A lesson in statistics and feedstuff variability

 Resize text         Printer-friendly version of this article Printer-friendly version of this article

Editor's note: This Practice Builder was contributed by Dave Mertens, of Mertens Innovation and Research LLC in Belleville, Wis. Mertens is a former research dairy scientist with the U.S. Dairy Forage Research Center, who is now consulting with scientists, laboratories, and nutritionists to improve the nutritional evaluation of feeds.

As a nutritionist, you use statistics every day. When you take 20 bale cores and mix them together to create the composite sample that you send in for analysis, you are doing this to reduce the variation in the results you get back. Intuitively, we all can understand that a single bale core may not truly represent the 200 tons of hay in that stack. This is true because there is variation among every bale core that is taken. The analytical result we get back from the lab is the average of the 20 cores in the composite. We all easily grasp the value of the average or mean of a measurement, and readily use it to formulate rations. However, there is another statistic that you should understand more fully — standard deviation — which measures variability in analytical results. You can use this statistical tool to evaluate feeds and formulate rations more accurately.

Oftentimes, we don't want to confuse ourselves with the fact that if we send in two samples of the same material, we'll get two different numbers. Just like two bale cores (or two silage grab samples) aren't the same, neither should we expect that two samples (or even splits of a sample) would give the same result. Some variability is natural and unavoidable, but we can use our knowledge about the size of the variation to reduce our risk of incorrectly pricing hay or formulating a ration.

Every stack of hay or silage has a true value for protein, fiber, etc., but each sample we send for analysis is only an estimate of that exact value. The only way you are ever going to get the exact value is to sample every pound of feed, which isn't feasible. But if you take multiple samples, you can obtain an average that is closer to the true value AND you can determine the variation (standard deviation) of that average. The standard deviation (StDev) gives you some idea of how close to the true value your average might be!

Averages and StDev are simply tools to help make decisions. Measuring variation is the key to assessing how close the measurements taken are to the real true result and the key to determining the risk of making decisions.

Comments (0) Leave a comment 

e-Mail (required)


characters left

Farmall® 100A Series

From field to feedlot, you need a tractor that can multi-task as well as you do. Case IH Farmall™ 100A ... Read More

View all Products in this segment

View All Buyers Guides

Feedback Form
Leads to Insight