Posted 29 August 2010 - 01:45 PM
I read the Chapter about the distribution of sample proportions (starting on p. 458) and solved the Excercise. The goal of the excercise is calculating the probability for having more than 40% desired items in a sample if the expectation is 25%.
I found that one gets the same solution if one does it the way I already knew from the previous chapters:
I assume a binominal distribution ~B(100, 0.25) and estimate the result using the normal distribution X~N(n*p; n*p*q) = X~N(25, 18,75) with an applied continuity correction.
If the goal of the capter is to make calculation like the one in the exercise, why does one do it in a different way than?
These things made me think a lot but I could not find a satisfying answer - but maybe you can (?)