Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I've had much the same problem as a TA as well. It's really painful having to go back to correct a few scores and sometimes I think I probably still missed something.

I actually made a web application for grading exams which alleviates this somewhat. Basically you grade by creating a rubric, which is a list of items and how many points to deduct for each item (could be additive too, but the professor I wrote it for didn't want that so I never implemented it). When grading you just check the boxes for the items that apply to a specific submission. You can change the weights in place and have all the scores update. Also, since all the exams are scanned and each question is cropped out you can much more easily go between instances of one problem. It's not completely releaseable to the outside world unfortunately but maybe in a month or so it could be. A lot of other professors have asked to use it but unfortunately we had to turn them down for now since there are still a few steps that are somewhat manual, and my time is currently prioritized for research work so I can't really clean up that last 10% and make it releasable right now.



That's not a bad system. However, I have to grade on the paper directly and I don't want a computer near me if possible when I correct. Too many distractions. I just grade each question then I add them all up if I predict the grade is gonna be bad or I subtract if the grade's gonna be good.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: