Using 30 as the initial sample size of our validating studies
Reynaldo O. Joson, MD, MHA, MHPEd, MSc Surg
August 2, 2012
You and I will be validating a lot of resolutions and hypotheses that we have consensually arrived at after doing critical in-depth analyses of issues that we tried to settle or resolve.
For practical purposes, in doing a research study to validate our resolutions and hypotheses, I suggest we use 30 as an initial sample size. If there is a need to have a bigger number in the sample size, then we will increase it.
Below are the reasons why 30 is the magical number for sample size.
The number 30 came as a result of simple sampling simulations from different parent populations (Uniform, Normal, Exponential, Triangular) and by the time the sample sizes reached 30-32, the distribution of the means started looking normal. That is the reason for the rule-of-thumb.
The central limit theorem states that the sampling distribution of any statistic will be normal or nearly normal, if the sample size is large enough.
How large is “large enough”? As a rough rule of thumb, many statisticians say that a sample size of 30 is large enough.
By the time you reach a sample size of N=30 …., the shape of the distribution of sample correlation coefficients is virtually identical to that of a normal distribution …..