|
Evansville, WI | There is nothing wrong with what you are doing. I have my own on farm plot as well to see how the new numbers looks next to proven ones in my system. What you are doing is smart in that you proceed slowly with what you see as potential new vatieties.
Not to put words in Tom's mouth but his point is usually because someone claims the results one side by side as fact, when it easily could be just random variation. Usually also associated with someone claiming that everyone should do it that way. Tom is a crop consultant so when he recommends something to one of his customers he likes to be reasonably sure what he is recommending is accurate. I come from a similar background, so I am pretty cautious about making recommendations, that said there many things I do on my farm that are not the result of replicated trials.
The reason that replication is helpful is to increase the odds that what you are seeing for differences in yield are not just due to random variations therefore replication increase the odds that what you see for differences is repeatable. Obviously this is not practical to do on farm variety trials is this manner, something like fungicide vs no fungicide is a lot easier.
An example how it might be useful is say you had a variety do really well in your plot. Unbeknownst to you it ended up on an area that got a double dose of manure a couple years ago so the extra fertility let it really shine, if you had replicated the trial, it may have averaged that one higher fertility strip, and might have shown that hybrid to be just average at best.
I have read a lot of your posts and pictures you have posted here, so I have no doubt about the success you and your brother have had, as your post always add a lot of valuable information about what works for you in a situation.
A couple if good ideas I got from you are the fat duals with mud scraper, and I really wish I would have saw your 22 inch rows before I got into and fed up with 20s and went back to 30s.
Edited by nviney 11/18/2017 15:26
| |
|
yield plots with statistically significant differences
