Suppose you manufactured something, toothpaste tubes for example, and one in 10 of them (10%) had problems i.e. a hole in them. If you changed the manufacturing process and the failure rate dropped to 8%, how would you work out if that was statistically significant or not?
I know how to do the t-test but for that you need standard deviation etc...How do I calculate whether a change in rate is statistically significant?
This distribution is called the sampling distribution of xbar . This is a new data set composed of x bar檚. These new data points have a mean and a variance, 未2xbar.
未2xbar = xbar/鈭歯
where n is the size of the sample taken from the population.How do I calculate whether a change in rate is statistically significant?
If p is a proportion, the mean = np, where n is the sample size.
Standard deviation = SQRT [ np(1-p) ]
You can then use the t-test.
Subscribe to:
Post Comments
(Atom)
No comments:
Post a Comment