Teaching vs. Grading: the Problem with Tests (and its Implications)
I just recently wrapped up the first mid term for the Lean Launchpad Entrepreneurship class, and the experience has been educational for me; it’s also given me some interesting insight on how organizations work in Kazakhstan, particularily in Universities. But first, some background.
Before coming to Kazakhstan, I would’ve considered myself a corporate trainer, and certainly not a school teacher. Ironically, I grew up quite disliking my high school teachers, so I found myself surprised when I discovered I really enjoyed working with young people — it’s one of my favorite parts about working in KZ.
A major difference between corporate training and teaching in a school class, however, is the problem of grades. At a corporation, I’m quite indifferent if executives don’t show up — they’re the customer, and I’m paid to give them the maximum skill per unit of effort, not to babysit. Undergraduate universities, however, are at that strange shade of grey where students are legally adults, but in many ways still treated as children. And thus, my least favorite part of teaching arises. Grading, that unavoidable annoyance.
Different schools approach this problem in different ways, and how they approach it gives an interesting window on their organizational culture and teaching philosophy. At one school, I’ve found teachers and students to have a alarmingly adversarial relationship; grades are a heavy-handed tool to force student compliance. At another, students are distinctly less grade-driven, and have spent time reading additional material simply because it was useful for their business.
The Problem of Cheating
I suspect that this has to do with the legacy of Soviet organizational behavior. One university I work at is tremendously bureacratic, with little connection between result and reward. This attitude filters down to students, who, quite naturally, respond to individual incentives. This leads to a lot of bad things: increased cheating, incessent bargaining for individual grades, and worst of all, increased cynicism. After all, why study if 1) the material taught has no impact on real life, 2) the system is not a test of material learned but rather of individual relationships, and 3) you see cheaters being obviously rewarded?
This puts me in an interesting position. Becuase if I’m to be completely honest, I cheated a lot in high school. At that point in time, I rationalized it pretty easily — if the teachers don’t care about teaching, after all, why should I care about learning? And if everyone is cheating, well, then I care too much about getting into college to not be the best.
Luckily for me, I met a lot of teachers at University that changed how I thought about this. At Berkeley, professors were obviously passionate about educating their students, and I felt that I had no right to game a system that was made in good faith. At that level, I think most students felt the way I did. And it was a credit to Berkeley that students there considered cheating to be something only done by immature high school students.
So this experience puts me in an interesting position. At one point or another, I’ve been all of the following: the student cheater organizing sophisticated rings in class, the studious learner disgusted by cheating, and now the professor trying to figure out how to minimize cheating. It kinda feel like I’m in a comedy movie, especially when I catch students red-handed. You’re not the first to try that, kid.
Why is this important?
How a person views cheating, I think, is the clearest sign of how modern or international someone is in Kazakhstan. At one university, I’ve met administrators who (in an apparently genuine belief) justifies non-enforcement of rules by arguing that cheating is clever and thus entrepreneurial. This is usually when my palm meets my forehead. Even 24 years after independence, the tentacles of the Soviet Union still show up in random places here.
This is important not just because of moral or social reasons, however. It’s a practical one. For example, I use a prisoner’s dilemma-type game in another class to teach students why trust is so important for a country’s economic development. The less business people have to worry about being cheated (corruption, broken contracts, etc.), the more they can invest and focus on growing business as well as incomes. The Congo might have more natural resources then Scandanavia, but its pretty obvious where I’d rather do business.
Which is why, particularily as business teachers, we need to emphasize this to our students — being clever at cheating is about as impressive as stealing purses in a restroom. All you’re doing is stealing someone’s elses grade.
As I tell my (sometimes disbelieving) students, my job is to teach and train, not to give them a hard time. In fact, not punishing cheaters would make my job a whole lot easier, since I don’t have to go through the bureacracy of an administration that, if I’m honest, prefers to look the other way. At the same time, though, that still leaves us with a problem to solve: how do you design a system that maximizes learning, minimizes cheating, and is still useful for the customer?
First, we need to tone down the adversarial nature of grading. It shouldn’t be a student fighting (or worse, begging) a teacher who is trying to give a low grade. Grades should be like the referee at a foot ball game, fair to high scorers but not there to fight the players.
Next, teachers here could learn to standardize grading better. I’ve spent far too much time explaining what a grading curve is here, and not only to students. This leads to unfair grading, and is demoralizing when students realize the fastest way to a better grade is to pester the professor at the end of semester.
Testing Tests (as a system)
So how to make, or at least nudge, our way to a better system? Well, let’s first be honest: tests are often terrible measurement tools. Speaking as a well-versed exam crammer, I can’t tell you a single thing I remember from an undergraduate test I studied for.
That being said, examinations such as multiple-choice exams are good for testing information retention, and there are plenty of well-designed tests that are accurate, reliable, and valid (the CFA exam, for example). In that vein, this is what a good school test should do: make it easy to tell who studied (have high spread), should test application of knowledge over just rote memorization, and be practical for large groups of students (essays are awesome, but terrible for grading).
And most importantly, I think students should be able to clearly see the application to their own experience. For if schools aren’t here to give skills to improve lives, what are we here for?