Epidemiology for nutritionists: 2. Sampling

There are two major objectives in sampling for a research study. The first is the avoidance of bias due to selection: to achieve this a random sample must be drawn and a high response rate achieved. The second objective is to achieve an efficient use of resources by selecting a sample of a size sufficient to detect an effect of clinical importance yet not so large that effects too small to be of interest are detected. Both these aspects will be discussed here.