Top Colleges – Methodology
In order to identify top colleges for each major, we relied on five weighted variables, namely those for major emphasis, major share, academic rating, peer assessment, and graduate earnings. Variables and their corresponding weights (in parentheses) are detailed below:
Major Emphasis (25%)
Indicates the number of students at an institution studying a specific major. If a high percentage of students at a particular college are studying a major, it is likely that the major attracts a relatively large portion of the institution’s resources. To measure major emphasis, we relied on data collected by the National Center for Education Statistics as reported by The Integrated Postsecondary Education Data System (IPEDS).
Major Share (25%)
If a college enrolls a relatively high percentage of all students in the United States studying a particular major, it is likely that the college has been identified by prospective students, employers, and other stakeholders as a leader in that subject area. Though this assumption may result in bias against smaller institutions or programs, it is important to account for the fact that larger programs often have the ability to attract more tuition revenue, meet fixed costs, and reinvest the major’s offerings. To measure major share, we relied on data collected by the National Center for Education Statistics as reported by IPEDS.
Academic Rating (25%)
To construct an academic rating for each college, we rely on ten different indicators of academic quality, namely the average SAT/ACT composite of incoming students, the percentage of incoming students ranked in the top 10 percent of their high school class, student-to-faculty ratio, class size, the percentage of faculty who are full-time employees, the percentage of faculty with terminal degrees, mean faculty salary, freshman retention rate, six year-graduation rate, and graduation performance. This last variable compares an institution’s actual graduation rate with its predicted graduation rate, after controlling for the high school class rank and SAT/ACT scores of incoming students as well as the percentage of undergraduates receiving a Pell Grant, a grant primarily given to low-income students.
After collecting data, we normalize and subsequently sum values across all ten indicators, assigning each institution a total score, and by extension, its academic rating.
Peer Assessment (12.5%)
To derive a peer assessment score, we relied on questionnaires distributed to both college admission experts (including school counselors and independent education consultants) and higher education administrators, and then compared responses with “strong programs” data published by U.S. News & World Report, Niche, and other reputable sources that rank colleges within a specific discipline (e.g. CSRankings.org for Computer Science, The Hollywood Reporter for Film, Foreign Policy for International Relations, etc.). Alone, peer assessment surveys are subject to a fair amount of bias. However, when evaluated in sum and considered along with more objective data—such as that examining major emphasis and class size—they can provide more complete and corroborative insight into the strength of a particular program/major.
Graduate Earnings (12.5%)
To measure graduate earnings, we relied on salary data provided by College Score Card, PayScale, and the Equality of Opportunity Project, all of which indicate both early-career and mid-career wages of students by college. We analyzed both early-career and mid-career wages because the former is typically an indicator of how employers perceive the quality of a particular major/program, while the latter indicates how well a major/ program may have prepared students for work. In addition, when able, we used data indicating graduate earnings by field of study, as doing so provides a more accurate measure of the relationship between a specific academic program and graduate earnings. For example, PayScale provides salary data for a student attending Boston University and majoring in engineering, which is different from salary data for students attending Boston University and majoring in business.
After collecting data, for most fields, we normalized and subsequently summed weighted values across the above five indicators. Institutions earning the highest values were included in our lists. For areas of study without a predominance of institutions offering majors (e.g. game design, financial engineering, etc.), we relied primarily on variables for academic rating, peer assessment, and graduate earnings, and also totaled the institution’s number of faculty and course offerings within the subject area.