A facebook status dialogue - welcome to...
RAPHAELLE (me): After all this, I'm going to be dreaming about code. Nightmare or best dream ever? Hmmm, let's test that. Two sample t-test: are dreams about code better than dreams about everything else (when quality of dreams is on a 10pt scale)?
BEN: p>.05?
JOSLYN: nurds
ERIC: This has got to be the hottest status update you've ever posted.
RAPHAELLE: Ben- Traditionally p<0.05, although since a dream rating scale is soft (read: not absolute/scientific), perhaps I should say that p<0.1 would still be statistically significant. Joslyn- Why thank you. :) Eric- I love you.
ROSS: This is almost as good as listening to you two [referring to Eric & I] argue about philosophy.
BEN: My dreams go to 11...
ERIC: http://xkcd.com/670/
KATIE: I'd like to see some confidence intervals as this is a sample size of n=1 I think that leads to some validy issues. N of 1 randomized trials are no bueno no matter what p value you select. And no, this conversation is not hot.
RAPHAELLE: Haha, Ross. Good times. :) Ben and Eric, here is my response http://xkcd.com/552/ Katie, wiki says n of 1 trials can be effective and useful... and xkcd says there's someone for everyone http://xkcd.com/314/ ... I'm just on the edge of the normal curve. ;)
ERIC: I think #552 might be my fav xkcd evah.
Dec 4th