Why Selective Colleges (and Outstanding Students) Should Be Less Selective
Current admissions processes at top universities are expensive and inefficient.
Posted Dec 03, 2009
High school seniors trying to get into the best college are on a fool's errand. Chance factors (eg., ones first year roommate or Bio 101 teacher) will have a bigger effect on success and failure -- satisfaction and disappointment -- than the tiny differences among schools that are within the range of acceptability. Decision scientists Detlof von Winterfeldt and Ward Edwards wrote many years ago about "the principle of the flat maximum," to describe situations in which random variation or measurement error was larger than the differences among the things being measured. It's a pretty good bet that when it comes to differences among highly selective institutions, high school seniors are operating in the region of a "flat maximum."
And the same is true of selective colleges and universities trying to decide which, say, 100, out of 5,000 great applicants should be admitted. The differences among the well-qualified applicants are trivial in comparison to the error in the tools used to predict their college performance. And yet, just as students compete for admission to selective universities, universities compete for the "best" students. Thanks in part to the perverse effects of the U.S. News and World Report rankings, the profiles of admitted students, and the percentage "yield" of those who have been admitted, are public information that plays a significant role in the ranking of each school. Every school is terrified that if their numbers drop, they will begin a kind of "death spiral" in which lowered rankings and reputation lead to worse students which lead to worse rankings, and so on, until a school falls out of the group of the select. Thus, we have merit-based financial aid crowding out aid based on need, ever more elaborate recreational facilities on campuses, glitzy view books and videos, fancy web sites, fun-filled weekends for admitted students -- untold millions of dollars spent each year trying to convince the students you've admitted that yours is the school for them -- even though, if they chose not to come, there would be a half-dozen perfectly good but rejected students who could fill their space.
So we are collectively engaged in a college admission "arms race" that is almost a complete social waste, for once a set of "good enough students" or "good enough schools" has been identified, it probably doesn't matter very much which one you choose; or if it does matter, there is no way to know in advance what the right choice is. Hair-splitting to distinguish among excellent students (or schools) is a waste of time and effort; the degree of precision required exceeds the inherent reliability of the data.
Actually it's worse than a social waste. This intense competition for the best students and the best schools can actually be destructive -- even for the winners. It creates an orientation to education where only results matter. It makes the stakes so high that students can't afford to take risks or pursue idiosyncratic interests. Everything they do is calculated to produce better credentials. This pressure to achieve extracts an enormous psychic toll from high school kids. Psychologist Suniya Luthar has found that the incidence of substance abuse and psychopathology is as high among children of the affluent as it is among children of the inner city, in large part because of achievement pressures that the affluent experience. Psychologist Jean Twenge makes a similar point in Generation Me. And it's not just high school. The tape unwinds all the way back to pre-school, as nervous parents worry about securing their kids every little edge that might determine the thickness of the envelope they get from Harvard in the spring of their senior high-school year.
Given the futility and the very high financial and psychic costs of the current selection process, there is a step that institutions can take that will dramatically reduce the pressure and competition that many of our most talented students now experience. Schools can scrutinize applications and decide which applicants are good enough to be admitted. But then, instead of struggling to choose the "very best" from among the good enough applicants, schools can just place the names of all the "good enough" students in a metaphorical hat, and draw the "winners" at random. Though a high-school student will still have to work hard to be "good enough" for Stanford, she won't have to distort her life in the way she would if she had to be the "best."
Nobody likes the idea of important life events being determined by a roll of the dice, which may explain why in the four years since I initially made this suggestion, in The Chronicle of Higher Education, no institution has taken me up on it -- even as an experiment. But college admission is already a crap shoot, and our failure to acknowledge this is a collective exercise in self-deception. Picking a fifth of 5,000 "good enough" applicants at random might be just as good a way of producing a great class as the tortured scrutiny of folders that is the present practice.
This modest proposal may seem preposterous, but it isn't. There is little doubt that any random fifth of the applicants who might survive an initial screening would make a fine first year class at Harvard. In an article in the New York Times a few years ago, Harvard's Dean of Admissions said that they could go through the process, pick their class, and then chuck them all out and start again, and still have a fabulous entering class. Stanford could fill their entering class with kids who had perfect scores on their SATs and still have plenty with perfect scores left over.
Further, while admissions people like to believe that they have the discernment and diagnostic ability to look at 5,000 wonderful applicants and pick from them, with high accuracy, the 1,000 "superwonderful" ones, there is a huge literature on decision-making, much of it reviewed in a classic article in Science twenty years ago by Robin Dawes, David Faust, and Paul Meehl, that makes clear that people in such positions are much more confident of their abilities than the data warrant. In other words, picking a fifth of the 5,000 at random might be just as good a way of producing a great class as the present practice.
It is easy to understand why admissions officers might deceive themselves into inflating their powers of discernment. After all, every year they admit a class, and it does just fine -- even better than fine. So they must be doing something right. This is an example of what psychologists call "confirmation bias." The real test of discernment would be to do what the Harvard Dean of Admissions said: admit the near-miss rejects and see how well they do. To my knowledge, no one has done this particular experiment.
The benefit of random selection from a pool of good enough applicants is not just that it will save time and money. Far more important, making college admissions less competitive will allow good students to be students again, experimenting, following their interests, and correcting their weaknesses, instead of trying to look perfect in their applications. Adolescents will once again be able to devote at least some of their time to figuring out what kind of people they are and want to be, and come to college with a lot of that work already behind them. The result of this decreased selectivity, I'm convinced, will not be worse students at our most selective institutions, but more interesting, more curious, more mature, and more highly motivated ones.
And there is another potential benefit. We like to believe that the U.S. is a meritocracy. Success is about talent and hard work. Luck has nothing to do with it. It may be true, when systems are working as they should, that people deserve what they get. But the converse of this claim is certainly not true: people don't always get what they deserve. There just aren't enough rungs on the top of life's ladders for everyone to fit. If talented and hard working people are forced to confront the element of chance in life's outcomes when they (or their kids) fail to get into the "best" college, they may be more inclined to acknowledge the role of luck in shaping the lives of the people around them. And this acknowledgment may make them a good deal more empathic toward others, and a good deal more committed to creating more room, for themselves and others, at the top.