Determinants of Out-of-State Tuition at Public Universities in the United States: A Comprehensive Model

In size, scope, and quality, the American system of public higher education remains unmatched by any other in the world. Hundreds of state-funded colleges and universities, many founded under federal land-grant programs, educate millions of students and generate massive amounts of original research each year. An impressive number of these institutions are competitive with the most elite private universities in the world, with cutting-edge facilities and faculties that boast multiple Nobel winners. They are distinguished, at least in theory, by financial accessibility: the system was founded upon the ideal that every academically qualified student should have the chance to receive a postsecondary education. Regardless of their state of residence, students are free to attend any public university in any state. Because state governments focus on their own constituents' best interests, all public universities in the United States require out-of-state students to pay higher tuition than in-state students. Nevertheless, they are generally viewed as comparatively affordable alternatives to their private peers, and a high school senior looking to leave his or her home state is likely to consider at least one top-ranked public university. Because of this, and because public universities have not been excluded from the recent trend of tuition inflation, it is worthwhile to explore the institutional characteristics associated with higher out-of-state tuition levels. This paper represents the determinants of out-of-state tuition at public 4-year universities in the United States utilizing a multiple linear regression model with 21 independent variables. There are two reasons for this abnormally large number of independent variables: first, the literature suggests that a very wide variety of factors are relevant in determining tuition levels. Second, the model incorporates several interaction variables and one multi-level qualitative variable. I estimate the model using cross-sectional data from 383 public 4-year universities in the United States, although I make several adjustments to the data and the model after obtaining my initial results. My paper builds upon the existing body of research in several ways. First, I draw data from a much larger and more diverse sample than previous researchers. For example