Roger Andreiuk, an agronomist with the Reduced Tillage Linkages group in Alberta, presented the following steps to develop and set up productive on-farm research plots during a recent Direct Seeding Advantage conference in Nisku, Alta.
- Goals and objectives of testing: Determine what you want to test. This should be something that you need to know before you adopt a practice or product on your farm. Decide what you need to measure to get the answers you require.
- Treatment and check selection: Keep it simple when selecting treatments to test. Select only one or two and be sure to include a check or control treatment. The check is your standard current practice against which the new technology will be compared.
- Site selection: Consider field variation when picking a site. The previous history of the site, including herbicide and fertilizer use, previous cropping patterns, old yard sites and manure applications can all have an effect on crop growth and site selection.
- Design and plot layout: If only two treatments are to be compared, such as fungicide versus no fungicide, a simple strip of each treatment may suffice, with replicated measurements taken.
Read Also
VIDEO: Green Lightning and Nytro Ag win sustainability innovation award
Nytro Ag Corp and Green Lightning recieved an innovation award at Ag in Motion 2025 for the Green Lightning Nitrogen Machine, which converts atmospheric nitrogen into a plant-usable form.
If comparing more than two treatments, a simple design for uniform land is the completely random design. This design allows all treatments an equal chance of being assigned to any given plot. Width of each plot should be somewhat larger than the width of the swather or combine header so that you can harvest a full width down the middle of the plot.
- Record keeping and data collection: Keeping records throughout the season is important because unexpected results can often be explained by conditions during the growing season. These records should include soil tests, weather records, insect and disease occurrence and crop development. Yield measurements should be collected on each plot in the experiment.
- Data evaluation, analysis and interpretation: This is the last step in the on-farm testing process, but the method of evaluating results should be planned before starting the experiment. Learn the lingo of statistical analysis and know what the numbers are saying so you can make a good decision.
The following websites may help with information about statistical analysis. The Agstats program can be used to evaluate results.
- www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/sag3024?OpenDocument
- pnwsteep.wsu.edu/tillage handbook/chapter10/100999.htm
- yolo.een.orst.edu/steep/AgStatsWeb/
- Partner up
Find someone else to work with you on the experiment. Often another farmer, researcher or company rep can offer new ways of looking at on-farm tests. Partners may also bring computing and analysis power to the table, as well as help in the field.
