<p>Good Luck!</p>
<p>Hunt, re your post #117, I meant that top 25% scores were were irrelevant for determining chances at the top schools. The scores of applicants to those schools are so uniformly high that everyone above a certain threshold with scores is essentially in the same boat-- and the school is looking at other factors. Maybe top 75% would be meaningful – that is, someone who was below the 25% level can assume that their chances of admission would be significantly diminished – but its the other factors that take over at that point.</p>
<p>In other words, I pretty much agree with what mathmom said in post #118. Once you have hit a certain threshold, different factors come into play. </p>
<p>I made that statement because I think that too often on CC there is this mantra of deciding reach/match/safety based on test scores. This does have value at less selective colleges – at a school where the typical admitted student has SAT’s of 1800, the ad com will probably sit up and take notice of a kid who applies with a 2200, and probably throw some merit aid that way as well. </p>
<p>Let’s look at it this way: my son had top scores, great grades, and almost no EC’s. Pretty much certain to be admitted at the vast majority of colleges based on his grades and scores alone – but completely out of the running at Ivies or equivalent schools. I figure 80% of the applicants to Ivies have great grades and test scores – and the 20% who don’t are eliminated from the start. So - to keep the math easy, if 1 out of 10 students are admitted to a college, and 2 out of 10 are eliminated because of weak grades/scores – then among students with the good grades & scores, the chances are now 1 out of 8. That’s still dismal odds – and grades and test scores aren’t going to help that group, because they’ve already passed that screen. So now you have to look at all the holistic stuff. </p>
<p>There are no published “stats” that can tell us where a kid stands, but I think that a parent who is capable of being realistic and objective should be able to get a sense of it.</p>
<p>
I wasn’t referring to “show the love”. I’m sorry that you don’t seem to get what I am talking about. </p>
<p>
Actually, I think most people can and do make different resumes, depending on the nature of the job being applied for. An applicant drafts the resume to emphasize the skills relevant to the job. I suppose if the background and experience is very narrow, there’s no point in doing that – but if a person has wide and varied experience, then it makes sense to emphasize the parts that are the strongest qualifiers for the job being sought.</p>
<p>OK, I believe the point I am going to make has been made by others obliquely (at least based on my skim through of the previous postings). I think there is a fallacy here and that is assuming everyone has the same probability of admission. If Harvard admits 8% of its applicants, it does not mean that every one has 8% chance. Some have 50% percent chance, some have a 10% chance and some have a 0.001% chance of getting in. Your chance depends upon a lot of factors (scores, grades, EC’s, essays etc.)</p>
<p>Now let assume that Harvard and Yale have the same probability of admitting a student and let us say it is 50% each. So if that student applies to both Harvard and Yale, there is a 25% probability he will rejected by both and 75% probability that he will be accepted by either Harvard, Yale or both. So applying to both Harvard and Yale has increased his chances of an acceptance to an ivy league from 50% to 75%. </p>
<p>If the same candidate also has a 50% chance in Princeton, then applying to Harvard, Yale and Princeton will increase the persons probability of acceptance to one or more of those three to 87.5% (1-.5<em>.5</em>.5).</p>
<p>Now if the probability of acceptance to Harvard and Yale is only 10% then there is only a 19% chance that the candidate will make it to Harvard or Yale, if he or she applies to both. If the same probability is applies to for Princeton also, the probability of acceptance has gone to 27.1%. </p>
<p>If a person has 1% chance of getting and they apply to three ivy leagues, then their probability of acceptance in or more of these institutions is statistically only 2.97%, not a significant improvement. So it really does not make sense to apply.</p>
<p>So here is conclusion and caveats:
- Applying to more colleges always increases your chances, but it may not increase to a point that you have a real chance of getting in. If you are already in the very strong zone, applying to more colleges will ensure that you get into at least one college. If you are in the weak zone, your chances go up but not to a place that it matters. Having a 3% probability will not get you in to a top college.</p>
