hi,
delta is the effect size, the difference between the means of the two populations. The second formula seems to be for a one-tailed t-test : https://onlinecourses.science.psu.edu/stat500/node/46
regards
I can't find reference to these formulas in the slides posted by my professor. I also can't find any versions of these formulas anywhere else online - guess my google-fu sucks.
I get what Z_alpha is (critical value), I know what sigma is (population standard deviation), and n I guess is the sample size. The hell is h? and w? and the delta?
http://i.imgur.com/9RgKmZ6.png?1
And what do these formulas achieve? Thank you for reading.
hi,
delta is the effect size, the difference between the means of the two populations. The second formula seems to be for a one-tailed t-test : https://onlinecourses.science.psu.edu/stat500/node/46
regards
leadgame (03-19-2017)
They're both sample size calculations. The first is when you know the population standard deviation (or are willing to assume a value for it), know what confidence level you want, and know how wide you want the interval to be. They call the width of the interval w. Sometimes people instead characterize this using the half-width (half the width of the interval) and they call that h.
The second does look to be a sample size calculation for a two sample z-test.
I don't have emotions and sometimes that makes me very sad.
leadgame (03-19-2017)
Cheers. How do I mark this thread as answered?
The link answered a lot of other questions I had as well, thank you!
Tweet |