3 ms·

> They will never fully open the black box, because people will optimize for its contents. This is laughable. There are many, many people who have worked at ad

by csa 8y ago

> They will never fully open the black box, because people will optimize for its contents. This is laughable. There are many, many people who have worked at admissions at elite schools, and they will tell you the contents of the "black box". What they would say goes something like this: 1. There are 10-30% of the entering class that we know we want. 2. The rest we have to fill with essentially clones (smart but unexceptional students). 3. We choose the clones that we think will make our school a better place in the short term and long term. Due to the multitude of variables that are introduced in selecting from the clones, it looks like our decisions are capricious. Maybe they are for the clones, but the clones aren't where we are looking to find our big winners in the admissions process. Read some of the stuff by Cal Newport. He's closer to the right track than anything else I've seen.

bilbo0s 8y ago
That doesn't tell you the contents of the black box, it only tells you the output of the black box. What people want to know is, how does the school "know" they want the initial 10 percent? How does the school "know" what clones (drones) they want? Etc, etc, etc.
- csa 8y ago
  For the top 10-30%, I will give you two heuristics. 1. Sniff test: You read something in their application and say to yourself “damn, that’s really cool”. 2. More robust test: They have worked on a hard problem and have done something interesting with it. This could be research, starting a business, starting a non-profit, something substantive in their community, something creative (usually externally validated like being a top performer), etc. Again, check out Cal Newport’s stuff. He says very specifically that it’s not a formula, because it’s not (I suppose it is at one level of abstraction, but...). For the rest of the class, there are a ton of variables that are usually out of the control of the applicant (one variable being who is on the admissions committee that year). Again, a quick heuristic is to ask what the applicant is giving the university. If the answer is not fairly compelling, then their chances of getting in decrease dramatically. If you’re looking for specifics, these are harder to find, and generally vary from year to year depending on the applicant pool. A few random ideas: - Kushner’s dad pledged $2.5 million over 10 years. It’s probably more than that now. - PA residents get preferential admissions treatment for Penn. They need something else going for them if their grades and scores are subpar. - Each elite boarding school gets a range of spots any given year. - Academically strong athletes are strong, esp. if they are great athletes by Ivy standards. - Academically weak athletes better be damn good and better have at least one more desired trait. Some of these folks do 13th grade at an elite boarding school to prep them for elite universities in a number of ways. - Being a legacy doesn’t get you nearly as far as people think it does. The number of legacy rejects typically far outweighs the number of legacy accepts. Again, that applicant better have another desired trait or two. - Having social capital and being able to demonstrate that goes a long way. What is “social capital”? Well, an applicant might say something like “I want to study economics at $SCHOOL because I want to understand better the fundamentals that my father, the finance minister of $COUNTRY, regularly discusses with me about his job and how that impacts my country and the counties around it.” The size of $COUNTRY can vary wildly, but these folks are unassumingly at elite schools. Does that better answer your question?

by csa 8y ago

bilbo0s 8y ago

That doesn't tell you the contents of the black box, it only tells you the output of the black box. What people want to know is, how does the school "know" they want the initial 10 percent? How does the school "know" what clones (drones) they want? Etc, etc, etc.

csa 8y ago
For the top 10-30%, I will give you two heuristics. 1. Sniff test: You read something in their application and say to yourself “damn, that’s really cool”. 2. More robust test: They have worked on a hard problem and have done something interesting with it. This could be research, starting a business, starting a non-profit, something substantive in their community, something creative (usually externally validated like being a top performer), etc. Again, check out Cal Newport’s stuff. He says very specifically that it’s not a formula, because it’s not (I suppose it is at one level of abstraction, but...). For the rest of the class, there are a ton of variables that are usually out of the control of the applicant (one variable being who is on the admissions committee that year). Again, a quick heuristic is to ask what the applicant is giving the university. If the answer is not fairly compelling, then their chances of getting in decrease dramatically. If you’re looking for specifics, these are harder to find, and generally vary from year to year depending on the applicant pool. A few random ideas: - Kushner’s dad pledged $2.5 million over 10 years. It’s probably more than that now. - PA residents get preferential admissions treatment for Penn. They need something else going for them if their grades and scores are subpar. - Each elite boarding school gets a range of spots any given year. - Academically strong athletes are strong, esp. if they are great athletes by Ivy standards. - Academically weak athletes better be damn good and better have at least one more desired trait. Some of these folks do 13th grade at an elite boarding school to prep them for elite universities in a number of ways. - Being a legacy doesn’t get you nearly as far as people think it does. The number of legacy rejects typically far outweighs the number of legacy accepts. Again, that applicant better have another desired trait or two. - Having social capital and being able to demonstrate that goes a long way. What is “social capital”? Well, an applicant might say something like “I want to study economics at $SCHOOL because I want to understand better the fundamentals that my father, the finance minister of $COUNTRY, regularly discusses with me about his job and how that impacts my country and the counties around it.” The size of $COUNTRY can vary wildly, but these folks are unassumingly at elite schools. Does that better answer your question?