<p>2) My example assumes that probability of getting in an ivy league is the same across the board. Each college is different and so one person may have 10% chance in one ivy league school and 15% in another. So applying to more colleges depends upon ones situation.</p>
<p>3) There are only a limited number of top colleges. If there were say 100 top colleges, a person with a 10% chance will have a good chance in one of them. Given that there are only a few, applying to all of them may not get an marginal candidate in.</p>
<p>4) These statistical analysis is based on a randomness and lack of bias. There are many confounding variables (your record, your school and demographics, race, if you are legacy or athlete, what the school is looking for at time e.g. more drummers for marching band etc.). So in many ways it is a weighted coin and hence applying to more colleges may not help if there is a bias working against that person. For example, if a person has low test scores, this may held against them even if everything else is great.</p>
<p>So to answer the OP’s question: Yes applying to more colleges in theory will increase your chances, but if starting basis is low, the increase in chance may not be significant. Also, this is not an easy question answer without knowing your statistics and if you post that then some one may be able to give a better answer. This where statistics leaves off and qualitative analysis comes into play as many things are situation specific.</p>
<p>Hope this helps.</p>
<p>
</p>
<p>The only students who have a “50% chance” of getting into H and / or Y are students who are legacies for the last 200 years and who have multiple buildings named after their families on the campus, with the promise of more. It’s a silly assertion.</p>
<p>^PG, I’m sure Rachael Flatt will also be admitted to all 9 of her applied schools.</p>
<p>(Figure skating on the brain with the Opening Ceremony in less than an hour!)</p>
<p>
</p>
<p>Great point! I have wondered about that for some time too. </p>
<p>I also suspect that the desire to boost yields has led each of the top elite schools to find ‘hidden gems’- students who who are not likely to be on the radars of their competitors. Kids from unknown schools, remote areas, or just students who are not necessarily strong ‘on paper’.</p>
<p>Pizzagirl, you are taking that one statement out of context. I was just using an illustration to answer the OP’s question that applying to more colleges does increase your chances of getting into a college, statistically speaking. The purpose of that illustration was to point out that in many cases even though your chances go up by applying to more schools, it may not be significant to make that student a serious contender. In some cases it does.</p>
<p>Hence there was no assertion in my statement, it was simply the use of a hypothetical example to illustrate the point of how individual probabilities affect one’s chances in applying to multiple schools. Also, in the end I did mention that there is a point where statistics leaves off as there are so many confounding variables.</p>
<p>Maze, agree with you.</p>
<p>One more question? “The biggest hint is the uniformly high yields at HYPSM averaging 70%.”</p>
<p>How do they do it? Looking out of top 25% stats that is I can think of.</p>
<p>
</p>
<p>What are the current actual figures for each of those five colleges?</p>
<p>^^ According to the link below, for the class of 2012, H=76%, Y=69%, S=72%, M=66%. P was reported as unavailable but I suspect it was pretty close to the rest.</p>
<p>[Admissions</a> yield for 2012 hits 39percent - The Chicago Maroon](<a href=“From Lance to Laundromats, band fad clasps campus wrists – Chicago Maroon”>From Lance to Laundromats, band fad clasps campus wrists – Chicago Maroon)</p>
<p>I also came across the following site (no idea how accurate it is or what the numbers mean):</p>
<p><a href=“http://mathacle.blogspot.com/2009/05/harvard-yale-princeton-stanford-and-mit.html[/url]”>http://mathacle.blogspot.com/2009/05/harvard-yale-princeton-stanford-and-mit.html</a></p>
<p>vicariousparent;</p>
<p>The last chart is a summary from the annual HYPSM cross-admit results here on CC. The data confirms a very low rate of cross-admission.</p>
<p>the most interesting part is that even on CC which attracts may Ivy bound students, only around 60 cross admits per year are recorded among the hundreds of admits to HYPSM. That is the total number of students admitted to more than one of HYPSM (among the CC crowd). As far as admitted to more than two of HYPSM the number was even dramatically lower. It is much less frequent than we assume.</p>
<p>If someone calculates the fraction of cross-applicants that ends up being cross-admitted, they would be able to come up with a pretty good measure of how much randomness there truly is in the system.</p>
<p>
</p>
