Moreover, as this was a naturally occurring study, children were not randomly assigned to conditions (adopted before 6 months, adopted between 6months and 2 years, adopted after 2 years) as it would have been unethical to do so. Therefore, those children adopted earlier may have been the more sociable ones which could explain why they were chosen by the adoptee parents and consequently had less development issues. This would be a confounding variable, therefore lowering the internal validity of Rutter’s research into the effects of institutionalisation.