Assessing the fit of the logistic model: a case study of children with the haemolytic uraemic syndrome

Regression methods for binary data are frequently based on the logistic model. There has been a considerable amount of very recent work concerned with methods for assessing the fit of this model to data. The purpose of this paper is to illustrate the application of these developments in model choice and assessment in terms of a case study of seventy‐five children with the haemolytic‐uraemic syndrome.