<p>“Washington University at St. Louis and some other schools (The Tufts syndrome) are reputed to reject candidates who they think will go a higher rated institution.”
What I have heard in the 6 years I have been on CC, and have derived from the data at my own son’s HS, is that Wash U has the reputation of also WAIT LISTING top students who DON’T apply there ED. Wash U really wants to “feel the love” of an applicant. For them applying ED is the best way to indicate that intent, and they reject or wait list thousands of RD tip top students who are accepted at other top colleges, partially in order to keep their yield numbers up.</p>
<p>
</p>
<p>This is definitely true. I really lament the fact that I didn’t apply to more schools. I foolishly only applied to HYP and the university of Texas, it definitely makes me feel slightly inferior to some of my other classmates who got into multiple ivies because I only got into one. Not necessarily because I wanted to go to those schools (because I still would have chosen Yale over them) but it is definitely good validation to know that you got accepted to multiple Ivies. Getting into one ivy and not the two others made me feel like my admissions was a fluke.</p>
<p>mazewandererr-
I suspect you may be in for a pixel-lashing from siserune for a couple statements in here, but one thing you wrote made me curious -
</p>
<p>I’d agree this is possibly true, but on top of this they would also have to gauge the likelihood that the applicant would apply to those schools, since in many cases they supposedly have no way of knowing. I think that’s a bit of a stretch.</p>
<p>bovertine: It is in CC that I first came across the Tufts Syndrome, where colleges reject applicants who they feel will go elsewhere. Again, I think top institutions would not consider this as factor, but there are a lot of give away and an experienced counselor could make some judgments. Again the idea is not to say this happens, it is point out the admissions from one top school to another are not totally independent.</p>
<p><a href=“QM:”>quote</a> in response to your question whether the multiplicative approach is the only viable one for an individual applicant who is working with pH, pY and pP (as opposed to rH, rY, and rP): yes. I think so.
[/quote]
</p>
<p>I’m glad we agree on that much. However, the extent to which this is the “only viable option” is much, much stronger than you indicate. What is really true is that any mathematical model of the selection (whether deterministic, probabilistic, or some combination of the two) that lacks independence has to explicitly involve spooky actions at a distance.</p>
<p>
</p>
<p>It isn’t necessary to assume that probability plays any role. Independence must still hold: revealing Harvard’s (random or deterministic) decision about Joe’s application provides no information about Stanford’s (random or deterministic) decision. Any model of the selection process that lacks this feature forces Stanford’s decision algorithm to depend on variables that are specific to Harvard’s processing of Joe’s application, such as whether his application reader had a headache. </p>
<p>The obvious routes for information sharing — collusion, leaks, espionage, ED, athletic recruitment — were ruled out by assumption. Alternative channels are supernatural, and I assume you would agree that models involving telepathy can be deemed “not viable”.</p>
<p>
</p>
<p>Please note that tokenadult’s AP Stat teacher source was originally quoted in the form of an equation, i.e., “Prob[H and S] is not equal to Prob[H] Prob[S] because those aren’t independent events”, when clearly discussing single-applicant outcomes.</p>
<p>Please also note that as soon as the non-independence claims were posted in the first thread where this came up — see link above — the guy arguing with tokenadult (bob9775) called it out as nonsense and said that he “questioned the teacher’s expertise”. It is a professionally disqualifying level of nonsense for someone whose job is to teach statistics, as it indicates a serious confusion about basics, such as the difference between probability and statistical inference. </p>
<p>We have had this nonsense served up for several years now as authoritative, FAQ-ed, statistically certified truth. There has been no retraction of any specific nonsensical, incorrect or misleading statement out of all those that were pointed out (and in this particular case, analyzed well past the point where the error is made manifest). To the extent that reliability of information in this forum is important, it is important to make very clear which factual or mathematical claims here are correct, and which are not, with no room for equivocation that leaves the answers in doubt. </p>
<p>
</p>
<p>“Have you taken statistics?” and “my source is a stats teacher” are not what I would call interesting ideas at the math-to-English interface. As for your discussion of cross-admits and non-independence, I will post about it later.</p>
<p><a href=“mazewanderer:”>quote</a></p>
<p>2) Even with this the multiplicative approach is the correct statistical approach. Statistically the confidence interval (in laypersons terms though not really correct, the margin of error) will be very wide. Generally one would expect that multiplicative approach will lead to over stating of the results, and you will not have confidence in the results due to the introduced bias and the wide margin of error.
