Information costs and student deaths-in-the-family

I enjoy playing poker; it satisfies a lot at one time. Competition. Interaction. Deception. Strategy. Asset management. Game theory. There's just a lot there.

I've got a (likely bad) habit of translating interactions in life to similar situations in poker. It can only get so far, of course, but it does help me rationalize decisions ex ante. I feel better about outcomes good or bad if I know I made a well-rationed decision.

After reading Justin's piece on cheating in the classroom, and a once-off line concerning the seemingly well-timed death of students' family members (very last line) from our good friends at Division of Labour, I got to thinking...

In poker-- especially in games with people you don't know-- if someone bets at you, sometimes it's worth matching the bet even if you're fairly certain that you don't have the best hand, just to see what hand they are playing. Poker is a game of information, and information is scarce, so if the price is right, it can be very valuable in the long run to acquire this information. In part, this information helps keep people a bit more honest throughout the game as well.

Along these lines, isn't there an optimal amount that I need to "call down," or challenge the claim that the student in fact does have a relative who recently passed? And isn't this percentage strictly greater than zero? By never challenging a student, they know they can always get me in a position of never questioning their claim-- get me to fold, in poker terms. At some point, somewhere, it has to be beneficial to challenge one of them.

The devil is in the details, of course.

- How do I time this? (Earlier would be better, in my opinion, and I think first would be ideal. If it weren't first, students could always come back with the "You didn't question Sally last week" rebuttal.)

- Which student do I choose? (Obviously, a student exercising this explanation more than once would seem to be an ideal potential challenge, but this necessarily pushes it towards later in the semester. I've been given the death-in-the-family explanation a lot, but I don't think I've ever had the same student give it twice in a semester. [By the way-- isn't that pretty compelling evidence that all of these can't be real? If you had two deaths in the family over a semester-- unlikely, but plausible-- you could use the excuse twice, but who would dare use it twice as a lie?])

Further, what are the overarching goals of the challenge? The idea is to instill an idea of credible threat-- a misleading death-in-the-family claim, if caught, would hopefully instill a pretty large sense of shame. Hopefully this will reduce its use in the future. Even if the challenge were unsuccessful-- that is, a death in the family actually occurred-- I would bear the cost of coming across as an ass (recommendations and all), but would this deter future uses of the excuse? If there were a deterrence effect, would the student simply shift to another type of excuse, or would this shock them into taking a test on time?

Perhaps most importantly; given the potential costs, which all-inclusive could be sizable even when long-run benefits may outweigh them, am I risk loving enough to take the chance?