The most important keys for effective research are replication and randomization. A replication is a repetition of a group of treatments and randomization is choosing at random the pattern of the treatments within replications. Both of these elements are important to reduce the effect of field variability skewing results, increase confidence in your data collection, and provides more data throughout the field. Many farmers split the field in half to compare treatments, however this does not account for soil type changes, low spots, hills, etc. Randomization is especially needed to ensure that one treatment is not unknowingly favored over another. We require at least 3 replications and encourage using 4. The digital ag team has some pre-designed layouts in eFields that can be used to build your own study from, and can be found at this link https://digitalag.osu.edu/efields/get-involved/study-implementation.
To make your on-farm research process easier, remember to evaluate your equipment and know how many planter passes are needed for the header to collect data. If using more than one pass, taking the combine through the center of two passes can also eliminate treatment carryover, especially in a fungicide or fertilizer treatment. To measure yield accurately use either a weigh wagon, a grain cart with scales, or a calibrated yield monitor.
The information that is collected from on-Farm trials done with Extension educators can be published in our eFields annual report, show casing on-Farm research done throughout the state, giving producers and educators a platform to share their information.
Source : osu.edu