Brengden (2005)

investigated aggression in MZ and DZ twins..
1. to discover the extent to which social + physical aggression is explained by genetic (shared) + environmental (non-shared) influences
2. to discover whether the overlap between social + physical aggression is explained by the direct effect of one type of aggression on the other

234 twin pairs from the longitudinal QNTS
44 pairs of male MZ, 50 female
41 pairs of male DZ, 32 female

twin study design
the QNTS collected data regularly, beginning at 5 months
At 72 months the data reported in the article was collected

Kindergarten teachers rated the social + physical aggression of each child on a 3-point scale (never, sometimes, often)
in response to questions such as 'says bad things or spreads nasty rumours' (social aggression) and 'hits, bites or kicks others' (physical aggression)

The level of the twins social and physical aggression was also rate by their classmates.
- 1st a research assistant checked that each child could recognise everyone in their class
- then each child was given a booklet of photographs of all the children and was asked to nominate the 3 children on one page who best matched the description of behaviour
- such as 'tells mean secrets about another child' (social) and 'gets into fights' (physical)

- only 20% (teacher ratings) to 23% (peer ratings) of social aggression was explained by genetic factors
- physical aggression was mostly explained by heritable factors - 63% (teacher) to 54% (peer)
- social aggression is mostly explained by non-shared environmental influences - 60% (teacher) to 54% (peer)
- high physical aggression led to high social aggression, but the opposite was not true

The data suggests that genetic characteristics predispose some children to aggressive behaviour in general (e.g. irritable temperament)
However, further analysis in the form of 'modelling' indicated that the specific form the aggression takes is more influenced by environmental factors

Increased generalisability due to relatively equal numbers of male and female. Due to genetics being different for men and women - a study of only one gender would not be representative of the target population and results would be inaccurate of aggressive behaviours relation to genes

a strength of the study is that the researchers used ratings of aggression from 2 sources. Each twin was rated by both their teachers and the other children. it is common for research to use teacher ratings, but this was the first study of social aggression to collect data from both sources. The two sets of ratings were essentially the same, which provides some reassurance that they are a valid measure of aggressive behaviour

A weakness of this study is that it assumes MZ and DZ twins experience the same environment therefore assuming that environmental influences are shared equally. This may be an incorrect assumption, for example it is more likely for a parent of DZ twins to behave more aggressively to one child than the other because they are dissimilar – however this isn’t the case for MZ twins. Undermining the validity of the twin study method as findings relying on equal environmental influence may be incorrect.

Derks et al (2006) conducted a study of aggression in adolescent twins and used statistical methods to provide evidence that the equal environments assumption is reasonably valid. This therefore means that Brengden’s results are accurate

This study can lead to effective interventions to reduce aggression – since this study highlighted the progression of aggressive behaviour, interventions should focus on preventing /
reducing aggression in the early years. Furthermore, this study paves the way for future studies to research factors to block the path from physical to social aggression – ultimately reducing both types.

interventions (application) could make aggression worse by labelling children as aggressive, they will conform to the label and be aggressive

sample / generalisability
procedure (ratings) / validity (triangulation)
findings / internal validity / CA (Derks et al 2006)
conclusions / applications / CA (labelling + SFP)