- May 14, 2014
- Reaction score
The author of the video claims that his 72% vs. 76% difference is significant, but statistically his data is not good enough to show significance. If you do a tolerance analysis taking into account the possible errors/differences in volume measurements, grain & hop absorption rates, grain bill weight measurement, SG measurement, grain moisture content, and grain potentials, you find that most efficiency measurements at the homebrewer level are only accurate to +/- about 3% - 4%. By using the same grain in both batches, he eliminates the possible errors/differences in grain moisture and potential, but the other sources of uncertainty are still there. To claim significance with this small a difference in result, he would have to repeat the experiment multiple times.Ran across this YouTube video on Anvil mash efficiency. Pretty well spells out everything I was thinking that was causing low efficiency
Just because the result is not statistically significant, doesn't mean it isn't real. It might or might not be real - this experiment by itself is not adequate to determine that. Since there might be a real difference, you should probably run your own side by side experiments to see what results you get. Just remember, a single experiment is not likely to prove a difference, unless the difference is much greater than the expected variation from run to run.