Hacker News Reader

Comment by noodlesUK | Hacker News Reader

At my university, almost all of our marking was pseudonymised. We were assigned a random candidate number at the beginning of each year, and that is what went on our important papers/exams. The less important coursework often didn’t bother with this, and used our student numbers instead, but the general idea was the same.

We didn’t put our names on any of our work other than our dissertation (and a few trivial assignments that didn’t impact overall marks). It wasn’t that hard to de-anonymise, but it meant that the system had a bit more integrity.

It’s a really straightforward system to implement and I don’t know why it isn’t done more frequently.

I also think that our VLE sorted assignments by time of submission rather than any identifier.

View on HN

Wouldn't a possible outcome here though be that it just randomly reduces grades instead of reducing them in a way that's related to the students? If the issue is the sorting the random candidate numbers would still be sorted. It solves the problem of bias related to the individual but it doesn't solve the problem of bias related to the way that the submissions are sorted.

A random identifier coupled with a random sort order seem like the way to go here.

University exams, this probably makes a lot of sense. After all, the exam is the exam and whether a student is well-spoken and actively participates in class shouldn't matter for an exam grade. I'm less convinced that blinded conference proposals are a good idea--an argument I've had with various people. If you know based on past experience that someone will almost certainly hit a home run, I'm less inclined to pick a random person without obvious qualifications for the same topic--although just picking friends of the committee can obviously go too far.

You could try to work around that by first grading all anonymized proposals, then grading all potential speakers without knowing their proposal. In the third round you deanonymize and look at the weighted average of the two grades. You probably still need some judgment calls because the combination of speaker and topic can be important. But the score would give you a good base to work of.

Maybe you could make it even more impartial by allowing conditional scores in the first two rounds. Like "Jim is a 6, but a 8 if his talk is about molecular biology" or "this Lessons Learnt talk is a 5, but if it's by X, Y or Z it's a 9"

Yeah, but I'm not sure conference proposals by themselves actually have a lot of value given that, in many cases (ask me how I know), the presentations won't actually exist until week or two before the the event.

Certainly a talk by X that's totally unconnected from anything they're directly involved with has less value.

I had classes like that, where at the beginning of the quarter, each student gets assigned an username of the form "<course id> <three alpha characters>" and all participation is based on username from then on. Even though the usernames are seemingly random, certain usernames started gaining reputations on the class discussion forums, and students come to recognize some names.

But computer science courses tend to have very objective rubrics for grading, so I am not sure the anonymity mattered much.

I think I get better feedback when the teacher knows who I am. Grades are secondary.

I'm not sure exam grades at the university level are really the place to get useful feedback beyond grades.

I think the point is that some automated systems like Canvas may hide the names, but they're still presented in alphabetical order. Pseudonyms don't help if you don't shuffle them.

Reply on news.ycombinator.com