April 27, 20179 yr The significance of sample size comes from manufacturing. This meant that to check for the accuracy of output per machine you need to audit x sample. i.e If Machine A produces 60,000 units you may want to audit 360 approx. sample. But this sample size is good for 60,000 units if produced by A. However if 60,000 are produced collectively by A & B then what should be the sample size. The fact is if Machine A is 3 yrs old and Machine B is 6 years old. Should you - Audit 360 overall Base sample size basis individual production Divide 360 proportionately basis production The intent is to understand the impact of the variable on sample size. Also, do you recommend stratified sample rather than a random sample?
August 1, 20178 yr Hello Tony, The sample size is dependent on several things, one of which is what are you trying to test. If your objective is to make sure that the accuracy of each of the machines meets a certain target, you should collect applicable sample size from each of the machines. However, if you want to ascertain if you meet an overall accuracy target, you can collect a combined sample from both the machines and validate statistically. This could be a random stratified sample based on production percentage by the machine.
Create an account or sign in to comment