Repeatability

Abstract

There are many sources of variability in measures in epidemiology. Subjects can vary in their response because of real change. Alternatively, the method of collecting data may be subject to variability because of differences in techniques or the inherent difficulty in some measures in ensuring consistency. Too much variability can introduce ‘noise’, making it difficult to find differences between populations. There may be a variety of reasons for failure to obtain consistent or reproducible results. This chapter describes various methods which can be employed to calculate the level of variability in measures used and how to minimize the sources of this.