‘Testocracy’ Is Here to Stay–Alas

In her new book, Harvard Law Professor Lani Guinier attacks “testocracy,” the over-reliance on standardized tests in deciding who gets into college, who has the chance to attend America’s premier institutions, and who is relegated to the cheap seats of community colleges and for-profit schooling.

Unfortunately, Guinier’s “Tyranny of the Meritocracy: Democratizing Higher Education in America,”” contains a number of unexamined assumptions or logical flaws that critics of the prevailing meritocracy (myself included) have been making for years.  For example, Guinier claims that colleges ought to select students on the basis of democratic merit because doing so would be good for the society — as opposed to over-relying on academic merit that can detract from social and economic welfare.

By what authority can Guinier make such a claim?  Who determines that colleges and universities are responsible for deciding what educational policies or admissions systems are in the public interest? State legislatures? Boards of Trustees?  If so, then we can agree that policymakers are free to assign colleges and universities with the mission of selecting students on the basis of “democratic merit” or any type of merit they choose.

Why Reform Pleas Fail

The bigger question: Why have institutions largely ignored pleas to reform the “meritocracy” as we know it? Like some other critics, Guinier asserts that colleges and universities “ought” to be reforming admissions to include democratic merit for the sake of the “public interest.” But perhaps she ought to ask why the testocracy remains so intractable, and why efforts to upend it have largely failed to fundamentally change the way Americans look at merit.

To appreciate how entrenched is the “testocracy,” consider some history.  The story basically starts in France at the turn of the 20th century, when Alfred Binet invented the first intelligence scale for school children.  His intention was to create a modest tool for placing French school children in their proper grade. American psychologists got hold of Binet’s scale and commercialized it for widespread use. Thus was born an IQ test known as the Stanford-Binet.  Certain American psychologists convinced the United States Army to use the test on Army recruits during World War I.  That fiasco produced the dubious notion that certain recruits whose families had immigrated to the United States decades prior — Germans, Norwegians and other Nordic immigrants — were deemed to have superior intelligence. By contrast, more recent immigrants, such as Italians, Poles and Jews, were often labeled as mentally defective due to poor performance on the Stanford-Binet.

Nevertheless, “visionaries” at Harvard University believed that such intelligence testing could identify students who would find their natural place as future leaders in society. From that vision, the Educational Testing Service was created, and the ETS adapted the Stanford-Binet to produce the first Scholastic Aptitude Test.

Binet himself had observed that doctors’ and professors’ sons outperformed the children of carpenters and bricklayers on his school placement scale.  Decades later, SAT scores largely demonstrated that America’s future leaders would also come from the very established, well-educated families who already dominated the ranks of Harvard, Princeton and Yale. If, before the SAT, the privilege to attend Harvard was passed down on the basis of blood heritage, after the SAT such privilege was reproduced by means of the “scientific” veracity of an objective test.  The test proved that the established order was also the properly deserved order.

And so the much-maligned “testocracy” ship sailed long ago, and it has never wavered from its course. This, despite its psychometric flaws, which can lead to wrong or incomplete assessments of human ability.  We know that the SAT and tests like it have a limited ability to predict future performance in school or other real-world endeavors.  We know that such tests tend to select for people who excel in logical-mathematical ability, as opposed to creative ability and deeper thinking styles. We know that such tests correlate to parental income and education–the “Volvo Effect,” as it were–to a far greater degree than other measures of academic performance, such as grades or portfolio assessments. We know all these things. And, yet they do not matter.

A Permanent Fixture?

Instead of arguing ad infinitum that colleges and universities “ought” to revolutionize the way we understand and measure merit, perhaps it’s time to concede that, realistically, that debate is settled, and that continuing to recycle the same arguments decade after decade is, at last, a completely useless exercise.

Admissions committees at selective colleges and universities — the theatre in which these issues actually matter — don’t care about the SAT’s predictive validity or lack thereof.  Parents do not care how well the SAT predicts academic success.  Students don’t care either.

The SAT’s capacity as both a fair and accurate tool does not matter.  And few people beyond the usual critics care because it’s in nobody’s interest to care — except perhaps to those disenfranchised students and families who might, by some democratic measures of achievement, deserve to be chosen, but aren’t.

Forget about predictive validity and forget that some children from economically and culturally well-endowed families start the meritocracy game far ahead of other children born with fewer economic and cultural advantages.  Higher education authorities largely give lip service to these inequities. Even affirmative action policies tinker with the prevailing merit system on the margins, affecting a relatively small number of students at selective colleges and universities.   And, as Guinier herself concedes, the racial or ethnic minorities who benefit from affirmative action, are people of color who often come from relatively well- educated and affluent families.

