Quote:
|
Wow, that's a hell of a post Henry. Definately more to think about than I was aware of. I understood a larger scale operation would give better answers and all, and I am not large scale at all, and the big guys don't seem to talk about it much. LOL But there is a lot I didn't concider. Thanks for the eye opener.
|
No problem. Experimental design can be extremely difficult, and in recent years has only gotten more complicated as we've found new problems that can ruin results; it's to the point that we've had to invent entirely new statistical methods for dealing with some questions.
Primary, though, is always the issue of sample size, and in breeding experiments, lineages are the proper sample size; no matter how many breedings there are, the results are always statistically non-independent (since, after all, one generation gives rise to the next one).
Quote:
|
I'd love to read the PDF's on the subject though. rick@sunsetbci.com Feel free to send away. Can never learn too much.
|
PM me to remind me if I forget; I'll have to wait until I'm back in lab tommorrow, as I don't have access from home to the full-text articles.
Henry