CA: Is there one make or break part of the application?
BD: There really isn’t. We don’t have any thresholds, we don’t have any minimums. It is not as though if your GPA or test score or the number of years of work experience is below a certain point, we won’t review you or we won’t consider you. We look at everybody, we consider everything.
Admissions officers like to talk about a holistic review, which sounds like a cop out term. But it is really true. Each piece of your application can have a different weight, a different significance, a different meaning, depending on what the rest of your profile looks like. So we don’t disqualify someone because some aspect looks lower than we would like, because there might be other things that compensate for that. We look at everybody with a pretty full lens to get a sense of your overall strengths and weaknesses and make an evaluation based on that.
For example, if your quantitative GMAT score is low, but you were an economics major and did very well, and are working in finance, we might rely on that data point less than if you were an English major who was working in the non-profit space, where we don’t have other data points to look at. So while saying something is holistic doesn’t give much insight, we address the relational nature of the different aspects of the application, and what that means; it gives a little more definition and contour around what we mean.
CA: This also underscores the idea that admissions is not run by Artificial Intelligence. These are admissions officers, taking each individual application on a case-by-case basis. You don’t see admissions going toward AI in the near future, do you?
BD: I don’t think so. I don’t think in the near future we will be completely algorithmic. As a former lawyer, I think there might be some Supreme Court reasons why that is the case. We cannot assign points to different aspects of the application. But I think that it’s also not the right way to approach admissions. We shouldn’t ever just give things to an algorithm and have it churn out an admit decision or deny result.
There is a machine learning part of admissions, in the sense that we can learn from past decisions and what those outcomes are. I think that component that is very valuable. We do already do this, to some degree, but we can always improve the iterative nature of the review process and learn from past decisions.
Because admissions is not pure science, it doesn’t mean it is not structured and it doesn’t mean it is not rigorous. We do have rubrics for how we make decisions and we try to be guided by behaviors and not just gut instincts and feeling. That can get dangerous, and allow bias and other bad things to creep in. We try to keep that in check as much as possible. We are human beings, so it is impossible to eliminate bias completely. But we are very conscious of that, and try to be structured and very rigorous about how we approach even the more artful aspects of the process.
Read on for more insights into the essay portion of admissions.
