Opinion Blog


Rick Hess Straight Up

Education policy maven Rick Hess of the American Enterprise Institute think tank offers straight talk on matters of policy, politics, research, and reform. Read more from this blog.

Education Opinion

The 2016 RHSU Edu-Scholar Public Influence Scoring Rubric

By Rick Hess — January 05, 2016 10 min read
  • Save to favorites
  • Print

Tomorrow, I’ll be unveiling the 2016 RHSU Edu-Scholar Public Influence Rankings, honoring the 200 university-based education scholars who had the biggest influence on the nation’s education discourse last year. Today, I want to run through the scoring rubric for those rankings. The Edu-Scholar rankings employ metrics that are publicly available, readily comparable, and replicable by third parties. This obviously limits the nuance and sophistication of the measures, but such is life.

Given that there are well over 20,000 university-based faculty tackling educational questions in the U.S., even making the Edu-Scholar list is an honor—and cracking the top 100 is quite an accomplishment in its own right. So, who made the list? Eligible are university-based scholars who have a focus wholly or primarily on educational questions (“university-based” meaning a formal university affiliation, including a webpage on a university site). The rankings include the top 150 finishers from last year, minus a few retirees, augmented by 50 “at-large” additions named by a selection committee of 26 accomplished and disciplinarily, intellectually, and geographically diverse scholars. The selection committee (composed of members who were already assured an automatic bid by dint of their 2015 ranking) first nominated individuals for inclusion, then voted on whom to include from that slate of nominees.

I’m indebted to the committee members for their assistance, especially given that they’re all extraordinarily busy folks. So, let’s take a moment to acknowledge the members of the 2016 RHSU Selection Committee. They are: Deborah Ball (U. Michigan), Camilla Benbow (Vanderbilt), Linda Darling-Hammond (Stanford), Susan Dynarski (U. Michigan), Susan Fuhrman (Columbia), Dan Goldhaber (U. Washington), Sara Goldrick-Rab (U. Wisconsin), Jay Greene (U. Arkansas), Eric Hanushek (Stanford), Shaun Harper (U. Penn), Doug Harris (Tulane), Jeff Henig (Columbia), Thomas Kane (Harvard), Gloria Ladson-Billings (U. Wisconsin), Marc Lamont Hill (Morehouse), Susanna Loeb (Stanford), Robert Pianta (U. Virginia), Morgan Polikoff (USC), Jim Ryan (Harvard), Marcelo Suarez-Orozco (UCLA), Jacob Vigdor (U. Washington), Kevin Welner (CU Boulder), Marty West (Harvard), Daniel Willingham (U. Virginia), Yong Zhao (U. Oregon), and Jonathan Zimmerman (NYU).

Okay, so that’s how the list of scholars was compiled. How were they ranked? Each scholar was scored in eight categories, yielding a maximum possible score of 200—but only a handful of scholars actually cracked 100. Surveying the results shows that a 100 puts one safely in the top 20 and a 75 suffices to crack the top 50.

Scores are calculated as follows:

Google Scholar Score: This figure gauges the number of articles, books, or papers a scholar has authored that are widely cited. A neat, common way to measure the breadth and impact of a scholar’s work is to tally works in descending order of how often each is cited, and then identify the point at which the number of oft-cited works exceeds the cite count for the least-frequently cited. (This is known to aficionados as a scholar’s “h-index.”) For instance, a scholar who had 20 works that were each cited at least 20 times, but whose 21st most-frequently cited work was cited just 10 times, would score a 20. The measure recognizes that bodies of scholarship matter greatly for influencing how important questions are understood and discussed. The search was conducted using the advanced search “author” filter in Google Scholar. A hand search culled out works by other, similarly named, individuals. For those scholars who had been proactive enough to create a Google Scholar account, their h-index was available at a glance. While Google Scholar is less precise than more specialized citation databases, it has the virtue of being multidisciplinary and publicly accessible. Points were capped at 50. This measure offers a quick way to gauge the expanse and influence of a scholar’s work. (This search was conducted on December 14.)

Book Points: An author search on Amazon tallied the number of books a scholar has authored, co-authored, or edited. Scholars received 2 points for a single-authored book, 1 point for a coauthored book in which they were the lead author, a half-point for coauthored books in which they were not the lead author, and a half-point for any edited volume. The search was conducted using an “Advanced Books Search” for the scholar’s first and last name. (On a few occasions, a middle initial or name was used to avoid duplication with authors who had the same name, e.g., “David Cohen” became “David K. Cohen.”) We only searched for “Printed Books” (one of several searchable formats) so as to avoid double-counting books which are also available as e-books. This obviously means that books released only as e-books are omitted. However, circa 2015, this still seems appropriate, given that few relevant books are, as yet, released solely as e-books (this will likely change before long, but we’ll cross that bridge when we come to it). “Out of print” volumes were excluded. This measure reflects the conviction that book-length contributions can influence public discussion in an outsized fashion. Book points were capped at 25. (This search was conducted on December 14.)

Highest Amazon Ranking: This reflects the author’s highest-ranked book on Amazon. The highest-ranked book was subtracted from 400,000 and the result was divided by 20,000 to yield a maximum score of 20. The nature of Amazon’s ranking algorithm means that this score can be volatile and favors more recent works. For instance, a book may have been very influential a decade ago and continue to influence citation counts and a scholar’s visibility, but it may no longer sell many copies. Thus, it would tend to have a low Amazon ranking. The result is an imperfect measure that conveys real information about whether a scholar has penned a book that is influencing contemporary discussion. (This search was conducted on December 14.)

