Tuesday, September 28, 2004

"When there are more than two groups the only sensible option for a parametric test is to use one-way analysis of variance.
It makes the assumption that the data is continous, at least approximately normally distributed and that the variances of each group/sample are homogeneous...."
Erm, I know it means something, and I know its something that I SHOULD know, but its like the book is talking in another language, possibly from another planet.
I've been wondering about it all since my tutor spent a good 45 minutes explaining it to me and talking me through it, but I've come home and I'm still lost.
I'm guessing that this is gonna result at some point in many hysterical tears, dramatic cries of "That's It! I quit uni and i'm off to join the circus/become assistant kit manager at Charlton/Tramp/whatever!" and I've got till the middle of November to work it out and get my I.S finished.
Ah sod this, The deadline is ages away. I'm gonna have dinner and go to the pub to watch Man U.
Meanwhile, me and The Brat are making plans to go to see the Mighty Addicks thrash Liverpool again!

No comments: