West Central Iowa | shwaga - 1/30/2025 12:32
To make interpretation easier, we display the difference from 100% rather than the full percentage. Some users found it difficult to understand how a value could be above 100% (e.g., 102%, 104%). By showing just the difference, we present a simple deviation from the trial mean (-2%, +2%, etc.). We went back and forth a lot on how best to display these metrics, and while we're still evaluating if relative yield is the best approach, it does allow data to be compared across environments—which is our main goal.
Regarding graphing no-till vs. tilled with side-by-side bars for each hybrid—if varieties are tested under those practices, they will display as you described. The data shown here only includes trials that tested this tillage practice, but if a conventional till practice were included, that would also appear in the results. Essentially, it pulls from the management practices within the trials you input.
Curious—do you typically analyze data like this? If so, how do you normally do it—Excel, another tool, or just field observations? Always interested in how others approach this!
Appreciate the feedback!
Were you using upside down bars for below average entries? I wouldn’t put upside down bars on anything even if something is 98% of average, also if it is labeled as a comparison like no-tilled and tilled if no-till is not available I would put a blank where the bar should be and an N/A. I know it has to be hard to fit everyone’s preferences for charts, but I am good at reading charts and what you have there is not a good chart, and the labels don’t technically work.
I like to look at variety/hybrid I’m going to try in a plot, along with looking at trial data, and talking to people I trust that have grown or represent the product. Then trying the product on a small area often split planter for corn. We also breed some corn and are close to having some products. We are in the process of switching over to gblup and other statistical methods via iPat. |
Seed Trial Insights and Tools: Let’s Discuss
