Byung Kyu Park's Personal Website

It’s been busy 8 weeks, but the summer session is ending next week. I don’t know if the professor plans on grading the exam on his own again, but either way, it would be a … relief, in some sense, anyway.

I’ve been fantasizing about what I’d do when the summer session is over. I might re-subscribe to WSJ (or not; I’m still thinking about it; WSJ’s always had too much material for me to even skim through, even without the rigors of GSI’ing for a summer session). I might resume exercise routine (I’ve discontinued it for some time in the interest of what limited time I had during the week). Oh, and I might actually be able to take some of the weekends off—maybe for hiking near Lake Chabot again.

Well. Perhaps I shouldn’t count my chickens just yet, after all, I still have one set of review problems to come up with, two sets of homework to grade, and three sets of solutions to write.

And I only get 2 weeks of break, counting the welcome week as a “break”.

While I was thinking about my GSI duties for the upcoming 7A midterm, it occurred to me: once a grading rubric with enough detail is written, the grader is nothing more but a set of trained eyes—that is, anyone else with enough competence to spot mistakes and not make his own mistakes will end up assigning, for each exam, the same grade as I do, if we work out of the same grading rubric.

Perhaps this will be … relevant for properly judging the merits of the so-called holistic grading. One of the “merits” of holistic grading, even in the science courses, that I hear is that the resulting grade is more consistent—i.e. two graders grading using holistic grading assign more similar grades than two graders using standard method (i.e. grading by rubric). If there is any basis in this claim (as in, someone did a study and found that), I wonder if that basis is simply that of a bad study, in particular, poor control.

If holistic grading seemed to yield a more consistent grade, that may be more a result of holistic graders having a common guideline (after all, holistic grading does have guidelines, and for essays, “anchor papers”), while the graders-by-rubric did not—at least when we grade midterms, we write our own rubric, often with little central guideline common for all graders.

After all, how could a grading system which basically says, “I’m giving you a 5 because I think you deserve it, and I’m giving you a 3 because I think you deserve that, and no, I don’t need to justify myself to you—my personal judgment is better than any line-by-line justification,” can be more consistent than a grading system that requires that the grader justify every point being taken off?

At the moment, I don’t have the time to look for the actual studies which would either confirm or contradict the above scenario (i.e. the study supporting holistic grading had poor controls … or not), but I should … at some point.

And if this turns out to be true, then, well, the only benefit of holistic grading (at least the way it would be done for classes, with a single grader, i.e. the GSI, using the so-called “holistic grading”, rather than actual two separate readers, like they do for the standardized tests) is that it’s easier. But then, you know, cutting corners when building a dam is also easier. Does that mean the contractors should cut corners?