5. Determining the size of sample for rigorous conclusions.

ABSTRACT: Sample size is among the most common concerns while planning an empirical research study in health and medicine. Its role in rigorous conclusions is well known. The procedure for determining the sample size is different for the estimation of a population parameter than for testing a hypothesis on a parameter. The former requires specification of the tolerable margin of error, and the latter the minimum clinically important effect to be not missed when present. But all research studies must be subjected to the highest rigor, and several other factors, such as inter-subject variability and the design, need to be considered for determining the adequate sample size. This communication compiles all these considerations together at one place. This also provides a general method for computing sample size for the two types of statistical inferences, which is not generally available in the literature. This may sensitize the reader to the ingredients required for calculating the correct sample size and the intricacies involved. The sources and tools for the calculation of sample size are also presented. This may enable the researchers and statisticians to determine the right sample size for rigorous results in their study.