Brendgen et al

• Aims


  1)To investigate the extent to which social and physical aggression can be explained by genetic and environmental factors.2)To investigate whether physical and social aggression are caused by the same genetic and environmental factors.3)To investigate whether one type of aggression directly influences the second type of aggression (phenotype to phenotype).
• sample


  -Participants used were part of an ongoing longitudinal study (Quebec Newborn Twin Study).

  -Recruited at birth in Montreal, Canada, between 1995 and 1998. For same sex twin pairs, zygosity was assessed on physical resemblance and using genetic markers.

  -Attrition averaged approx. 2% per year, leaving 234 twin pairs at the final assessment at 6 years

  -In 175 out of 234 of the twin pairs, they were not in the same class, whereas in 59 they were.
• procedure


  -Before data collection, written consent was gained from the parents.

  -The sample was followed at 5, 18, 30, 48 and 60 months, with the final set of data being taken at 6 years.

  -Peer (done in the classroom) and teacher ratings (were done at the same time as the students did their peer reviews).
• Peer ratings


  -the children were given booklets with photos of all the children in the class they then had to circle 3 students that best fit each of 4 behavioural descriptors (2 for social aggression and 2 for physical aggression). For each of the behavioural descriptors, the total number of received nominations was calculated for each child. This was then turned into a total social aggression score and a total physical aggression score.
• teacher ratings


  -the children's physical and social aggression using the Preschool Social Behaviour Scale and the Direct and Indirect Aggression Scales. With regard to social aggression, teachers were asked to rate students on a 3 point scale for things like, to what extent the child 'tries to make others dislike a child'. With regard to physical aggression they rated the child on things like to what extent the child 'gets into fights'. They were then given a total score for physical and social aggression.
• what do results show for physical aggression


  -MZ correlations were approx. twice as high as same-sex DZ correlations (for both teacher and peer ratings). E.g. from teacher ratings, there was a correlation of .61 for physical aggression in MZ twins, but the correlation for physical aggression in same-sex DZ twins was only .25.
• what results show on social aggression


  -MZ correlations were similar to same-sex DZ correlations (for both teacher and peer ratings). For example, from teacher ratings, there was a correlation of .35 for social aggression in MZ twins and a correlation of .34 in same-sex DZ twins. Overall, the correlations are modest (moderate/not strong enough to suggest behaviour is purely linked to genetics).
• what can be concluded from results on social vs physical aggression?


  social aggression may be explained more by environmental factors than by genetic factors, as the relationship for social aggression in MZ twins was no stronger than the relationship for social aggression in same-sex DZ twins (and only a moderate relationship in both cases). So non-shared environmental factors must also play a role in physical and social aggression because neither of the results were 100% for the variance in either of these types of aggression. However, shared environmental influences may only play a negligible (very small) role in the cause of physical aggression.
• what is the possible cause of physical aggression?


  For teacher rated physical aggression, heritability accounted for 63% of the variance of physical aggression and non-shared environment accounted for 37%. For peer rated physical aggression, heritability accounted for 54% of the variance of physical aggression and the non-shared environment 46%
• what is the possible cause for social aggression


  For teacher rated social aggression, heritabilityandshared environment accounted for 20% each and the non-shared environment accounted for 60%. For peer rated social aggression, heritabilityandthe shared environment account for 23% each and non-shared 54%.
• conclusions


  -The findings support the phenotype to phenotype explanation of the link between physical and social aggression in children.

  -The results suggest a substantial contribution of genetic factors on individual differences for physical aggression (as twins who share more genes showed greater similarity in behaviour).
• AO3 inter rater reliability


  -both teachers and peers rated behaviour on social aggression and physical aggression increasing validity and reliability as it was from more than one source
• AO3 supporting validity


  Translation of the measures used was checked by bilingual researchers and was translated twice to ensure it was done correctly. This increases thevalidityof the measures used.
• AO3 supporting reliability


  The findings that there were observable individual differences in social aggression as well as in physical aggression at 6 years of age, based on teacher and peer ratings, supports previous research done in this area. As does the fact that physical aggression appears to be mostly explained by genetic factors. This suggests there isreliabilityin the findings.
• AO3 sample


  There was asmall sample sizewhich precluded the examination of sex differences with regard to genetic and environmental effects on social versus physical aggression.
• AO3 generalisability


  Physical aggression is already diminished at school entry whereas social aggression may not be fully developed until age 8 not representative of general population
• AO3 attrition rates


  The twin pairs that dropped out of the study due to attritionmay have influenced the resultsdifferently if they remained in the study. For instance, the fathers that had dropped out of the study had a slightly lower level of education than the fathers that stayed in the study. This may have had an impact on the scores that the twins received if this influenced the twin’s behaviour (i.e. if it was an environmental factor). However, as attrition only averaged approx. 2% per year, it is unlikely that this number of twin pairs would have significantly altered the results.