<p>It would be a better study if it also controlled for average SAT level. Publics have excellent resources but have an obligation to educate a broader range of students. The difference between publics and privates could be due to differnces in average student ability. </p>
<p>This is not a criticism of public universities. It is a criticism of research methods. </p>
<p>Publics may have a smaller proportion of PhD material than privates. Privates tend to have higher average SATs. Yes, I know that publics have large numbers of top students. I am talking about the proportion of top students. Ideally, it would be factored into the normalization process.</p>
<p>LACs may seem like superior producers of PhDs than universities but it is probably a function of differences in curriculum. Universities offer majors in fields that have a job market for bachelors and masters level students:</p>
<p>education
engineering
business
and so on.</p>
<p>So when normalizing, you should divide by the number of students who were undergraduates in that particular PhD major who had SATs or GPAs above a certain level. Dividing by the entire enrollment works against universities.</p>
<p>You can make a living in these fields without a PhD. The traditional Arts and Sciences curriculum at LACs encourages encourages students to seek PhDs so they can make a living in that field.</p>