Farm Progress is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Serving: IN

No Replications Means No Real Conclusions

No Replications Means No Real Conclusions
Trial confirms need for replication.

Some farmers scoff when an Extension person asks if they would do a plot for them. The farmer knows he's probably going to want it down twice, and that he must follow very picky details. Nevertheless, some inquisitive farmers are working either Bob Nielsen, Purdue University Extension corn specialist, or local private consultants to establish trials on their farms. What the farmer gets out of it is a chance to see how the hybrids, varieties and treatments interact on his own farm.

Many ask why not just do a strip trial and include each treatment once? The Indiana Prairie Farmer/Precision Farming trial at Throckmorton Farm near Romney, a Purdue research farm, made it very clear why one strip trial isn't enough.

"The bottom line is that when you get done, you don't really have anything if you just planted every treatment once," says Jeff Phillips, Tippecanoe County Extension ag educator who helped organize and direct the plots. "You've got some yield numbers, but you can't do anything with them. That's because you don't have anything to compare them to."

What Phillips is driving at is that you don't know whether one treatment that yielded better was because a treatment actually worked, or if that treatment happened to land on better soil. Or maybe none of that plot was in an area affected by early-season ponding, while another plot was affected.

To be able to draw conclusions, Phillips needs enough data to run statistics. Then you can say within 90% probability, if you run at a least significant distance of .10, that the difference was due to the treatment on the one plot, and not just due to chance or location on the trial grid.

The other factor is that when you have more than one rep, you discover that sometimes a treatment will rank higher in one rep than another. Again it may be due to soil conditions or topography. But if you hadn't replicated, you could have drawn the wrong conclusion from only one treatment planted only once.

Even as it is, Phillips says, the results are only for this year's conditions- planting late into dry soil. That's where multi-year data would be helpful. Planted earlier when soils were cooler may have made a big difference in results. The only way to find out would be to repeat the experiment in a cooler, wetter spring

TAGS: Extension
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.