ABSTRACT
Sample size calculation is one of the most commonly performed tasks of a collaborating biostatistician. Sample size and power calculations are essential in any form of research that involves hypothesis testing. This encompasses studies of the various forms of study designs discussed in Chapters 1 and 2. If too few patients are recruited, there is a possibility of the study being under-powered to detect any meaningful effect, and conversely if too many patients are recruited, the study can be over-powered, and this can then result in wastage of time and money, potentially putting patients in the study at additional risk, as well as causing unnecessary delays in the study (Earnest, 2010). Declaring a trivial effect size statistically significant can lead to subsequent confusion with clinical significance. Increasingly, journal submissions request for sample size calculations to be explicitly stated in the methods section of the manuscripts, and articles are often sent for external statistical review. Some journals also have statisticians as members of their editorial boards, and this highlights the emphasis that is placed on statistical methodological issues, including appropriate statistical analysis plan and sample size calculations.
