<p>Even if a school had rejected applicants who were more desirable under that school’s OWN CRITERIA in favor of less qualified students, I don’t think that this necessarily means that the school was engaging in gamesmanship or was necessarily trying to “protect their yield” for unseemly PR purposes. Couldn’t it simply be prudent management?</p>
<p>Here’s an example. Let’s say that a school historically has a 33% yield and they want a class of 2000 students. So, before the applications come in, they decide that they are going to accept somewhere in the range of 6000 students. </p>
<p>Then the applications come in. Assume that they get 25,000 applications, but 5,000 of those are no-brainer rejections. So they are left with 20,000 qualified applications.</p>
<p>Now assume that the school is for some reason hot – on an upward trajectory – and the applicant pool is better than the school has ever seen. The middle 50% of qualified applicants – the middle 10,000 – is stronger than any applicant pool that the school has previously seen. But the kids in this pool of 10,000, while better than usual, are not signficantly different from the kids who historically have matriculated at the school. Because of this, the school is confident that its past data is applicable to this population. </p>
<p>Now let’s look at the top 5,000 students in the applicant pool. Let’s assume they are off the charts, much better than the kids who typically enroll at the school. The school has very little experience with this population. Almost all the students in this range who have been accepted in the past have declined their offers, but the school has never had anywhere near this number of applicants in this range.</p>
<p>Finally, assume that there are all sorts of institutional needs that have to be met, from oboe players, to sports team participants, to humanities majors, to a cappella group performers, to diversity goals, to gender balance, etc. </p>
<p>Now, let’s say the admissions office has three goals: (1) To make sure you have an entering class of as close to 2000 students as you can get WITHOUT going over, (2) to make sure that the class will meet all of the institutional needs that you are tasked with filling, and (3) to make sure the class is AS STRONG AS OR STRONGER THAN past entering classes.</p>
<p>What would you advise the admissions office to do? Remember, the school has already decided that it needs to accept 6000 students to get a class of 2000. </p>
<p>Would you advise them to accept ALL of the 5000 of the off-the-charts students, and only 1000 of the middle 10,000, and to cross their fingers and hope that their previous experience and data will apply equally to the unknown population? In other words, would you advise them to accept mostly “REACH” students and hope for the best?</p>
<p>Or would advise them to try and shape the class and to meet their institutional needs primarily from the population that historically has matriculated at the school and for which you have scores of data, selecting, say 5000 from the middle 10,000? If you go this route, would you then say, “What the hell, let’s accept all 5000 of the top group; they probably won’t come anyway?,” thereby increasing the number of acceptances from the 6000 you planned on to 10,000, and hoping and praying that the yield numbers work out and that you don’t end up with a class of more than 3000? AND thereby increasing your acceptance rate?</p>
<p>Or would you accept only some portion of the top group – perhaps the kids who give you some reason to believe that they will act similarly to the population you regularly draw from – and increase the number of acceptances slightly but not dramatically over the planned 6000, allowing you to accept more students that are a “high reach” for your school but still allowing you to manage the process based on data and experience? (Keep in mind that this last approach will allow you to meet ALL of your goals and still end up with a STRONGER class than before.)</p>
<p>Wouldn’t this last approach lead to results that have been called the “Tufts Syndrome”?</p>