Independence criteria for Z-tests and p-charts
February 22, 2010 at 8:15 pm #53313
Christian JParticipant@Christian-J Include @Christian-J in your post and this person will
be notified via email.
I’m relatively new to six-sigma, so forgive me if my question appears basic! I’m looking at a production process on which I am trying to reduce the number of defective items. I would like to be able to monitor the output of the process in some kind of control chart, and would like to create a hypothesis test to show that our improvements have had an effect. It was recommended to me to display the output via a p-chart (unequal sample sizes), and I’ve been considering carrying out a z-test to test the hypothesis that the proportion post improvement is better.
The thing that has me stuck is the need for independence. Obviously for production processes each of your values is not independent because it is impacted by what comes off the line before it. Reading up that would seem to discount both the p-chart and the z-test as both seem to require independence.
If that is the case, my next question is what am I able to use?
Thanks in advance!
The forum ‘General’ is closed to new topics and replies.