What trends are typical when Munsell testing a group of people? Are there effects due to age, gender, education, or ethnicity?
I haven’t noticed anything systematic like that in our testing over the years, but we haven’t looked for such effects either. You might inquire with the folks at GretagMacbeth about whether there are data on such effects (much of this sort of thing is described in the manual that accompanies the tests). I’ve never seen data on age vs. results, but as people age their lens yellows, so that would suggest that the blue colors would become harder to discriminate with age. However, a little more light in the blue end of the spectrum might counteract that effect. That brings me to one thought on your results. You might be seeing a systematic trend to more errors in the blue due to some aspect of your viewing conditions being different from normal (e.g., maybe your daylight sources are at a lower correlated color temperature or the luminance is low). There are definitely more errors at lower luminance levels.