Pushing Up Ivies: Institutional Prestige and the Academic Caste System

By Robert L. Oprisko, Kirstie Lynn Dobbs, and Joseph DiGrazia

Not all university professors are equal; they are not all peers. In academic hiring, candidates are often evaluated almost entirely upon their research output rather than their teaching prowess or service. Using Oprisko’s procedural theory of honor, we can understand the hierarchic nature of academia as a prestige-based system.  Those who teach larger course-loads have less time to produce research and are less likely to embody what it means to be an exemplary academic: publishing in highly ranked journals.  Thus, where one is employed influences scholarly productivity and, by extension, how “good” an academic is at being an academic.  At the top of the academy is the R1 professor who sits in judgment of those who seek to become his or her honor-peers, (“social equals” that are capable of “conceptually compet[ing]” with one another for distinction as professors), the graduate student.  These academics are enforcing the process of face – gate-keeping PhD attainment and providing quality-control for the academy.  Our research suggests that the current system remains mired in a prestige-based caste system that diminishes the importance of individual merit in favor of affiliated honor, or institutional pedigree, when hiring.  In this first in a series of articles analyzing academic hiring, using political science as a case-study, we argue that the de facto academic hiring policy is to focus on candidates from elite universities and that this policy is harmful to the vast majority of PhD-granting programs.

Chasing prestige stratifies academic programs (and their PhD-graduates) into “born” winners and losers, with the biggest winners of meritless prestige being the Ivy League. We do not mean to suggest that these programs are not exceptional, nor that graduates from elite departments have no merit. Nonetheless, they enjoy a significant professional benefit from affiliation with the well-regarded departments and universities from which they graduate.  Our data suggests two possible hypotheses: 1) elite institutions have better job placement because search committees use PhD alma mater as a tool to weed out candidates for consideration or 2) graduates from elite programs tend to secure better jobs because of sorting normalization – the best scholars will seek out the best programs and vice versa.

We question the validity of the second hypothesis because it implies that hierarchical sorting of graduate students into graduate schools is infallible, ubiquitous, and contains zero externalities. Further, taken to extremes, this argument implies that the best scholars always come from the best graduate schools, always come from the best colleges, always come from the best high schools, etc. ad infinitum, until professional academics must have matriculated from the correct preschool, which judges the student not on their personal talent, but on the rational-action of their parents.  Such an argument further reinforces not a merit-based system of achievement, but rather a network-based system of affiliation where elites (typically the wealthy) have access to educational paths that will allow them to succeed because they are better situated than peers from less-prestigious programs. Research shows that high-achieving, low income students attend lower-tiered universities rather than the highly-selective institutions they qualify for based on merit, which creates an “undermatch,” due to financial pressure, lack of appropriate information, and geographical location.

Previous research has continuously reinforced the importance of institutional prestige as a major factor in hiring job candidates.  Arthur Stinchcombe likens the process of placement to clan-like marriages where PhD candidates, “correspond to women as being people of low status and power,” are, “exchanged partially to show mutual respect [between institutions], partly to improve each others’ material positions,” between appropriate institutions.  Shin-Kap Han shows that academic exchanges represent markets organized by the “prestige principle” such that a structure similar to Immanuel Wallerstein’s World Systems Theory emerges.  There is a core of Class I departments who are true peers and place ubiquitously, a semi-periphery of Class II departments who are in a continuous state of competition with one another and place accordingly, and a large periphery of Class III departments who struggle to place even within their own class.  Val Burris notes that “The most prestigious departments hire almost exclusively from the graduates of similarly prestigious departments to an extent that exceeds anything that can be explained by the meritocratic application of universalistic standards regarding past or potential scholarly productivity.”  Similarly, Pierre Bourdieu found the unequal distribution and self-reproduction of social-capital to function with a form of inertia, what Somit and Tanenhaus referred to as institutional or departmental tenure.

Our research seeks to examine the impact of this hiring within the political science academy by taking a snapshot of the discipline, mining data on PhD-granting institutions from the CVs of their faculty. Using the U.S. News and World Report’s ranking as a proxy for institutional prestige (the graduate ranking methodology reflects nothing more than reputation by academics), the initial results described domination by the top five schools (20% of all placement), specifically, and top eleven schools (50% of all placement), generally.

