Physics 4410, Spring 1998: Peer Grading

This semester we will institute a new form of homework grading. Each week when the homework set is due, I will randomly pick one of you to be that week's grader. You will be responsible for producing a clear, complete solution set, which will be copied and put in the library. You will then grade everyone else's homeworks that were turned in that week. You will have one week to produce a solution set, and possibly 2 more days to finish grading. This time schedule will be strictly enforced, since your classmates will be counting on getting their homeworks back in a timely fashion! You will only be assigned to be grader once during the entire semester, and will not know in advance which week you will be grading. I will of course always be available to help you if you need it, either in producing your solution set or making any difficult grading decisions.

You will be excused from grading your own homework. (Since you also get to drop your lowest homework, that means that two homeworks are ultimately dropped.) I will give the grader a grade based on the solution set they produce. This grade will count as roughly four homeworks (i.e. one third of your entire homework score for the semester) as an incentive to produce a careful, readable and timely solution set!

You will get slightly more detailed instructions the week you grade, but the particular grading scheme you pick is up to you; most "fluctuations" in grading styles should average out over the semester, as long as you are careful, consistent and fair. Your grading will not be anonymous - anyone with complaints about their score will be asked to talk with you first, and hopefully any possible mistakes can be resolved fairly without even involving me.

This is a novel system, and I encourage feedback and suggestions. It will be a lot of work for you, and involves some serious responsibility, but it's only one week in the semester. (Don't forget, the week you're grading you still have to work on your next problem set, as usual) I suspect that you will learn this one assignment inside out; years from now when you've forgotten most of atomic physics, you'll probably still remember how to solve that one set of problems! For those of you thinking of grad school, experience with grading will prove a valuable skill as well.



Generic Grading Instructions, Physics 4410, Spring '98

You are assigned to grade your classmates' homeworks this week. You have 5 days to come up with a complete, readable solution set (i.e., assuming the set was due Wed, your solution set should be done by this Monday.) You must first come see me by Friday afternoon (i.e, just 2 days after you get them), so that I can look it over and ensure there are no serious mistakes.[1] There will probably be a couple of minor corrections to make; you then have 2 more days (i.e, over the weekend, until Mon) to produce your final version of the solutions, which I will copy and put in the library. Please sign your solution, so people know who to direct any questions to.

Your classmates (and I) would appreciate getting the graded homeworks returned by this day (Mon) too. If it proves too much work, you may take 2 additional days to complete the grading, meaning homeworks absolutely must be graded by Wed, 1 week after they were turned in). [2]

You will be evaluated by me on your solution and grading. It counts 1/3 of your total homework score for the semester, almost as much as a midterm! Your grade will be based primarily on the completeness, correctness, and readability of your solution set. Severe point penalties will be taken off for being late on any of the deadlines mentioned above.

You may make up any grading scheme you like. For most problems, you should decide how to distribute points to different parts in a fair way. Don't try to grade too "finely"; something like 3-5 points per "lettered part" of a problem is usually enough to clearly distinguish strong, partial, and poor solutions. In the end, total up the points and record the score as a percentage of total possible. Keep a copy of your "grading scheme" after you're done - if someone complains that you've graded unfairly, it's important you be able to explain how/why you did what you did. On the back side of this sheet is a list of names - please fill it in completely and return it to me. There is space for comments if needed, generally you won't need to fill that in. This week involves a lot of work for you, but remember that your classmates are counting on it, so do your best!

Steve Pollock - x2-2495 (campus), -3764 (NPL),

Email me with questions

4410 main page Prof. Pollock's page. Physics Dep't
Send a message