Going back a bit in my career, I've looked at (real, raw) data on this topic in a different way and come up with a conflicting conclusion on the academic perfor

Going back a bit in my career, I've looked at (real, raw) data on this topic in a different way and come up with a conflicting conclusion on the academic performance side. Students taking the same course appear to perform significantly better when they take that course <= 9:00am. This trend is stronger the more students take the course. The difference tends to be on the order of the difference between, say, a B and a B-. (and I also chose specific courses where students are likely to have a variety of choices). I found that it tended to be the better students who self-selected into taking earlier courses, IIRC. The authors add confounding factors when they structure the question as "# of early courses taken" and compare it to overall GPA. For example students who are able to take later classes tend to be Juniors & Seniors, and they may have different GPA's for other reasons-- program retention requirements for example often have higher requirements than for freshmen/sophomores. Students also have different constraints if the have a job, or commute and live further away. Again, I'm only commenting on the academic performance side of the conclusions, not attendance or the other (interesting!) areas the study covered. My guess is that if the study's authors controlled for variables like student class standing, employment, age, and other factors their results might be weakened a bit. Or it could be that their results don't apply uniformly outside their own school (Though mine came from a school that was substantially similar in terms of market niche)