One of the ways you can reduce bias in observations is to have someone else
read the data. In that way He/she does not know what the treatments are.

We used to do this in turf trials for grubs in Ohio. The University would
put out the treatments, in August all the commercial technical reps would
evaluate the plots by digging grubs and recording the results. I will tell
you that this method kept you honest.  If you dug "your plot"
you would not know it. You did not know if it was your treatment or your
compactors. Honesty was "enforced" by this method.

We all awaited the final results by the end of the day. If the plots you dug
were the poorest, the bias would be upward, unless you thought it was a
competitors treatment, if the best, could it be your competitors? This
worked great as no one knew the treatments until the end of the dig. Final
data was statistically analyzed and the final report was forwarded in about
a week

Dan Veilleux
Boone area
In the Mountains of NC

-- Visit www.honeybeeworld.com/bee-l for rules, FAQ and  other info ---