The author mentions HERI schools which he used as a subset of universities to represent all universities. Furthurmore, the HERI schools were chosen based on whi

The author mentions HERI schools which he used as a subset of universities to represent all universities. Furthurmore, the HERI schools were chosen based on which schools allowed him to collect data, or had publicly available data. This automatically biases his scale towards public universities and colleges. Private universities would be less likely to allow the author to conduct a study on their campus or provide data. Additionally, schools range in difficulty and "study time." The reason some schools are known as "party schools" and others are known as hard schools are because some require more work than others. I'm sure MIT, CMU, Berkeley and Stanford are much more time consuming that schools like Cal Poly, Penn State, Michigan State, and Texas A&M. With a biased sample, you can't make broad scale assumptions. This paper is just poorly conducted research.