Education Press Mentions: This measures the total number of times the scholar was quoted or mentioned in Education Week, the Chronicle of Higher Education, or Inside Higher Education during 2015. The search was conducted using each scholar’s first and last name. If applicable, we also searched names using a common diminutive and both with and without middle initials. In each instance, the highest result was recorded. The number of appearances in the Chronicle and Inside Higher Ed were averaged and that number was added to the number of times a scholar appeared in Education Week. (This was done to give K-12 and higher ed equal weight.) That total was multiplied by two. Ed Press points were capped at 30. The intent is to use a “wisdom of crowds” metric to gauge a scholar’s influence on the public discourse last year. (This search was conducted on December 15.)

Web Mentions: This reflects the number of times a scholar was referenced, quoted, or otherwise mentioned online in 2015. The search was conducted using Google. The search terms were each scholar’s name and university affiliation (e.g., “Bill Smith” and “Rutgers”). Using affiliation served a dual purpose: It avoided confusion due to common names and it increased the likelihood that the mentions were related to their university-affiliated role, rather than their activity in some other capacity. If a scholar was mentioned sans affiliation, that mention was omitted here. (That likely tamps down the scores of well-known scholars for whom affiliation may seem unnecessary. However, since the Darling-Hammonds and Hanusheks fare just fine, I think that’s okay.) As with Ed Press, searches included common diminutives and were run with and without middle initials. For each scholar, we used the single highest score from among these various configurations. (We didn’t sum them up, as that would have padded some scores with a lot of duplication.) Points were calculated by dividing total mentions by 30. Scores were capped at 30. (This search was conducted on December 15.)

Newspaper Mentions: A Lexis Nexis search was used to determine the number of times a scholar was quoted or mentioned in U.S. newspapers. Again, searches used a scholar’s name and affiliation, diminutives, and were run with and without middle initials. In each instance, the highest result was recorded. Points were calculated by dividing the total number of mentions by two, and were capped at 30. (The search was conducted on December 15.)

Congressional Record Mentions: We conducted a simple name search in the Congressional Record for 2015 to determine whether a scholar had testified or if their work was referenced by a member of Congress. Qualifying scholars received five points. (This search was conducted on December 15.)

Klout Score: A Twitter search determined whether a given scholar had a Twitter profile, with a hand search ruling out similarly named individuals. The score was then calculated from a scholar’s Klout score, which is a number between 0 and 100 that reflects a scholar’s online presence. The Klout score was divided by 10 to calculate points earned, yielding a maximum score of 10. (This search was conducted on December 14.)

Scores are intended to acknowledge scholars whose widely-referenced body of work influences our thinking and who are actively engaged in public discourse. That’s why the scoring discounts, for instance, academic publications that are rarely cited or books that are unread or out of print. Generally speaking, the scholars who rank highest are those who are both influential researchers and influential public voices.

There are obviously lots of provisos when perusing the results. Different disciplines approach books and articles differently. Senior scholars have had more opportunity to build a substantial body of work and influence (and the results unapologetically favor sustained accomplishment). And readers may care more for some categories than others. That’s all well and good. The whole point is to spur discussion about the nature of constructive public influence: who’s doing it, how much it matters, and how to gauge a scholar’s contribution. If the results help prompt such conversation, then we’re all good.

It’s worth noting that some academics dabble (very successfully) in education, but that it’s only a sideline for them. Two who come to mind are James Heckman and Raj Chetty. These accomplished individuals, for all their gifts, are not eligible for the Edu-Scholar rankings. (I’m sure they’ll survive.) For a scholar to be eligible for inclusion, education must constitute a substantial majority of their research and publication. Otherwise, among other challenges, the acclaim a Heckman or Chetty has earned writing about larger economic questions would play havoc with the rankings. This decision helps ensure that the rankings serve as something of an apples-to-apples comparison among scholars who focus on education.

Two questions commonly arise: Can somebody game this rubric? And am I concerned that this exercise will encourage academics to chase publicity? As for gaming, color me unconcerned. If scholars (against all odds) are motivated to write more relevant articles, pen more books that people read, or get more proactive about communicating in an accessible fashion, that’s great. That’s just good public scholarship. If I help encourage that: sweet. As for academics working harder to communicate beyond the academy, well, there’s obviously a point where that turns into sleazy PR, but most academics are so immensely far from that point that I’m not unduly concerned.

Tomorrow’s list is obviously only a sliver of the faculty across the nation who are tackling education or education policy. For those interested in scoring additional scholars (or themselves!), it’s a straightforward task to do so using the scoring rubric. Indeed, the exercise was designed so that anyone can generate a comparable rating for a given scholar in no more than 15-20 minutes.

And a final note of thanks: For the arduous task of coordinating the selection committee and then spending dozens of hours crunching and double-checking all of this data for 200 scholars, I owe an immense shout-out to my ubertalented, indefatigable, and eagle-eyed research assistants Jenn Hatfield and Kelsey Hamilton.

The opinions expressed in Rick Hess Straight Up are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.