<p>I think (and demonstrate below) that this FAQ is, like so many of the others, an unfortunate marriage of bogus logic and persistent reposting.</p>
<p>
</p>
<p>The ideas you call wrong are either literally, 100 percent, mathematically correct (e.g., it really is correct to multiply the probabilities), or correct in substance with minor errors that don’t affect the argument for sending multiple applications.</p>
<p>
</p>
<ol>
<li><p>Every applicant understands that the chances at any given school are not the same for everyone (in particular, they are not simply equal to some known base rate); superstars have a higher probability and underachievers a lower one. This means that whatever calculations are posted based on overall admission rates are not evidence of how people are reasoning about their individual admissions chances, but rather, of how they are attempting to model the effect of multiple applications, using as simple a formula as possible.</p></li>
<li><p>The “wrong” reasoning of replacing the per-school probabilities (however the applicant arrives at those) with the average is essentially correct. The error introduced in the calculations from this approximation is small, and does not affect the conclusion.</p></li>
<li><p>The only substantially incorrect statement one sometimes sees here, is to mis-estimate the overall chances of success (under either a few- or a many-application strategy) by mis-estimating the admissions chances at the individual schools (e.g., 10 percent when it might be 1-2 percent across the board for a given applicant). But such a mistake, if present, strengthens the argument for multiple applications: it is at the lowest probabilities that sending more applications makes the most difference to the chances of (an) acceptance. </p></li>
<li><p>Even the simple “wrong” calculation of adding the probabilities (so that doubling the number of applications to similarly selective schools basically doubles the odds of admission) is fundamentally correct: expected number of acceptances is a better measure of outcomes than whether admission occurs. That measure does have the additive behavior, and for low probabilities it closely approximates the ostensible statistic of interest, the odds of at least one acceptance. For high probabilities, it is a measure of how many offers the applicant can compare.</p></li>
</ol>
<p>
</p>
<p>That student reasoning (no quotation marks needed) is correct. The selections are independent (no quotation marks needed) in the statistical sense. Inferring this from the fact that committees act without coordination and in different locations is absolutely correct. </p>
<p>
</p>
<p>The students correctly apply the right formula, i.e., one minus the product of the rejection probabilities at the different schools. If the acceptance probabilities are not too low at the individual schools, this calculation shows that acceptance at one or more Ivies (or whatever the “reach” level of school) is made very likely by a large number of applications. If acceptance probabilities are low, chances are dramatically improved by multiplying the number of applications. An applicant doesn’t need to decide which is the case in order to act on these calculations, because the result is the same: send more applications subject to limitations of time, money and interest in attending.</p>
<p>
</p>
<p>A few questions: </p>
<p>-The above was posted to the AP Stats discussion list and nobody, not one statistics teacher or professor, commented on the minor detail that it’s completely wrong. Is that correct? </p>
<p>-That is, somebody seriously claimed that Probability(admission to schools A and B) is <em>not</em> equal to Prob(admission to A) x Prob(admission to B), and none of the list recipients raised an eyebrow. Did I understand that right?</p>
<p>-And those totally wrong comments have been reposted as (the basis of) an authoritative, knowledgeably sourced FAQ on College Confidential for several years, is that also correct? </p>
<p>Answers should be quite revealing!</p>
<p>
</p>
<p>uh, you absolutely can and should use the multiplicative rule. The colleges act independently in the statistical sense of the term <em>because</em> they act independently in the layman’s sense. An AP Stats teacher who tells you otherwise is confused at a very basic conceptual level and should not be teaching the subject.</p>
<p>Hi, siserune, I responded to tokenadult’s post with some hypothetical mathematical illustrations in posts #43, #46, and #57. The statement that the decisions at H and Y are not statistically independent is correct, applied to the aggregate outcomes of the pool of applicants. However, this lack of independence results from the existence of underlying variables (strength of applications), giving each applicant personal chances (pH and pY) at H and Y that are correlated with each other. Nevertheless, for each applicant, the decisions at H and Y are made independently and do not affect each other, so that the multiplicative probability calculations are correct–if pH and pY are known.</p>
