A team of Extension specialists has put together a guide to interpreting crop research results. The team includes Lizabeth Stahl, University of Minnesota; Sara Berg, South Dakota State University; Josh Coltrain, Kansas State University; and John Thomas, University of Nebraska-Lincoln. The team members recently answered questions about how the phrase “not significantly different” is used in research reports and tables, what it means, and how you can use it to make smarter decisions on your farm.

Dakota Farmer: If you see that one hybrid yielded 5 bushels per acre more than another, how can it not be significantly different? Five bushels per acre added up over many acres can be a lot of money!
Extension specialists team: The result has to be from a statistically sound trial to conclude that a yield difference is due to the difference between the hybrids and not some other factor or factors, or even random variability. We look at what what’s called “the least significant difference,” (LSD).

What is LSD?
In a hybrid variety trial, it is the minimum bushels per acre that two hybrids must differ by before they can be considered “significantly different.” The formula is complicated, but factors that go into the calculation include the number of replications, and randomization.

If you split a field and plant different hybrids in each half, can you determine the LSD between two hybrids?
No, there is no way to calculate the LSD if you simply split a field in half and put one treatment on one side of the field and a different treatment on the other. In this scenario, you have no way to sort out if a difference in observed yields was due to underlying factors such as soil type, planting population, drainage, compaction, disease, insect pressure, harvest issues, topography, etc., or the treatment.

What does an LSD number mean?
When you see the LSD calculated at the 0.05 significance level, this means we can be 95% certain that the treatments (or hybrids, etc.) really did differ in yield if the difference between them was equal to or greater than the LSD. A significance level of 0.05 or 0.10 is most commonly used in agricultural research.

How do you end up with “no significant difference?”
This can occur when there simply isn’t a difference in yields, or when there is so much variability in the results that you can’t make a conclusion with confidence. For example, in an actual University of Minnesota trial recently, strip tilling outyielded moldboard plowing in the first replication by 13 bushels per acre, but moldboard plowing outyielded strip tilling by 14 bushels per acre in the fourth replication. Chisel plowing was the lowest-yielding treatment in the second replication, while it was the top-yielding treatment in the third replication. Due to this variability, statistical analysis revealed we couldn’t say with confidence that any one of the tillage systems resulted in a higher yield than another. Other factors we couldn’t account for appeared to have impacted results at this site, highlighting the value of conducting research over a number of locations and years.

How do you use the LSD?
Resist the temptation to put economics to yields if they are not significantly different. Doing so could lead to poor and costly decisions in the future.