dean, for what it is worth, i have had some experience in the design, execution, and publishing of a handful of scientific studies, (although that was several years ago).
after reading your suggestions for a design in your previous post, it is not clear to me exactly what hypotheses your are trying to test.
my take on the small question, is that there is interest as to whether or not regressed bees have an advantage over nonregressed bees in resisting collapse from varoasis.
if this is the prime question, my initial thoughts on a study would be these:
the experimental variable would be regressed vs. nonregressed bees.
the control variables would be everything else. i.e. type of hive, comb material, feeding or not, genetics, manipulations, and anything else that would mean managing the two groups as identically as possible.
both groups should located nearby geographically to control for differences in weather and available forage.
treatments should not be part of the protocol, because we would want to know specifically what difference cell size makes.
the measure would be regular mite counts, and ultimate success rate in terms of survival.
controlling for genetics would be challenging, because it might be hard to get genetically similar colonies that have are both regressed and not regressed. it might be easier to use already regressed bees for the experimental group, and shake out genetically similar bees onto standard cell for the control group?
it would take someone versed in statistics to figure out an adequate number of colonies needed for both groups to make the data meaningful.
and, it may take a couple of years of observation.
this is my first go at the thought process, i'll keep thinking about it.
and, i'm good for $25 as well.