To find the effect of program rank on hiring, we present a regression model using median graduate placement rank (i.e. the median rank of the school that each department places its grads in) as the dependent variable.  Testing our data with a regression allows us to see if there is a significant relationship between U.S. News rank, which is based upon perception, and placement, which is based upon action.  Model One shows the bivariate results of median placement rank regressed on US News and World Report departmental rankings. In both measures, lower numerical values correspond to more prestigious rankings. In the US News and World Report rankings the top ranked program is ranked number one and “lower” ranked programs have higher numerical values for their rankings. Similarly, the program with the lowest value for its median graduate placement rank is the program placing its students in the most prestigious programs.  The results show a highly significant positive relationship between graduate placement and departmental rankings.  A one-position increase in graduate departmental rankings is associated with a 0.277 increase in placement ranking and a similar increase in university ranking (using undergraduate ranking data) is associated with a 0.343 increase in placement (p<0.001).  Model Two shows the addition of a dummy variable for Ivy League schools and shows that Ivy League schools place their graduates substantially better than non-Ivy League schools, even controlling for US News and World Report ranking.  Holding US News and World Report ranking constant, Ivy League programs, on average, place their graduates in programs ranked 15.78 positions better (p<0.05) than non-Ivy League graduate programs and 40.46 positions better (p<0.01) than non-Ivy universities.  This result indicates that Ivy League programs have a substantial and statistically significant placement advantage even over similarly ranked non-Ivy League programs.  In Model Three, we add two other major conferences to the model: the Big Ten and the Pac-10.  Neither of these conferences has a significant effect on placement net of US News and World Report rank.  The addition of these programs does not substantively change the effect of Ivy League membership or US News and World Report ranking.

oprisko scatter

We used institutional affiliation to determine the benefit that hiring prestigious candidates has on a graduate department’s rank, which measures other academic’s perception of departmental prestige.  The Institutional Ranking Difference (IRD) measures the difference between the rank of a department and the mean ranking of the departments from which its tenured and tenure-track faculty matriculated. The numbers are more meaningful when viewed holistically as they show the relative positions of each institution.  This new system of ranking reinforces the impact of institutional prestige and the extent to which the departmental ranking is decoupled from the rankings of the institutions from which it is drawing faculty. At the top of the IRD ranking-system are the ten departments who appear to derive a net benefit from their hires: Stanford, Michigan, Duke, Harvard, Princeton, Yale, UCSD, Columbia, Wash U-St. Louis, and UC-Berkeley are the only departments with positive scores in the IRD model. (All of these programs are ranked in the top thirteen.)  This relationship makes intuitive sense – these elite programs embody the prestige that other programs are chasing.

Going down the list does not suggest a tiered prestige system in which a candidate is equally competitive (with candidates from programs with higher prestige) at and below their rank; the drop increases to -68 points on a scale that only ranks 86 institutions.  Effectively, all PhD granting institutions seek human resources from the same few departments, but benefit to substantially different degrees.  The linear trend toward increasingly negative IRD scores demonstrates that programs are not gaining institutional prestige by hiring graduates from highly prestigious departments.  As Somit and Tanenhaus say, “Departments at less prestigious institutions have great difficulty in achieving independent distinction, at least in political science.”  The political science academy’s proclivity for prestige-based tenure-track hiring appears to disproportionately benefit well-funded, mostly private universities (and their graduates).  When faculty searches focus on hiring from prestigious departments, there is no evidence to suggest that the employer-institution will enjoy any practical benefit; hiring for prestige will not increase a program’s rank.  Departments that focus on prestige are, however, buying into a hierarchy that does not value the students they train or the institution that employs them.

INSTITUTIONAL RANKING DIFFERENCE (Ranked Programs Only)

