No Rising Tide

There is this saying that “a rising tide lifts all boats.” Apparently it’s got something to do with the free market economy, but I’ve seen it used also in the context of education, in the sense that in a group that studies together everyone benefits from everyone else’s progress.

I have always found that nearly a truism, and one I wholeheartedly agreed with. For the past two and a half years, I have been a vocal protagonist of study groups, of helping one another, of sharing solutions, shortcuts, learning techniques and mnemonics, of asking the professors questions publicly in lectures rather than privately after them, so everyone would profit from the answers. I still have no doubt that, in a general sense, this is the right thing to do. It can’t help but make us better students and, in time, better software engineers.

Unfortunately, as I belatedly realized this term, under the system prevailing here it won’t get us better grades. Possibly to the contrary.

I have discussed many times before how our professors retroactively adjust the “grading ceiling” (the minimum points required for an A+, a number relative to which the requirements for all other grades are determined) to match the actual outcome of an exam. So if hardly anyone managed to answer all the questions correctly, that ceiling will be lowered to ensure a reasonable distribution of grades. Hardly any professor wants an exam in which nobody gets an A. At the very least it would reflect badly on their ability to come up with a fair exam. Or even on their teaching skills.

That’s a nice policy to be sure, and has often resulted in my getting unexpectedly good grades. If the expectation for an A+ is lowered from the standard 95 per cent of the points achievable in the exam to say 80 per cent, and you’re the only guy who managed to collect 85, that’s great for you. Instead of a B+ you get an A+.

However, it cuts both ways. If everyone does well in the exam, the ceiling will be higher, and the same performance will result in a lower grade. At the very least, the professor might conclude that the exam was too easy and grade the replies more harshly. Or ask tougher questions next time.

Which has lately given me the sobering idea that in trying to make sure that everybody gets it I might inadvertently contribute to making it harder to get a good grade. For instance, in lectures I’ve been in the habit of asking questions whenever something was not entirely transparent to me, and if necessary asking again and again, offering interpretations and systematizations of my own, until the concept became crystal clear. Yet last week in the final distributed systems lecture I suddenly had this nasty idea that I might benefit more if only I understood it–and maybe my peer group, our original semester group. But not all those other people I hardly know. For everything we know and they don’t gives us a relative edge on them, puts us higher on a correspondingly lower grading scale.

Is that a nice thought? No it isn’t. And in fact I fought the temptation and kept asking questions. But in a way even having this thought is the inevitable conclusion from a system in which often the performance of the students, rather than the expectations of the professors, determine the grading scale. If you really think it through, in this here system the best scenario you can get is a disastrous exam in which only you do well, or at least only a small number of people. For then your excellent grade is a foregone conclusion. So from that point of view, best make sure only you get it. Ask the professor privately and don’t share his reply. Keep your insights, your learning techniques, and your solutions to yourself. The more everyone else is kept in the dark, the more you shine. Dog eat dog.

Is that really how it’s meant to be? Or am I just having morbid thoughts that late in the game?

Advertisements