Survey design and sample selection

broached is through identifying respondents who have separated recently and then collecting information about their past retrospectively. This is problematic because of recall problems and because of the likely contamination of response as a result of the separation. Objective analyses of the impact of marital separation require data collected preand post-separation, which in turn require longitudinal methods. Second, the HILDA Survey involves interviews with all members of the household aged 15 years or over. This again marks the survey as distinctive. Information about household members is usually sought of one adult only responding on behalf of others in the household – a method that precludes asking subjective questions, and raises concerns about the accuracy of some of the data collected. For example, while it can be expected that one household member will know the employment status of all other household members, is it reasonable to expect them to know more specific details of their employment arrangements (such as hours of work)?