To the resident DCUM College Admissions Stats Guru...
![[Post New]](/jforum/templates/default/images/icon_minipost_new.gif){kind=icon}
Anonymous
|
From time to time I see a poster on here who contributes to discussions by manipulating data and spitting out conclusions which I mostly can follow. However, being stats challenged, I still have a couple of questions and I'm hoping you can chime in with a Stats for Dummies primer for College Admissions (and to everyone else: please feel free to interject your own questions on this thread if Stats Guru is willing).
My primary question is about a frequent response I see here related to Early Admissions and chances for acceptance, especially as related to subsequent Regular Decisions. Can you please explain why sometimes you say that the chances for acceptance for X Student for are Y% in ED but dribble off to Z% in the RD rounds separate and apart from the published acceptance rates by those colleges for those rounds. You seem to be making some kind of anti-cumulative argument that I can't follow related to the merits of an X Student's application. I can't find an example thread at the moment but you've stated this several times and I'm wondering. |
Anonymous
|
My guess is the poster is weighting in information about the fraction of the admitted class a certain school is known to accept ED vs RD.
|
Anonymous
|
Betting everything on one school ED is great, *if* you get in. If you are deferred or denied you have squandered your chance to possibly apply EA (non-binding to several schools instead of one).
Regular decision odds are getting tougher at top 50 schools.... However, every school is different. Looking for a statistical theorem that can be applied equally to all schools is unwise. Consult the "Common data set" for the schools you are interested in. |
Anonymous
|
How much of a difference ED / SCEA makes depends on the school you are talking about. Have you seen this data?
Of course what you can't see in this is how many "hooked" students were also ED students -- meaning they would have gotten in regardless of when they applied. https://www.iecaonline.com/wp-content/uploads/2018/05/Early-Decision-and-Regular-Decision-Acceptance-Rates-April-2018.pdf?fbclid=IwAR3oXgn9bEVEhtxyp8BHS36zyX4r3jUAqFx7pLNbiIPXEL9nYnWMqZl_hsg Most selective colleges and universities put out a news release about its accepted students and include the acceptance rates for early vs. RD candidates. |
Anonymous
I'm not the Stats guru, but here are the few points, I've gleaned from looking at it: 1. The composition of applicants to ED may differ from RD. ED both includes recruited athletes, legacies that may have lower stats than average but considerable "hooks." Outside of those cases, people usually apply ED to a school that they are not a sure thing at, but have a reasonable shot at. So likely less variability in scores/applicant quality once you take out the athletes/legacies. It's helpful to compare the stats of students admitted ED and RD to get a sense of the advantage at a particular school. 2. Schools know that each ED admit is a definite "yes" so they are inclined to admit those who are a good fit for the school's needs. They don't have to worry about yield with ED. So even if they fill 50% of the class with ED, doesn't mean your chances are so much worse applying RD because they will accept far more candidates than will attend RD. 3. A NYTimes article that I can't find right now reported on a study on ED that controlled for these and likely other factors and found that the overall advantage of applying ED to selective schools conferred an average advantage of about 1-2% chance or something around there. So an advantage but not as outsized a one as sometimes thought. But individual schools vary. For some, controlling for all these factors, ED was more of an advantage. For others, it had no advantage. |
Anonymous
| Does anyone know if there is a source to figure out what percentage of an incoming class is hooked? That might help us calculate the chances of a regular kid with the stats but without a hook |
Anonymous
It certainly isn't published in the CDS or anywhere publicly I've ever seen, and certainly not a school by school. Some media have written on the question though https://www.newyorker.com/news/q-and-a/an-investigative-journalist-on-how-parents-buy-college-admissions https://www.theatlantic.com/education/archive/2018/10/college-sports-benefits-white-students/573688/ https://www.thecrimson.com/article/2018/6/30/athlete-admissions/ https://www.nytimes.com/roomfordebate/2011/11/13/why-do-top-schools-still-take-legacy-applicants/in-college-admissions-athletes-are-the-problem https://www.ncsasports.org/blog/2014/04/07/recruited-athletes-with-sub-average-academics-can-receive-preference-in-admissions/ |
Anonymous
A recent WashPost article said 40%. That's a huge generalization so don't take it as fact. |
Anonymous
|
Some schools take a huge portion of the incoming class from ED applicants (e.g. U Rochester took ~60% of the class of 2023 via ED this year). That makes RD much more competitive.
ED is for the rich, basically. |
Anonymous
Anonymous
|
According to this opinion piece 40% of spots in top colleges go to hooked applicants and they have a 2 - 20 times higher likelihood of admission depending on what their hook is. So that tells me that the acceptance rate of an unhooked student at an Ivy is much lower than the published acceptance rate
https://www.nytimes.com/roomfordebate/2011/11/13/why-do-top-schools-still-take-legacy-applicants/in-college-admissions-athletes-are-the-problem Thanks to the PP who posted the article in their post |
Anonymous
Can't you still apply EA to other police but you agree to withdraw those applications if you get in to the ED school? |
Anonymous
Yes. You've only squandered your chance to apply ED or SCEA elsewhere. |
Anonymous
| Op again: what I was really driving at was the overall affect of all of the number of applications submitted including ED and RD. If X Student puts in 1 ED plus 9 RD applications, the Stats Guru seems to massage the data in a way that suggests that Application #1 (presumably the ED application) has Y% chance and each successive Application 2-10 has a diminishing Z% chance of acceptance. The percentages that the Stats Guru comes up with don't appear to follow the published acceptance rates but rather some type of calculation that X Student will get accepted at ANY college. That's what I am trying to figure out. |
Anonymous
They’ve developed their own algorithm to predict based on the data publicly available, which may or may not be accurate. |