<p>The card game analogy in #57 is intended to show how correlated underlying variables (in that case, deck composition) can lead to a lack of statistical independence in the overall tabulated data for a group of players, even though for each individual player, the multiplicative probability analysis is valid.</p>
<p>I think that for many applicants, the actual probability of admission is very near to zero, and multiple applications really do not help. For a few, it is close to 1. I think the multiple application strategy really starts to make sense when the applicant’s chances are about twice the raw odds in the pool–just personal thinking, but this would mean that the applicant has something going for him/her that might resonate with admissions at a particular school.</p>
<p>I suspect that the number of cross-admits is actually much greater than 60, but the people in question just aren’t posting it. On the other hand, it cannot be extremely high, or the yield figures at HYPSM+C could not all be as high as they apparently are–actual data on this would be nice. Caroline Hoxby did a study some time ago on the preferences of the cross-admits. Aside from the “Harvard is clearly the top choice of all reasonable people” and “Caltech? Where did that come from? Must be an artifact of the methodology” prejudices that are apparent in the study, it’s useful–and might give a better indication of the number of cross-admits.</p>
<p>Don’t forget that early admissions can also skew the potential number of cross-admits.</p>
<p>siserune:
I was getting all ready to debate you when I realized I pretty much agree with everything you wrote here.</p>
<p>I think the only possible way multiple applications could hurt, is if the mere process of completing multiple application affected the underlying probabilities of admission to each school for that particular applicant. What several people have mentioned, what one poster cleverly referred to as the “Heisenberg Uncertainty Principle of Admissions”, and what you allude to with </p>
<p>
</p>
<p>I personally believe, although I have no data to back it up, that the multiple application downside is generally less than the upside.</p>
<p>I believe there is a point of diminishing returns, but that may not be the same for each applicant–and some applications are more similar than others.</p>
<p>^ ^ ^ ^ </p>
<p>CURRENT figures on cross-admits, from all the colleges in question (which, it appears from this thread, are not just the eight colleges that are in the Ivy League </p>
<p>[Ivy</a> League - Wikipedia, the free encyclopedia](<a href=“http://en.wikipedia.org/wiki/Ivy_League]Ivy”>Ivy League - Wikipedia) </p>
<p>proper) would indeed be helpful, the more detailed, the better. Every year there are a few people in the happy position of having been admitted to all eight of the eight colleges in the Ivy League. Every year there are quite a few more who applied to all eight but were admitted to none. And there are also various cases of applicants applying to some but not all of the Ivies and being admitted to some subset of the colleges to which they applied. It would be helpful to have exact published data on all the cases that have occurred in recent years. </p>
<p>But I hope no one seriously entertains the idea, which was the basis of quite a lengthy thread a few years ago, that simply applying to all eight so boosts one’s chances that an applicant can be assured of getting into at least one of the eight. Wrong idea 1 is that merely increasing the number of applications steadily increases probability of admission to some subset of the whole athletic conference in a straightforward, readily calculated way. And I hope too that students are unafraid of applying to as many highly selective colleges as they wish (possibly including all eight colleges in the Ivy League) if the students have some reasonable preparation for thriving at each of the colleges to which they apply, and sufficient time, energy, money, and other resources to make thorough applications. Wrong idea 2 is that Ivy League admission committees envy or disparage applicants who apply to other Ivy League colleges. Each idea is wrong in a different way, but both ideas are wrong and might mislead students into suboptimal admission application behavior.</p>
<p>Most kids I know who got into an elite school were also rejected by several. I can’t imagine my son will apply to fewer than 15 or so to increase his odds (unless he gets in on EA somewhere really good, in which case wd eliminate safeties). Common app makes this far less painful than in “our day”.</p>