I was playing with the idea of integrating CV as part of an advanced statistics course/workshop; Say, the students are required to achieve a reputation of 100 by the end of the semester. The students would get experience in applied problems and a sense of community, and the community would benefit from some (hopefully) good answers. I was wondering if anyone had any experience with this? Any pitfalls to be aware of?

This sounds terrific for a graduate or advanced level course where the students are intrinsically motivated and deeply interested in stats. +1 –  Seth Rogers Nov 10 '11 at 16:23

3 Answers 3

I am sympathetic to the general idea of capitalizing on SE as a teaching tool but am concerned about the potential for abuse in this case. Two people acting in collusion can easily generate thousands of points of reputation for each other merely by writing a lot of repetitive junk and upvoting each other's posts. We can catch that behavior, but if the students are smart--and many are--they can mask this behavior through extensive voting for others as well, making automatic detection difficult.

This is not theoretical speculation. It has happened (and, in one notable case, is allowed to continue because it's benign and the perpetrators appear to have a net positive effect on the site, although as a result one of them has a noticeably inflated reputation).

Requiring participation also threatens to reduce the quality of posts, especially towards the end of the semester (when demand usually goes up anyway), as procrastinators ever more desperately throw stuff on our proverbial wall hoping to reach the required minimum. As a moderator I don't welcome the additional janitorial work that would entail and would look extremely unfavorably on anyone I suspected of instigating this. In the worst case, you could wind up having no access at all to CV within your institution.

I'm sure there are creative and constructive uses of this site for pedagogy, in addition to the obvious one of offering it as a resource for learners. If you want to go in this direction, consider making participation optional and offer extra credit for answers, comments, or questions that in your opinion are especially worthy. In short, please do not delegate to this community your duty to motivate and assess the performance of your students.

One could use other aspects besides reputation as well. Such as a quality question/answer (based on evaluated content as opposed to upvotes, same as whuber said). To @JohnRos, how about you act a filter of sorts. Have the student submit the question/answer to you first, and then encourage the student to post that themselves if you approve and it is not redundant with other questions/answers. –  Andy W Nov 10 '11 at 16:20
+100, if I could. –  mbq Nov 10 '11 at 17:25
I am curious who are these perpetrators? How were they detected? –  mpiktas Nov 12 '11 at 14:16
@mpiktas It's probably best not to provide details of the detection mechanism. Suffice it to say that some cross-checks are routinely run whose results are visible to moderators. We don't have the ability to see any details about votes, but clearly that information is available to SE technical staff. –  whuber Nov 12 '11 at 16:58
@whuber: Another way to go about, would be to send the students to find interesting problems in CV to discuss in class. Except for the tag, how would you recommend "fishing for problems"? –  JohnRos Nov 13 '11 at 13:08
@John, the best way to fish for problems is just to read over the new problems regularly. (I do this with math@SE for a problem-solving course I teach; it turns up about one good problem a week.) –  whuber Nov 14 '11 at 2:31
@whuber: Pure curiosity, but what are the breadth of topics covered in the problem-solving course? –  cardinal Nov 14 '11 at 3:06
@cardinal Anything related to math, including (on occasion) some computer science, physics, and economics applications. We meet once a week for an indefinite period that usually stretches to 5 hours. Students often bring in interesting problems, but if not, I keep a collection of good ones and we can always look at old Putnam exams :-). After they create a solution on their own, I encourage them to find alternative solutions and then I draw connections among various branches of math that might be implicated, as a way to help them unify their undergraduate studies. –  whuber Nov 14 '11 at 3:39
@whuber: Thanks. It sounds like you're creating quite the stimulating atmosphere for your students. That is wonderful! –  cardinal Nov 14 '11 at 9:31

Aside from the (very valid) concerns from @whuber, it's still a great idea if one could this right. For example, an assignment could be to ask one good question and to give one good answer on CV. This would do away with the incentive to spam the site. The instructor / TAs would have to evaluate the questions and answers on their own to do away with the incentive for students to upvote each other.


John, you can additionally consider requiring your grad students to contribute to Wikipedia. If you see that their edits have withstood the criticism and additional edits of the community at large, that's a good sign that your students have mastered the material. (I do list the Wikipedia articles that I contributed to on my CV, although I don't know of anybody else who does that.)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .