Dear All,
Maybe someone could help me with this:
I'm trying to check that there is no error values in my database. Error values is combination of values that I can not check in whole database. It contains 12 million records. Then I want to take a sample or samples and test the point estimation of a population proportion:
Estimate of p=X/n
where p is the probability of finding a error, X is the number of records with error and n the size of the sample.
However Can I use this when probably I don't have any error value or at most very few? It's very likey that I will get X=0.
Do I need to use another method?
Many thanks,
Kind regards
Jorge
---------------------------------
Yahoo! Plus - For a better Internet experience
|