| |
BGonline.org Forums
Method of scoring a BG quiz
Posted By: Timothy Chow In Response To: Method of scoring a BG quiz (Jason Lee)
Date: Saturday, 29 September 2012, at 11:34 p.m.
Your scheme looks interesting. The main problem that I foresee is that skipping a problem doesn't seem to be quite "neutral," which I think was your intent. Say I'm a below-average participant. Once I recognize that I'm below average, I'm incentivized to skip problems that I sense are likely to be difficult, because I don't expect to be able to score better than zero. If weaker participants start dropping out, then the field will get stronger, and people who were average or even slightly above average will find themselves below average, and the effect could snowball. I think you'd want a scoring system to encourage participation, not discourage it. Note that the state of "no participation" is a stable equilibrium, i.e., if nobody is participating, then I have no incentive to be the one person to answer because I'm going to score zero either way. For this reason, I think you probably want to do something like give a small bonus to people who get the problem right—not so much that skipping a few problems is a disaster, but enough to give people an incentive to participate.
Here's another idea that is borrowed from Sid Sackson's game, Patterns II. You might consider devising a scoring scheme for the quizmaster as well as for the quiz takers. For example, maybe the quizmaster's score for a problem is the standard deviation of the quiz takers' scores. This incentivizes the quizmaster to come up with "interesting" problems. Of course, since the quizmaster is necessarily scored separately from the quiz takers, this really makes sense only if there is somehow going to be more than one quizmaster.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.