So apparently copying homework makes for lower grades. Duh!
Well it’s something I’ve noticed too. I have taught differential equations at UIUC 3 times. Twice the longer version (286) and once the shorter version (285). First time I taught from the Edwards and Penney book (standard at UIUC). Second time I was making up my notes, but still mostly taught from the book and assigned problems from EP. The third time I taught only from my notes, leaving EP as optional reading only. The last time I assigned problems only from my notes, which do not have a solutions manual (on purpose).
The first two times, with problems assigned from EP, the overall class grades were actually rather similar. However, the third time, assigning homework from my notes, the overall grades were significantly better. In fact, no curve needed to be done on the final at all. There were other factors at play, and material was a little different, but overall I would say the exams were similar enough in both scope, length, and toughness. The differences were seen even on the first exam which was very similar in scope and content for both 285 and 286. And the difference was pronounced, definitely about a grade point higher on average.
My teaching style was not different and I didn’t cover material differently. The type of homework problems were similar. The big difference in my opinion was that more students suddenly had to work out homework on their own. I remember my grader telling me that when problems were from EP that approximately half the class has their solutions more or less word for word from the solutions manual. (yes, students are not supposed to be able to buy the solutions manual, but … they do).
Now I don’t think you will gain many friends among the students by making them work out problems they have no ready made solutions for. In fact I heard many stories how this significantly reduces your student evaluations. My evaluations were comparable across all three semesters, so I guess the fact that the grades were higher because of it cancelled out any negative effects of making students do more work. (I do not read too much into the numbers from the student evaluations, I simply look at the handwritten comments for any useful constructive criticisms, which tend to be rare)