I definitely agree that competitiveness has something to do with it, but I think you could have a highly competitive culture without as much cheating. I think

I definitely agree that competitiveness has something to do with it, but I think you could have a highly competitive culture without as much cheating. I think it's the "winner take all" mentality along with "gateway by proxy" that creates such a toxic environment. For instance, we decide who will become a physician largely by grades and MCAT scores. Many people who would have been good physicians will not be accepted to med school, and people who get into top schools are better positioned. Med school is hard, but the vast majority who get in make it through, so "getting in" is the most important thing. Unfortunately, we won't be deciding who will be a physician by seeing who is best at being a physician, it doesn't work that way. Instead, we use a GPA and exam as a stand in. This system can be malevolently cracked, with huge payoffs for the people who aren't caught, and potentially life altering consequences for those who don't cheat and fall short (I know a guy who was borderline and didn't get into med school - after two tries, he gave up on being a doctor). Compare this with being a programmer. I'm not saying there's no cheating in CS classes, I'm sure there's plenty. But if you really want to be a programmer, there isn't a magic proxy that you can crack. And if you're at the 49%ile, it's not like you don't get to be a programmer the way some people don't get to become physicians. But imagine if decided that every year, 50,000 people were allowed to study to become "programmers" (with fines and imprisonment for those who programmed without this credential), and we chose them them all based on a logic test that correlates pretty damn well with programming ability. Now I think you'd see really horrendous amounts of cheating. It isn't just the competitiveness, it's the added reliance on a crackable proxy. Character matters, but we should also see widespread cheating as perhaps a sign of a deeply flawed system - and if it's clear that we can't stop the cheating, we should understand that the system is profoundly unfair to honest people who accept the results. If cheating on a single exam can make such a difference, and it's clear that we are unable to control the cheating, then maybe we should try to be less reliant on a single, mass produced exam. One of my math professors in college said he figured he could assess a student's knowledge of an area of math after an hour of discussion. No "mass produced" exam needed, just take it where it goes, probe a bit, allow the student to consult the books of ask for a reminder if needed. It would be nearly impossible to cheat on such an exam. I understand (and so did the professor) that this is idealistic - it wouldn't be the same for everybody, it wouldn't scale, and so forth. But it's still a direction some of these struggling institutions might want to investigate.