One of Hacker and Dreifus’s recommendations about how to improve American higher education is to “Demand the ‘Golden Dozen’ deliver.” Yet even with the questionable metric for “delivery” that they’ve chosen, they wrongly conclude that Williams – and some other members of the Golden Dozen, such as Princeton and Yale, are failing to deliver. 

Hacker and Dreifus begin their critique by defining the “mission” of the “Golden Dozen.” They claim to find it in statements by Princeton that the university’s “mission is to educate national leaders.” Is that a claim Williams makes about its mission? Harvard? Brown?  As an alum, I take pride in the leadership accomplishments of Ephs, and I’m sure most of the Williams community does as well. But is that the College’s “mission”? The statement of “Mission and Purposes” adopted by the Board of Trustees in 2007 doesn’t go that way: 

Williams seeks to provide the finest possible liberal arts education by nurturing in students the academic and civic virtues, and their related traits of character. Academic virtues include the capacities to explore widely and deeply, think critically, reason empirically, express clearly, and connect ideas creatively. Civic virtues include commitment to engage both the broad public realm and community life, and the skills to do so effectively… [T]he College’s greatest mark on the world consists of this: the contributions our alumni make in their professions, their communities, and their personal lives. 

At a minimum, Williams puts the “personal” on par with the professional and the community. And “community” can be defined in many ways other than “national.” 

Williams critic and Amherst alum David Hacker with co-author Claudia Dreifus. Photo by Tequila Minksy.


Hacker and Dreifus then set out to prove that the “Golden Dozen” are failing in their mission by proving that Princeton is failing to prepare “national leaders.” Naturally, that’s exactly what they conclude, which might come as a surprise to those familiar with Indiana governor Mitch Daniels (Princeton ’71, a much-discussed possible 2012 presidential candidate), Bob Ehrlich (Princeton ’79, former governor of Maryland and 2010 Republican nominee for Senate), Tom Kean (Princeton ’57, former New Jersey governor and chairman of the 9/11 Commission); Eliot Spitzer (Princeton ’81); Donald Rumsfeld (Princeton ’54); Bill Frist (Princeton ’74). And that’s hardly an exhaustive list even of those in the news this past year. 

How do they reach their conclusion? They looked at a single class of matriculants: those who enrolled in Princeton as part of the class of 1973. Why Princeton ’73? 

[F]irst, it was of manageable size to study, yet also large enough to be a reliable sample. Second, it was the first class with women and a relatively sizable black representation . . . Third, and most relevant for our purposes, its members have reached their mid-fifties, and were old enough to have achieved much of what they would do in their lives.” 

Really? It’s true that several of the illustrious alumni I’ve named above graduated Princeton in the 70s, or even more recently. But it hardly follows that a class in its “mid-fifties” has “achieved much” of what it will do in terms of “national leadership.” Particularly if you’re going to cherry-pick certain pinnacle positions, such as United States Senator or CEO. 

More importantly, as much as Hacker and Dreifus might want to cloak their conclusions in the language of formal analysis, it’s erroneous to suggest that even a randomly-chosen single class from a single institution is a “reliable sample” for what they’d like to measure, especially if they define it as “national leadership” positions. No single class year can be a reliable measure of whether your institution produces senators when there are only 100 positions available and at least 30 potential class years at a time from which a sitting senator could probably have graduated. Hacker and Dreifus are a social scientist and a science writer, respectively. They should know this. 

Hacker and Dreifus analyze this class according to several other metrics. The one they rely on most heavily is the percentage of the class that can be found in Who’s Who in America. Really. I’m dubious about the value of Who’s Who as a source, but at least they’re not the first to use it. Clarence Lovejoy, author of the once-popular Lovejoy’s College Guide used it as early as 1940 in his So You’re Going to College. More recently, and more relevantly, Ohio University Professor Richard Vedder has used it in studies for his Center for College Affordability and Productivity. 


Poring through the pages of Who’s Who for members of Princeton ’73, Hacker and Dreifus find that “none . . . has served in cabinet or sub-cabinet positions, in either congressional chamber, as a federal judge or financial official, or as chief executive of a national corporation.” 

Nothing underscores their determination to find the class unaccomplished as the way they gloss over the fact that Princeton ’73 produced at least one world leader: Queen Noor of Jordan. Hacker and Dreifus dismiss her as someone who “wed” her way into prominence (never mind that she met and impressed the future King of Jordan while pursuing a career in her field and has done far more with the title than play “palace wife”). 