[/quote]
</p>
<p>Given that non-multiplicative approaches invoke supernatural forces, (a) I don’t think there is any alternative method of modelling, and (b) the more stable, more reliably estimated and strategically informative quantity is the expected number of acceptances. This is just the sum of the acceptance probabilities, whether those are assumed, estimated, modelled, guessed or prayed for. It is also completely oblivious to the presence or absence of correlations, in case one prefers the ghosts and goblins.</p>
<p>
</p>
<p>Wouldn’t a quick calculation based on best-, middle- and worst-case sets of acceptance probabilities (even just a single probability when dealing with similar schools, such as Ivy League) tell you most of what you need to know? Probabilities are also available from online chances calculators based on databases of hundreds of self-reported applicants (at each school) who give their SAT and GPA and other data together with the outcomes. Crude but far from uninformative.</p>
<p>
</p>
<p>By “probabilities are additive at the low end” I meant that the expected value (sum of probabilities) is almost the same as the probability of one or more acceptances, which was the thing we are supposedly most interested in. This is because multiple acceptances (which account for the difference between the two calculations) are rare enough to ignore, and that rarity is due to independence. I’m sure you knew this but it doesn’t hurt to say it in more detail.</p>
<p>
</p>
<p>Mult approach compared to what other approach? If you really care about bias you can simulate the presumed distributions, and relatively simple R code for this can be posted to CC if anyone cares to code something up.</p>
<p>
</p>
<p>This doesn’t spoil the independence (multiplicativity) of Joe the Applicant’s outcomes at different schools. As long as Brown’s determination that an applicant is Harvard- or MIT-bound is based on its own locally available information and not information that depends on processes specific to Joe’s application process at Harvard or MIT, we will have multiplicativity for Joe’s outcomes at Brown, MIT and Harvard. If Brown hacks into the Harvard admission files, that could introduce correlation.</p>
<p>re: #187</p>
<p>If a successful applicant posts information that identifies him or her on facebook or CC regarding EA acceptance at Yale or a “likely letter” acceptance from Dartmouth, it is possible for another school to know of these outcomes before making a decision of their own. I don’t know that schools would actually do this, but they could. It would therefore be advisable for the applicant applying to multiple ivies not to post such information until all schools have made their determinations.</p>
<p>Along the same lines, what does EA acceptance to Yale or “likely letter” acceptance at Dartmouth say about the probability of being accepted at a 2nd ivy?</p>
<p>chandelle,</p>
<p>Your point about GC is true maybe for the most of high schools but DS school is the largest school in the nation having graduation class 1400 kids plus every year. GCs have to take care senior and junior students at the same time. (S)he hardly know my kid. All paper works for letters and grade reports are computerized and have a very strict rule to follow. So one or more application did not impact the content of (s)he wrote for most kids. Everything is a number. I am not complaining actually I do like his school. Every year they send a lot of students to very selective schools. Since school is so large he has more choices of classes and actives he can do.</p>
<p>So the impact from GC is very limited. For sure DS doesn’t get a very personalized letter but nothing you can do. It is a public school and too big for GC knowing everyone.</p>
<p>
As others are explaining, it has no impact on the probability of being admitted elsewhere, but it does provide an indication that the candidate is qualified to be admitted at at least one Ivy.</p>
<p>Like to update what happened to S for this thread.</p>
<p>Result : All four ivies rejected him. No different for adding three more schools later on after the first one was in his list. However, he learned the lesson that at that level he needs well prepared. No pain no gain. No free lunch. Let’s see after four years later if he learned.</p>