has the flexibility to use information from sibships of varying sizes, and to identify those factors that affect differences between families, as well as those that lead to differentiation within families. We are thus entering an era when we will be able to better understand the development of singleton children, children with one brother or sister, and those with multiple sibling relationships

Alison Pike is in the Psychology Department, University
of Sussex
[email protected]

Tina Kretschmer  is in the Psychology Department, University of Sussex
[email protected]

Judith Dunn is Professor of Psychology in the Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, London
[email protected]

Ablow, J.C. & Measelle, J.R. (1993).  The Berkeley Puppet Interview: Interviewing and coding systems manuals. University of Oregon: Department of Psychology.
Atzaba-Poria, N. & Pike, A. (2008). Determinants of parental
differential treatment: Parental and contextual factors during middle childhood. Child Development, 79, 217–232.
Brody, G.H. (1998). Sibling relationship quality: Its causes and consequences. Annual Review of Psychology, 49, 1–24.
Buhrmester, D. & Furman, W. (1990). Perceptions of sibling relationships during middle childhood and adolescence. Child Development, 61, 1387–1398.
Coldwell, J., Pike, A. & Dunn, J. (2006). Household chaos – Links with parenting and child behaviour. Journal of Child Psychology and Psychiatry, 47, 1116–1122.
Coldwell, J., Pike, A. & Dunn, J. (2008). Maternal differential treatment and child adjustment: A multi-informant approach. Social Development, 17, 596–612.
Dunn, J. (2002). Sibling relationships. In P.K. Smith & C.H. Hart (Eds.) Blackwell handbook of childhood social development (pp.223–237). Malden, MA: Blackwell.
Dunn, J., Creps, C. & Brown, J. (1996). Children’s family relationships between two and five: Developmental changes and individual differences. Social Development, 5, 230–250.
Dunn, J., Kendrick, C. & MacNamee, R. (1982). The reaction of first-born children to the birth of a sibling: Mothers’ reports. Annual Progress in Child Psychiatry and Child Development, 143–165.
Hetherington, E., Henderson, S.H., Reiss, D. et al. (1999). Adolescent siblings in stepfamilies: Family functioning and adolescent adjustment.
Monographs of the Society for Research in Child Development, 64(4), 222.
Kretschmer, T. & Pike, A. (2008). Young children’s sibling relationship quality: Distal and proximal correlates. Journal of Child Psychology and Psychiatry, 50, 581–589.
Larson, R.W. & Richards, M.H. (1994). Family emotions: Do young adolescents and their parents experience the same states? Journal of Research on Adolescence, 4, 567–583.
Minuchin, P. (2002). Looking toward the horizon: Present and future in the study of family systems. In J.P. McHale & W.S. Grolnick (Eds.) Retrospect and prospect in the psychological study of families. (pp.259-278). Mahwah, NJ: Lawrence Erlbaum.
Pike, A., Coldwell, J. & Dunn, J. (2005). Sibling relationships in early/middle childhood: Children’s perspectives and links with individual adjustment.  Journal of Family Psychology, 19, 523–532.
Pike, A., Coldwell, J. & Dunn, J. (2006). Family relationships in middle childhood. York: York Publishing Services/Joseph Rowntree Foundation.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber