#1
can anyone give me a step by step on how to do a t-test (with the help of excel obviously). I've done it at school but can't remember how to do it.

Also, can some one explain Standard Erro to me and why, if at all , it is needed in a t-test

Thanks
"Like a midget at a urinal, I was going to have to stay on my toes"

"Like a blind man at an orgy, I was going to have to feel my way through"

#2
ugh. i hate t-tests...
all you have to do is have your two samples of data in two separate columns in excel.
then you type into any empty box "=ttest" and click on where it says "TTEST"
then you select the first lot of data
then press the comma key
then select the second lot of data
then the comma key

then look up ttest with the F1 help function to determine what to put for the next two things. it will change depending on what sort of result you want. it will most likely be 1, and 1. in fact, i dont think i've ever used anything different (maybe 1 and 2 once...).

edit:
the standard error is how far away from the true value your estimate is across a data range. sorry, i missed that the first time i read your post.
Quote by Machanix
We play guitar.... we're automatically on top of the world.

Life Is A Lemon And I Want My Money Back!

A Wasted Youth
Is Better By Far
Than A Wise And Productive Old Age

Good Girls Go To Heaven, But The Bad Girls Go Everywhere!
Last edited by instagata0 at Jul 21, 2009,
#3
Thanks for the reply, but the thing I need help with is the critical values from the table and the things after the test is actually done. (like the p values )
"Like a midget at a urinal, I was going to have to stay on my toes"

"Like a blind man at an orgy, I was going to have to feel my way through"