Sex differences in the relation between temperament and aggression

in 5- to 13-year-olds

Lisabeth F. DiLalla & Paula Y. Mullineaux

Southern Illinois University School of Medicine

Poster presented to the Society for Research in Child Development, April, 2003, Tampa, FL.

ABSTRACT

The purpose of this project was to explore the relation between temperament and aggression, including sex differences in this relation. Children were assessed at 5 years of age and again at approximately 11 years of age to determine whether early parent temperament ratings were correlated with early peer play and later teacher rated aggression. Forty-four children (19 girls and 25 boys) with complete data, who were part of the Southern Illinois Twins and Siblings Study, were tested at age 5 years and were re-assessed at a mean age of 11.7 years. At age 5, parents completed a temperament questionnaire and the children participated in a peer play situation. Aggressive and prosocial behaviors were coded. At follow-up at approximately age 11 years, teachers were asked to complete several questionnaires about children's aggressive behaviors and prosocial peer relationships. Results showed that temperament was significantly correlated with the aggression measures for boys but not for girls. There was no evidence for temperament as a mediator between aggressive responses in a stressful situation (the non-familiar peer play at age 5) and aggression at follow-up. This finding may help to explicate the mechanism between early difficult temperament and later externalizing problem behaviors. Both biological and social aspects specific to boys may drive this relation.

INTRODUCTION

Early difficult temperament has been shown to correlate with later conduct problems in young children (Bates, Maslin, & Frankel, 1985) and adolescents (Giancola & Parker, 2001; Gjone & Stevenson, 1997). The exact mechanism for this is not clear. It may be that difficult children evoke more negative environments (e.g., peer rejection, associations with difficult peers), suggesting reactive gene-environment (g-e) correlation. Also, difficult children may respond to stressful environments in a more aggressive manner than easier children, who have temperamental resources on which to draw, suggesting gene-environment interaction. Additionally, some studies have suggested that the relation between temperament or emotion and problem behaviors may be more salient for boys (e.g., Gjone & Stevenson, 1997; Eisenberg et al., 2001). To begin to explore these issues, children were assessed at age 5 and again at approximately age 11 to determine whether early parent temperament ratings were correlated with early peer play and later teacher-rated aggression.

We hypothesized that children with parent-rated difficult temperament would show more aggression in the lab during free play with an unfamiliar peer and also would be rated as more aggressive by their teachers several years later. Specifically, we hypothesized that temperament would be a more important predictor of aggression for boys than girls. Also, to explore g-e correlation, we hypothesized that children with more difficult temperaments would evoke more negative responses from the peer during free play.

METHODS

· The sample comprised 44 children (19 girls, 25 boys) who were part of a larger, on-going study (the Southern Illinois Twins and Siblings Study; DiLalla, 2002) and who represent a subsample for whom complete data were available. As part of the larger study, children were tested in a lab playroom at age 5 and follow-up information was obtained at a mean age of 11.7 years (range 11 to 13 years).

· At age 5, children were brought to a laboratory playroom and were allowed to play freely for 20 minutes with a same-sex, same-age, unfamiliar peer. Play was videotaped and later coded for prosocial and aggressive behaviors by trained raters. Parents completed several questionnaires at this time. Children received toys to thank them for participating.

· Six to eight years later, 146 parents of children aged 10-13 years were re-contacted and asked to provide the names of their children's teachers. 51 (43% of those who could be located) agreed to participate. Teachers were asked to complete several questionnaires about children's aggressive behaviors and prosocial peer relationships. 44 teachers returned completed questionnaires. Children and teachers were sent $5 gifts.MEASURES

AGE 5 YEARS:

Parents completed:

• •Behavioral Style Questionnaire (BSQ; McDevitt & Carey, 1978)

Four temperament dimensions were combined to form a single measure of difficult temperament (Cronbach's a = .56):

• •High Activity
• •Slow to Adapt
• •High Intensity
• •Low Threshold

Behavioral ratings of aggression during free play (inter-rater reliabilities > .80):

• •Aggression Total (Cronbach's a = .76)

Sum of 4 items scored on 5-point Likert-type scales:

• •Level of aggression during 1^st and 2^nd 10 minutes

• •How difficult was child during 1^st and 2^nd 10 minutes?
• •Rejection (Cronbach's a = .67)

Sum of 4 items scored on 5-point Likert-type scales (inverse):

• •How much other child enjoyed playing with target during 1^st and 2^nd 10 minutes
• •How much other child complied to target during 1^st and 2^nd 10 minutes

FOLLOW-UP (AVG. AGE 11.7 YEARS):

Teachers completed:

• •Teacher Report Form (TRF; Achenbach, 1991)

Measure of child problem behaviors in the school setting. Aggression scale (raw score) was used in this study.

