Selected methodological problems in research in child abuse: a developmental perspective.

Abstract

It is apparent from extensive literature reviews that only a few authors have attempted to enhance the quality of their research in the area of child abuse. Despite the refinements made by these authors there are still gross weaknesses apparent in much of the research. These deficits are clearly linked to the problems encountered in doing research in childhood psychopathology in general where there is fragmentation both on an empirical and theoretical level. The aim of this paper is to represent some of the many issues which have to be taken into account when conducting research into childhood psychopathology in general and child abuse in particular and when making comparative evaluations of the research. Amongst the issues discussed are: the purpose of the research; naturalistic v. experimental continuum; techniques of data collection; expectations of subjects; experimental bias; individual v. group observation; sampling issues viz definition of abuse; age at which assault occurred; time lapse investigation and last known incidence of abuse; definition of subject sample; demographic variables; technical issues and behavioural research, e.g. specificity v. generality of behaviour; unit of use and time sampling.