Their analysis provides no indication of the extent to which their result is influenced by the arbitrary choice of Princeton ’73. So suppose Hacker and Dreifus had picked Princeton ’72: then, their list would have included Supreme Court Justice Samuel Alito. Suppose they’d picked Princeton’ 74? Senator Frist. Princeton ’76? Justice Sotomayor — at least that would have supported their spin that they found it was primarily the Princeton women who were successful. (If Hacker and Dreifus had picked other Golden Dozen schools in 1973, they would have also found different results. For example, had they chosen the much smaller class of Amherst ’73, they would have found a governor, a congressman, and a Cabinet-equivalent appointee all graduating in the same year). 

Hacker and Dreifus also criticize Princeton for having an ostensibly low total number of listings in Who’s Who, finding that “only” 2.8% of the 934 members of the Princeton class of 1973 are listed, suggesting that number should be much higher. To their Who’s Who analysis, they add a couple of other measures of “success”: that the median income for class members responding to a 35th reunion questionnaire in 2008 was $175,000 for men and $115,000 for women; and that only 120 members of the class had children who had also gone to Princeton. 

In a moment, I’ll look at the merits of their analysis. But first I want to address the cheap shot they take at Williams at the end of their analysis, stating that they “suspect” that their methodology and sample would yield similar results at “Williams or Yale.” They then add that the “Golden Dozen” should seek to emulate liberal-arts schools such as “Haverford and Davidson, [whose graduates] are nearly twice as likely to end up in Who’s Who as products of Duke or Brown or Penn.” 

This is either extremely sloppy or deliberately misleading. Hacker and Dreifus do have a reason to “suspect” that they might get different results at Williams or Yale. To support their later laudatory comment about Haverford and Davidson, the authors rely on the work of the aforementioned Dr. Vedder in his article “Outcomes Based Assessment of Universities. And that article reveals that, after adjusting for enrollment, Yale and Williams are the #2 and #3 best represented undergraduate institutions in Who’s Who, trailing only Harvard. (Princeton is not far behind). 

 Haverford? Davidson? Down the list – a Williams graduate is about 50% more likely than a graduate of those (also excellent, also elite, also expensive) institutions to appear in the pages of Who’s Who. No wonder they moved the point of comparison for Haverford and Davidson to that misleading line about “Duke or Brown or Penn.” 

Of course, it’s not just the unwarranted attack on Williams (and Yale) that makes this analysis misleading. Hacker and Dreifus provide no context to assess their “statistics” – literally none at all. Don’t they think readers might want to know, for example, the median household income in the United States (just over $50,000) or the number of people in Who’s Who in America (leaving aside the dubious value of this publication as a source): slightly less than 100,000. Or what percentage of the graduates of Princeton are now outside the United States and listed in, say, Who’s Who in the World? (In the past, there was also a “Who’s Who in Asia, but it appears to be defunct). 

What’s a reasonable target for alumni in Who’s Who, then? Princeton has 58,000 living alumni; if 2.8% of them were listed, that would represent nearly 2% of the entries — from a single school with class sizes smaller than the average in the Golden Dozen. If 2.8% of each of the Golden Dozen school alumni were in Who’s Who, more than 20% of the book would come from those twelve schools alone. And remember, this is a list that leaves out, as Jeff pointed out, MIT, CalTech, Rice, Northwestern, UChicago, Georgetown, Swarthmore, and Pomona.  (Although, actually, because Princeton — like Williams — is much better represented in Who’s Who than the bigger schools like Penn and Cornell, the total Who’s Who representation of the Golden Dozen is not quite that high).

Yet Hacker and Dreifus are clearly suggesting it should be even higher. My view: ridiculous. We already live in a society largely governed by an insular group of alumni from a few elite institutions, particularly when you combine undergraduate and graduate attendance at those schools.  At Williams, if there isn’t a future Who’s Who member in your entry, there’s going to be one in your first seminar. On average, there will be more than a dozen more in your graduating class. Those are just the people who are being “nationally-recognized” for their professional achievements — one of the three categories of accomplishment in our mission statement.

So, again, even if Who’s Who listings are a good way to measure college accomplishments, the only reasonable conclusion is that if you want to be listed in Who’s Who — or to know people that are — Williams (or Princeton) is a good place to go. If Hacker and Dreifus had directed even an ounce of effort towards context, they would have pointed out that Princeton ’73 alumni are nearly 100 times more likely to be listed in Who’s Who than the average American and nearly 30 times more likely than the average American holder of a bachelor’s degree.

Hacker and Dreifus have invented a grouping called “The Golden Dozen,” which has a nice ring, and tried to tarnish it. Their attempt to do so with an outcomes-based criticism is ridiculous, at least as to the giants of name like Harvard, Yale, and Princeton, and our own diamond-in-the-“wilderness”, Williams. For these schools, there’s no reason not to embrace Hacker and Dreifus’s elegant title except its association with Hacker and Dreifus. If Hacker and Dreifus haven’t trademarked “The Golden Dozen,” maybe a campaign to take it back is in order.

Print  •  Email