The SAT as a Market Signal

From an economist’s perspective, the real reason the “testocracy” will not disappear is that the SAT and tests like it serve an important market signaling device in the higher education marketplace.  While most markets depend on price and value as the primary market signals for consumers and producers, higher education is a special case. Like it or not, higher education in the United States does operate under the glare of the “public interest.”  But the parameters the public interest are often vague and undefined, leaving colleges and universities free to enact business models they believe will maximize their own private interests while taking heed of some vague social contract.

Because of this vaguely defined public interest, educational authorities don’t allow price and value alone to determine who attends their institutions.  Were colleges to rely on strictly on price and value, the real prices that consumers would pay for elite higher education would most likely climb to unaffordable levels for all but the top 10 percent of households.

To be sure, elite education is already approaching this level of exclusivity, but it’s moderated by institutional aid, allocated on the basis of financial need.  Rich schools — like Harvard, Princeton, Stanford, or Yale — can subsidize capable students to such a degree that families need only pay a fraction of the true price, as determined by supply and demand.

Insanely Intense Competition

The role of the SAT is peculiar indeed.  Because institutions can’t rely strictly on a true market price, their allocation of subsidies is easily determined by a second-best solution: SAT scores.   It goes without saying that, in recent years, the competition for obtaining high SAT scores — sufficient to obtain access to richly endowed colleges and universities — has become insanely intense.

An SAT score is a simple market signal that correlates almost perfectly with the degree of selectivity of a given college or university.  Without the SAT or an equivalent clear market signal to partially substitute for price, the market would become insanely complex, involving transactions costs that would push up the price of higher education even higher, begging the question of who or what would bear the burden of these extra costs.

But the SAT avoids all that.  Owned by the College Board, a not-for-profit organization, the SAT serves some vague “public interest” in that it purports to measure merit of some sort, however flawed.  At the same time, the SAT easily sorts students into classes of student consumers who are most likely to “fit” the business model of a particular institution.  An SAT score in the 95th to 99th percentile easily identifies a pool of students from which a highly selective college or university choses which individuals to subsidize to one degree or another.  With highly sophisticated market segmentation techniques, risk-averse institutions know that for a given geographic location, a certain SAT score range will yield low-risk students who are “worth” subsidizing.

Low-risk students typically come from certain geographic locations and demographic markers, including income, wealth, home prices, education levels, and so on.   Low-risk students eventually graduate, find lucrative careers and join alumni organizations.   Alumni are called upon to donate back to institutions to build and maintain the very endowments that helped subsidize their education in the first place.  Alumni have children and grandchildren who become legacies who are provided generous admissions advantages. Thus the virtually self-perpetuating cycle goes on.

 Excluded from the Pool

To be sure, this highly competitive environment results in the exclusion of untold numbers of students.  This includes those who are essentially not allowed to even enter the competition because their unacceptable SAT scores exclude them from the viable application pool in the first place; and this competition also excludes the vast majority of students who, despite impressive credentials, are not selected from the viable applicant pool.

Yes, this competition yields results that might justifiably be called “unfair.” But the market seems to work with a fair degree of clarity and predictability.  That surely is worth a ton of value for both institutions and the larger society, and needs to be factored in any equation for reforming the meritocracy as we know it.

In order to compensate for these inequalities, Guinier argues that colleges and universities ought to employ selection methods that take into account “democratic” merit, including the many sorts of non-cognitive intelligences that help individuals function in groups and maximize the effectiveness of groups.  She believes that replacing “testocratic” merit with “democratic” merit would ultimately, be in the best interest of democratic society.

But I believe she discounts the degree to which existing admissions systems already take into account talents and skills that are not necessarily captured by classroom grades and test scores.  For the relatively high-scoring students who do make the pool of viable candidates, colleges do look carefully at non-cognitive factors and experiences that might tip the scales.

What’s more, a number of colleges and universities have in recent years become “SAT optional,” meaning that students can chose to submit test scores or not.  Obviously, those who choose not to submit scores must demonstrate merit in a variety of other ways that include Guinier’s “democratic” merit.  At some major universities, such as the University of California, academic measures still predominate, but lower-scoring students do have the opportunity to demonstrate how they have overcome barriers of poverty and disadvantage, providing a larger context in which to interpret their modest SAT score.

In the end, there is good news and bad for Guinier.  The bad news is that history, habit, and the nature of the higher education marketplace have made her book largely irrelevant.  Testocracy is here to stay and will continue to play a decisive role in American higher education. But, the good news for Guinier is that the enterprise is flexible.  Exactly what the public interest really is remains sufficiently vague. That means that colleges and universities are free to modify testocracy on their campuses as they choose. There will be ample opportunity for experimentation and discovering new and interesting ways to tap into the nation’s vast pool of undiscovered talent.  There’s no doubt the country cannot afford to waste human talent.  But, unless and until the courts or legislatures or boards of trustees settle once and for good exactly what “merit” should mean — a certainty never likely to be achieved — we would do well to stop pushing this particular boulder. For better or worse, it has become unmovable.