Some individuals are biased by having good taste. That poses a challenge for schemes of aggregating opinions. More in my latest newsletter:
@tomstafford for grant application review the idea of coming up with a score has always seemed like nonsense to me for exactly this reason. I wonder would a series of yes/no questions aggregated into a score by a common formula be better? Like "the proposed research has no fundamental errors that couldn't be easily corrected". Perhaps not.
@neuralreckoning i think this would be more consistent across judges (and maybe even easier to do), but I bet some researchers would resent be "tied" to the formula, such they couldn't give a great proposal a great score, but had to enumerate the value point by point
@tomstafford we could let reviewers give a score and then not use it. 😉
@neuralreckoning ah. The "programme manager" approach!