Program Rank Instiution Faculty Rank (Mean) Faculty Rank (Median) IRD – Mean IRD – Median
1 Harvard University 6.03 5 5.03 4
2 Princeton University 5.32 5 3.32 3
2 Stanford University 8.27 5 6.27 3
4 University of Michigan 11.43 7 7.43 3
4 Yale University 8.26 6 4.26 2
6 University of California – Berkeley 8.35 5 2.35 -1
7 Columbia University 9.15 5 2.15 -2
8 Massachusetts Institute of Technology 3.42 1 -4.58 -7
8 University of California – San Diego 10.03 9 2.03 1
10 Duke University 14.28 11 4.28 1
10 University of California – Los Angeles 8.58 6 -1.42 -4
12 University of Chicago 5.87 5 -6.13 -7
13 University of North Carolina – Chapel Hill 13.23 13 0.23 0
13 Washington University in St. Louis 13.58 10 0.58 -3
15 New York University 8.70 7 -6.30 -8
15 Ohio State University 10.37 7 -4.63 -8
15 University of Rochester 8.21 4 -6.79 -11
15 University of Wisconsin – Madison 10.90 9 -4.10 -6
19 Cornell University 11.28 5 -7.72 -14
19 University of Minnesota – Twin Cities 12.59 7 -6.41 -12
21 Northwestern University 11.42 6 -9.58 -15
21 University of Texas – Austin 13.41 11 -7.59 -10
23 University of California – Davis 12.84 7 -10.16 -16
23 University of Illinois – Urbana-Champaign 15.85 11 -7.15 -12
25 Emory University 19.19 12 -5.81 -13
25 Indiana University – Bloomington 15.48 15 -9.52 -10
25 Texas A&M University – College Station 24.25 21 -0.75 -4
28 Pennsylvania State University – University Park 23.47 17 -4.53 -11
28 University of Maryland – College Park 14.47 11 -13.53 -17
28 University of Pennsylvania 7.66 5 -20.34 -23
28 University of Washington 15.25 9 -12.75 -19
32 Michigan State University 20.55 15 -11.45 -17
32 SUNY – Stony Brook 27.12 17 -4.88 -15
32 University of Iowa 17.95 13 -14.05 -19
36 George Washington University 13.42 9 -22.58 -27
36 Rice University 21.74 15 -14.26 -21
36 University of Notre Dame 15.33 9 -20.67 -27
36 University of Virginia 15.00 7 -21.00 -29
36 Vanderbilt University 11.68 9 -24.32 -27
40 Florida State University 22.67 21 -17.33 -19
40 Georgetown University 10.66 6 -29.34 -34
40 Johns Hopkins University 11.52 6 -28.48 -34
40 University of California – Irvine 13.67 9 -26.33 -31
40 University of Pittsburgh 18.92 13 -21.08 -27
45 Brown University 9.53 7 -35.47 -38
45 Rutgers State University of New Jersey – New Brunswick 16.31 11 -28.69 -34
45 University of Colorado – Boulder 20.21 15 -24.79 -30
48 University of Arizona 28.18 26 -19.82 -22
48 University of Georgia 28.00 22.5 -20.00 -25.5
50 SUNY – Binghamton 17.56 13 -32.44 -37
50 Syracuse University 18.19 9 -31.81 -41
50 University of California – Santa Barbara 10.00 7 -40.00 -43
50 University of Florida 22.97 15 -27.03 -35
54 Arizona State University 20.80 13 -33.20 -41
54 Claremont Graduate University 10.29 9 -43.71 -45
54 University of California – Riverside 16.71 11 -37.29 -43
54 University of Kansas 23.36 22.5 -30.64 -31.5
54 University of Nebraska – Lincoln 23.69 19 -30.31 -35
54 University of South Carolina 23.86 20 -30.14 -34
54 University of Southern California 19.62 11 -34.38 -43
61 Boston College 15.33 8 -45.67 -53
61 Boston University 10.63 8 -50.37 -53
61 George Mason University 20.09 10 -40.91 -51
61 Purdue University – West Lafayette 20.84 17 -40.16 -44
61 University of Missouri – Columbia 39.56 51 -21.44 -10
61 University of North Texas 32.58 28 -28.42 -33
61 University of Wisconsin – Milwaukee 33.72 28 -27.28 -33
68 American University 30.53 26 -37.47 -42
68 American University, School of International Service 33.31 25 -34.69 -43
68 Brandeis University 14.70 6 -53.30 -62
68 CUNY Graduate School 19.59 8 -48.41 -60
68 Louisiana State University – Baton Rouge 28.63 27 -39.38 -41
68 University of Houston 26.79 21 -41.21 -47
68 University of Illinois – Chicago 23.97 21 -44.03 -47
68 University of Massachusetts – Amherst 15.58 9 -52.42 -59
68 University of Oregon 21.94 9.5 -46.06 -58.5
76 SUNY – Albany 25.00 23.5 -51.00 -52.5
76 SUNY – Buffalo 30.21 28 -45.79 -48
76 Temple University 16.52 9 -59.48 -67
76 University of Connecticut 31.71 26 -44.29 -50
76 University of Kentucky 22.50 18.5 -53.50 -57.5
76 University of New Mexico 22.20 20 -53.80 -56
76 University of Texas – Dallas 20.79 18.5 -55.21 -57.5
83 Georgia State University 19.84 17 -63.16 -66
83 New School 16.33 9 -66.67 -74
83 University of Oklahoma 32.87 27 -50.13 -56
86 Northeastern University 24.86 21 -61.14 -65
86 Washington State University 24.80 17 -61.20 -69

 

Robert L. Oprisko is a Visiting Professor at Butler University, Kirstie Lynn Dobbs is a PhD candidate in Political Science at Loyola University – Chicago, and Joseph DiGrazia is a PhD candidate in Sociology at Indiana University. Butler University research assistants Nathaniel Vaught, Needa Malik, and Kate Trinkle also contributed to the research in this article. This is the second article in a series on hiring practices in higher education. The first article, “Superpowers: The American Academic Elite,” can be found here. 

 

+ posts

Established in 1995, the Georgetown Public Policy Review is the McCourt School of Public Policy’s nonpartisan, graduate student-run publication. Our mission is to provide an outlet for innovative new thinkers and established policymakers to offer perspectives on the politics and policies that shape our nation and our world.

10 thoughts on “Pushing Up Ivies: Institutional Prestige and the Academic Caste System

  1. Excellent, report! Congratulations for your new webpage! Please keep us updated with policy based evidence. Do you have evidence for International Universities?

  2. If I’m not mistaken, I believe International Unis are of interest to Dr. Oprisko as he looks to expand on this research!

Comments are closed.