Grading Policy, Sir!

Dr. Pion’s blurb about exam design and grading reminded me of a few things. I taught for the Navy in the nuclear power program, long ago, when the school was in Orlando, Fl; physics, which included applications to plant operations whenever possible, and a class on principles of reactor operation. There are some distinct differences between exams in this context, vs how they were graded when I was a TA. Being in the military means never having to say you’re sorry to students or even explain yourself to them, if you don’t want to. That translates into not having to post or explain the grading policy on exams; the students’ job is to ‘learn the material, dammit,’ not to haggle for points on exams. Students could still put in for regrades, but it had to be for an obvious grading error, rather than for a dispute about how many points should have been deducted for their mistakes. That didn’t stop all whining, but it’s certainly a bonus when you can tell the offender to shut up — in navy parlance, “Secure that!” (or, “Secure that shit!” Optional for officers, pretty much mandatory for senior enlisted)

Since the material had a definite application, answers to questions had to display an appropriate level of understanding, which was a factor that could supercede any other policy that had been set up. There was a shorthand for the various types of errors — the usual suspects, like math errors (ME) or sign errors (SE), and the big red X for anything wrong, but there were others, too, in part because there were always several “discussion” problems, even on physics exams:

CE and GCE: Conceptual Error or Gross Conceptual Error, losing at least half of the points of the problem, and possibly all points. This included grossly unphysical answers, like solving for reactor power and getting a negative number or something several orders of magnitude too big or small. If you noted that the answer was unphysical, your calculated answers often reverted to simple math errors. A colorful variant of this (that I never used; it was more popular with the senior enlisted staff) was DAF: Dumb as Fuck.
BOD: Benefit of the Doubt. An ambiguous answer that could be right or could be wrong, but contains some correct information. Not something a wise student would put in for a regrade — BOD goes out the window.
S/G: Shotgun. Blasting a problem with as many answers and/or equations as you can, hoping that something is right, hoping for partial credit. But you get none, because you haven’t demonstrated anything beyond rote memorization.
DR: Dead Rat. This name was an evolution of the dead mouse principle — the student played with an answer too much and ended up killing it. Usually adding in extra tidbits that were wrong or contradictory, not recognizing that they had answered the question already.
ABA: American Basketball Association. Your showing is weak and you need to merge with another student to pass. Answer by accident (also RAWR: Right Answer, Wrong Reason, for those who tended to growl) As indicated, the work or reasoning is incorrect, but somehow the correct answer was obtained.

And there may be something i missed, because this was all many beers ago.

Update: ATQ or ATMFQ: Answer the (MF) Question! You did everything just fine except actually answer the question that was asked.

Thanks to some former cubicle-mates — BDJ for reminding me of that last one, and RTS for the prior entry.

0 thoughts on “Grading Policy, Sir!

  1. I’m tempted to start using this notation when grading my tests. I’m afraid I’d end up getting carpal tunnel writing s/g too many times, though…

  2. Glad you found that article entertaining. Depending on just how long ago that was, I might have one of your grads in my physics class this year. Ex navy nuclear propulsion. To no ones surprise, he never asks about grading. … and was quite interested when I told him a prof in the nearby E school was also ex nuclear.

    I really need to adopt that kind of grading shorthand, especially for sig figs on lab reports.

    I notice that you convert a GCE into an ME if the student notices the answer is nonsense. I do the same thing, since the biggest difference between an exam environment and a professional environment is that you have both time and colleagues to help you find some error that eludes detection by its creator.

    My tactic on s/g answers once was to average the grades for the separate solutions if one in particular was not identified as being the student’s final answer. (10+0)/2 = 5 of 10. These days I also subtract points for not indicating the correct answer. In engineering school they get zero if they don’t get a single answer and put a box around it.

    One of our math profs used to teach in the Air Force. His comments are mostly about attendance and classroom behavior, which are er slightly different at a CC.