Calculating Sample Size for DVT Testing
- August 10, 2018 at 11:45 am #56066
This is a really basic question. I took Stats over ten years ago and do have some understanding of concepts, but I am struggling with calculating sample size in Minitab.
I would like to generate sample sizes of variable or attribute data given confidence and reliability values of 95/95, 95/90, 90/80 etc. We do have data we can leverage for mean and standard deviation. Which method would you use?
Also, are there some good resources for gaining a more advanced knowledge of statistics as applied in a manufacturing/development environment? I have taken many classes on this subject but it has been a long time and I need a refresher.
Thanks so much!!
SAugust 10, 2018 at 12:43 pm #202917
@regnells If you have Minitab the sample size menu is at the bottom of the Stats drop down.August 13, 2018 at 12:55 pm #202930
@regnells for your application, I believe the Acceptance Sampling tools will be a more direct way to get the sample sizes you need. They can be found in the Quality Tools sub menu of the Stat menu.
For a validation test plan, you just assume an infinite lot size, so leave that blank in the dialog box. Set the AQL to an arbitrarily low number (.1 should be fine). The RQL is 1 minus your required Reliability time 100 (e.g., for a 95% reliability, enter 5). Set the alpha much higher than beta to force the Acceptance Number to zero (alpha = .5 should be fine). Finally, beta is 1 – your required confidence (e.g., for a 95% confidence, enter 0.05).
If you use the Variable data option, you will always get a smaller sample size, but this assumes your data is normally distributed. The attribute data option will give you a valid sample size regardless of the distribution of the data.
Minitab’s help has a lot of very useful information that goes beyond how to use the tools and fill out the dialog boxes. It also includes information on interpretation of the results, examples, and even details on the calculations.
You must be logged in to reply to this topic.