ABSTRACT
In a certain kind of empirical educational research, most of the work is done before you know what the results will show. That is why you do it. Once the research is written up, however, it is not easy to remember what it was like before you knew the results. Looking back on the piece of research with which this chapter is concerned, a study carried out at the National Children's Bureau (NCB) between 1977 and 1983, is not too difficult. Explaining how we did it or, rather, where you should turn if you want to know in detail what we did is straightforward. Trying to go beyond the basic methodology, however (a methodology fully explained in the various publications referred to later which have arisen from the study), in order to disentangle something hitherto unpublished about why we did the research in the way we did, is a less obvious task. More challenging still is to draw out the implications for research work of this type, that is, studies requiring analysis of statistics from large numbers of school pupils with some relevance to educational policy.
