ABSTRACT
Epidemiological studies require accurate measurements of both health outcomes, potential confounders and estimates of exposures that might drive associations with health. Traditionally measurement error has been based around the idea of repeated measurements of a value, for example, measuring blood pressure where repeated measurements will contain a component of error, often assumed to be random. This chapter reviews approaches to classifying missing values and discusses measurement error and the effect that it may have on estimates obtained from regression models. In real-life applications, data are often incomplete and there are many reasons why observations may be missing. The regression method requires that the missing values appear in a monotone pattern. Measurement error is a general term used to encompass situations where the observed data do not represent the quantity of interest exactly.
