I have the following problem: An engineer want to estimate the proportion
of defective light bulbs in a very large shipment of light bulbs. A small
sample of 5 out of 50 is defective. The cost of sampling of each lightbulb
is small whereas the setting up of the inspection process again is
expensive. How many light bulbs should he sample so that the estimate of
the proportion of defective items to within 0.04 with 95% confidence ?
My question is should I use 0.5 as p (the conservative method) in the
sample size method bearing in mind the cost of setting up
another inspection. This will give me 601 or should I use what books call
the educated guess with p = 5/50 in the sample size formula and get 217?
--
Thanks,
Jim.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|