The most essential concept of my blog today is to encourage you to think about two things: 1) how many participants are needed for my study, and 2) how do I select them from the larger group? I am often asked how many participants are needed in a study for it to be "statistically reliable." The answer is: it depends, but in general terms, surprisingly few! The "magic" number is usually 30. And, for sure, 300 is not 10x as "accurate" as 30. For example, if I were to randomly select 30 people from the United States to ask them who they would vote for in the next election, the result would be surprisingly similar to how the entire country might vote for the president - and we can quantify exactly how close it would be with some math. The important concept here is the idea of "randomness" - choosing random people is surprisingly difficult - how would you do it? If we can achieve some level of randomness, then there are many great tools online to help us to determine what sample size to use. Below is my favorite link. If you read the "small print" in many quantitative studies, you will find that the sample size is a limitation because it is too small. Concomitantly, one of my tips for success is to ensure that you have access to an adequate volume of data before you start any quantitative study!

http://www.raosoft.com/samplesize.html

Dr. S