Sample size guidance for surveillance data
Sample size calculation is an essential step in ensuring that analysis of the data which are collected through surveillance systems can support the desired public health objective(s). The main aim of sample size calculations is to determine the sample size needed from the population under study to possibly detect a statistically significant result.
Executive summary
This document provides guidance to support disease experts in doing sample size calculations according to different surveillance objectives. Its main aims are to:
• introduce the statistical terminology and key concepts used in relation to sample size and power calculations;
• provide guidance for doing sample size calculations for typical situations that involve the methods most frequently used in surveillance analyses;
• illustrate these calculations with examples and easy-to-use R scripts.
This document is organised as follows: Section 1 presents an overall introduction to sample size calculations for surveillance data and a list of objectives for this guidance. Section 2 enumerates the key statistical concepts and definitions used in sample size calculation and provides illustrative examples drawn from different sample size requirement situations, which are mostly relevant in the context of ECDC. Section 3 presents sample size calculations for surveys using simple random sampling, where formulae are provided according to the different situations being considered. Each outcome is illustrated with a scenario, along with the provision and explanation of R codes to obtain the required sample size. Most computations are done with available R packages.
Download
