« Reading Proofiness 1 | Main | Checking the numbers means more than checking just the numbers »


Feed You can follow this conversation by subscribing to the comment feed for this post.


I think the argument for formalizing the coin flip as the Harvard admissions process was that it would reduce the cost (in terms of time) of admissions for both applicants and administrators. If applicants only have to show that they are above the lowest 10-20%, they won't have to spend ridiculous amounts of time and money on application consultants or test prep programs to try and eke out the last little bit of advantage in their application.

Likewise the admissions committee will only have to determine whether a student is approximately qualified or not, they won't have to read long essays or writing samples or conduct interviews with every applicant looking to get a slight edge. Even though you and I know that the admissions process now is essentially a coin flip, most applicants (and parents of applicants) don't see it this way. If you formalize admissions as a coin flip, then people will perceive it as it is, namely random, and act rationally.


...the existing admissions decisions can be considered the first coin flip. If the 80-90% are truly indistinguishable, then the admissions committee is only flipping a coin...

Wouldn't a truly random process be superior to the pseudo-random decisions made by an admissions committee, which might hold hidden biases?


john: I get your point but I think powerful forces will prevent you from formalizing admissions as a coin flip.

Brad: yes, the current process is random only if in aggregate the admissions officers do not introduce hidden biases. See my last point about better science vs. pragmatic policy-making.


I think the essay is making a more subtle argument than "Harvard admission is a lottery." I think it's making the point that because no one wants to say that a process carrying a high level of significance( e.g., from the signalling from the Harvard credential, earnings power, etc.) is random, the admissions office generates distinctions among candidates that don't actually have the ability to predict outcomes (such as qualities of leadership, extracurricular activities, interviews, etc). These distinctions are things that they can point to in order to say "we've built a wonderful class of admitted students." And so these distinctions in turn drive counterproductive behaviors among prospective candidates - activity overload, for instance. The author's idea is that you could reduce these behaviors if you gave up thinking that these other factors were really useful and just went ahead and called the admissions lottery a lottery. I doubt it will happen!


Just two points.

1) "80 to 90 percent of Harvard applicants are qualified to be here. Harvard should identify that 80 to 90 percent" How? By means of a test, I suppose. Then, you say that "the two-way split is too simplistic. [...] there are three groups: the top 10-20 percent are sure-ins, the bottom 10-20 percent are less qualified". I think it is better to refine your idea. I would split all applicants in 25% top, 25% above-median, 25% below-median and 25% bottom. My first cousin maybe would choose to make a 20%-20%-20%-20%-20% cuts, and my last cousin (I have many cousins) could argue that a 1%-1%-1%-...-1% (the 100 percentiles) is more elegant solution. Now the problem is: how to identify the percentile each applicant belong to? By means of a test, I suppose. If I am right, such percentile test is practically equivalent to the real admission test, so the need of a lottery becomes insignificant.

2) "if two football (soccer) teams could not find a way to win after 90+30 minutes, we might as well flip a coin to decide the outcome, leaving it to "chance"; or just accept a "tie" as the outcome." Is your proposal serious or it's a joke? I ask only because I think that many european soccer supporters wouldn't be very happy of it...


Antonio: 1) Point taken; that's why I don't like Salmon's description of it as "randomizing admissions"... it's randomizing for a subset of applicants judged to be equally qualified. 2) Do ask your soccer fan friends and let me know what they say; as a soccer fan, I see penalty kicks as practically the same as flipping a coin. Penalty kicks (missed ones) can destroy players and I'm sensitive to that.

The comments to this entry are closed.

Get new posts by email:
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR, Wired.

See my Youtube and Flickr.


  • only in Big Data
Numbers Rule Your World:
Amazon - Barnes&Noble

Amazon - Barnes&Noble

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee

Next Events

Jan: 10 NYPL Data Science Careers Talk, New York, NY

Past Events

Aug: 15 NYPL Analytics Resume Review Workshop, New York, NY

Apr: 2 Data Visualization Seminar, Pasadena, CA

Mar: 30 ASA DataFest, New York, NY

See more here

Principal Analytics Prep

Link to Principal Analytics Prep