Odd Sampling Behavior
- July 9, 2006 at 4:43 pm #43958
linParticipant@bill Include @bill in your post and this person will
be notified via email.
I have a known distribution of parts in a container that I use repeatedly for a classroom exercise. We have the students take samples from the container (using replacement sampling) to show people how to do six sigma calculations from sampled data. One thing bugs me about it however. The predicted defect rate of the process is very consistent from class to class, as I would expect. The actual number of defects drawn by each class varies a lot but is consistently far below the predicted value. I would expect that the actual defect rate would vary around the predicted one if the prediction is any good. This discrepency has been true for over two years. No one class would ever notice this but I do since I see all of them. I suspect there is a flaw in the way I’m conducting the simulation but don’t see what it is. Any ideas?Bill0
The forum ‘General’ is closed to new topics and replies.