[eDebate] Question about MPJ systems
Thu Oct 18 10:10:42 CDT 2007
Labels are curious things. There is no ?functional? difference at all between ABCX and A+AB+X or between 1,2,3,4 or any other labels that you provide for 4 categories. I think the reason for the use of the ?grade inflation? labels is an artifact of the design of column labels in Bruschke?s site that are arranged to accommodate 9 different options, which I incidentally label 1-9 rather than giving any letter grades. Even there we can find an interesting psychology. If forced to use letters, I accommodate the old ABCX by using A+AA-B+BB-C+CX (with percentages that match up pretty well). But if you look at the nine labels on Bruschke?s site you?ll find A+AB+BC+CD+DF(X) (chosen because they best accommodate the 6-category system used at the NDT for several years). Does it make any functional difference ? NO. Does it make a psychological difference ? I certainly hope not.
The one other thing that we need to take into consideration is the percentages we assign to each category. As we top-load the categories having a higher percentage of A+?s, we have the apparent psychological advantage of having more A+?s. Of course, the meaning of any of those A+?s is reduced accordingly. In addition to the fact that ordinal pref reduces the incongruity between within-category and between-category differences, I also suspect that it gives us the most truth in labeling, permitting apples-to-apples comparisons rather than the multitude of labeling alternative we have at present.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mailman