• •Teacher's Checklist of Peer Relationships (TCPR; Dodge, 1986)

This checklist contains three scales: Social Competence, Reactive Aggression, and Proactive Aggression.

Means are presented in Table 1.RESULTS

• • One-way ANOVA tests demonstrated sex differences for the two peer play variables only (see Table 1).

• •Boys were more aggressive than girls, and girls were more rejected (enjoyed less, complied to less) than boys.



• •Correlations were calculated between Difficult Temperament and all Aggression measures at age 5 and at follow-up.

• •No correlations were significant for the total sample.



• •A different picture emerged when analyses were conducted separately for boys and girls (see Table 2).

• •No correlations were significant for girls.

• •Temperament was either marginally or significantly correlated with all other measures for boys.

• •Regression analyses assessed whether Temperament was a moderator between Aggressive responses in a stressful situation (the non-familiar peer play at age 5) and Aggression at follow-up (see Figure 1).

• •There was no evidence for Temperament as a moderator. The interaction term did not significantly predict follow-up Aggression.



• •Correlation analyses were used to assess whether more difficult children evoked more negative responses from peers during free play.

• •Difficult temperament was not significantly correlated with Peer Play Rejection for boys, girls, or the total sample.

CONCLUSIONS

These results suggest that the relation between temperament and aggression is dependent upon sex of the child. Boys were more aggressive than girls at age 5, and boys with difficult temperaments were more prone to aggression at age 5 and also at follow-up than were girls. This finding may help to explicate the mechanism between early difficult temperament and later externalizing problem behaviors. Both biological and social aspects specific to boys may drive this relation.

We did not find support for the hypothesis that children with more difficult temperaments evoked more negative responses from the peer during free play. There was no significant correlation between difficult temperament and the reaction of the peers to the target child during play. Thus, our results do not support reactive gene-environment correlation as a mechanism that might explain the relation between difficult temperament and aggression.

This study avoids the common confound of having parents rate both temperament and aggression because aggression was rated by trained coders at age 5 and by teachers at ages 11 to 13. However, the small sample size is a limitation and replication with a larger sample will be necessary. Future research will explore possible mechanisms for the sex difference in the relation between difficult temperament and aggression, including the interaction of child temperament and parent personality. Future research will also include family interaction styles as possible mediators.

TABLE 1

Sample Descriptives

^a Boys were significantly more aggressive than girls, F(1,31) = 14.29, p < .001.

^b Girls were significantly more rejected by peers, F(1,29) = 7.87, p < .01.

TABLE 2

Correlations between 5-year-old Difficult Temperament and Aggression

Girls Boys

(N = 19) (N = 24)

Aggression Total -.21 (N=15) .38 (N=18)

[Including 5-year-old sample

without follow-up data: .01 (N=61) .28* (N=61)]

Rated by Teachers (age 11 years):

TRF Aggression (raw) .15 .40+

Reactive aggression -.21 .35+

Proactive aggression -.05 .53**

Prosocial .00 -.49*

* p < .05

+ p < .10

References

Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: University of Vermont Department of Psychiatry.

Bates, J. E., Maslin, C. A., & Frankel, K. A. (1985). Attachment security, mother-child interaction, and temperament as predictors of behavior-problem ratings at age three years. Monographs of the Society for Research in Child Development, 50, 167-193.

DiLalla, L. F. (2002). Preschool social and cognitive behaviors: The Southern Illinois Twins. Twin Research, 5, 468-471.

Dodge, K. A. (1986). A social information processing model of social competence in children. In M. Perlmutter (Ed.), Eighteenth Annual Minnesota Symposium on Child Psychology (pp. 77-125). Hillsdale, NJ: Erlbaum.

Eisenberg, N., Losoya, S., Fabes, R. A., Guthrie, I. K., Reiser, M., Murphy, B., Shepard, S. A., Poulin, R., & Padgett, S. J. (2001). Parental socialization of children's dysregulated expression of emotion and externalizing problems. Journal of Family Psychology, 15, 183-205.

Giancola, P. R., & Parker, A. M. (2001). A six-year prospective study of pathways toward drug use in adolescent boys with and without a family history of a substance use disorder. Journal of Studies on Alcohol, 62, 166-178.

Gjone, H., & Stevenson, J. (1997). A longitudinal twin study of temperament and behavior problems: Common genetic or environmental influences? Journal of the American Academy of Child & Adolescent Psychiatry, 36, 1448-1456.

McDevitt, S. C., & Carey, W. B. (1978). The measurement of temperament in 3-7 year old children. Journal of Child Psychology and Psychiatry, 19, 245-253.