cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
nac
nac
Level III

determination of sample size

Dear All,

 

my sample population of each batch is 810.000 tablets for pharmaceutical product. I analyse 7200 sample for each batch. I want to reduce my sample size. Is there a statistical approach for this? 

 

I try to use Menu, DoE, Design Diagnostics, sample size and power, one sample mean part of jmp

I do not change alpha value. my Std Dev is 0.3mg. My lower/upper spec is 1.3mg lower/higher than mean value therefore ı write it on difference to detect part. When I write >10 on sample size part power is 1. I am not sure sample size >10 is really simulation my population (population size:810.000) 

Do you have any recommend this? I am not sure, ı use correct approach or not?

Is there anyone can help for this?

 

Best Regards,

 

 

 

3 REPLIES 3

Re: determination of sample size

Yes, it does not take a large sample to decide if the mean shift is at least 1.3 when the standard deviation is 0.3 but consider this: do you only want to decide if the mean weight is out of specification? That is, would you want to know if the mean is off target but not out of spec yet? If so, then you would use a smaller difference. For example, a one sigma shift would be a difference of 0.3 in this case.

I bet that the minimum sample size is still quite a bit smaller than the current 7200.

nac
nac
Level III

Re: determination of sample size

I do not just only interesting mean value is inside or outside of the spec. I am interesting all individual results of my batch that means 7200 samples from my population (population size:810.000) must be inside +/-1.3mg of my target value. And std dev of 7200 individual sample is 0.3. I want to reduce sample size. I am searching a scientific way. Is there any approach that you can recommend by using jmp?

 

Thanks for your help

 

Best Regards,

 

Re: determination of sample size

The sample size and power calculators are for questions or tests about the mean of the population. They do not address questions or tests about individual observations from the population.

Hopefully other members of the JMP Community with more experience with acceptance sampling and specificiations than I have will also offer some direction or suggestions.