Do you think my son has the qualifications for ivy league?

Parent Cafe
181

<p>Good summation JHS.</p>

182

<p>Cubsfa: Your post was excellent. A cautionary tale that should be posted elsewhere as well. If you had posted your qualifications before admission decisions, I would not have predicted those results for you at all. I think you're right about the area you live in making a big difference. Perhaps you'll have Georgetown Law in your future, though.</p>

183

<p>


</p>

184

<p>Has anyone posted the link about Andison? That certainly relates to this thread...an extremely talented and well qualified student who we all thought had a decent variety of schools on his list...his story has a happy ending, but it certainly speaks to the fact that being well qualified does not guarantee an acceptance to college. I'm not all that adept at posting links like that....maybe someone else could...not that I think the OP is reading this...the OP seems to have disappeared.</p>

185

<p>
[quote]
There are two sides to the analysis: (1) Very high test scores are not as unusual as people think. (2) The majority of kids accepted at the most selective colleges don't have very high test scores.

[/quote]
</p>

<p>I don't know how you can arrive at any of these conclusions. I see the data saying the exact opposite. </p>

<p>1. Number of high scorers: Clearly from the data posted yesterday I arrived at only 18,984 individual SAT scores over 2,200 in a single sitting which is the 99th percentile cutoff. This correlates strongly with the number of NMFs each year of about 20,000. </p>

<p>2. Numbers of high scorers admitted to top schools If you take the 25%/75% percentiles for HYPSM schools from the CDs reports, they are all in the 700/790 range guaranteeing a median above the 99th percentile of SAT scorers. This means at least half of their admits or about 3,500 students are from the 2200+ group. Even more interesting is that based on the 75% percentile at least 25% or 1,700 students are from the 2300+ group, a group with less than 5,000 students nationally. In conclusion, HYPSM consisting of only 5 schools absorb nearly 20% (18.5%) of the 99th percentile scorers (2200+) and over 35% of the superscorers over 2300. </p>

<p>If you take the next ten top national universities by selectivity (Dartmouth, Brown, Penn, Cornell, Columbia, Duke, WashU, Nothwestern, Johns Hopkins), their selectivity is only slightly lower than HYPSM and they all have a 75th percentile score well into in the 2200+ scoring group. A good estimate of the enrolled number of high scorers is provided by the Brown university statistics. They show 541 and 510 enrolled students with scores greater than 750 in math and verbal respectively. This corrsponds to 37.% and 35.4% of enrollment respectively. One could conservatively estimate from these numbers that at least 33% of students have combined scores over 1500 hundred and would be in the same group as th2 2200+ scorers. With a total enrollment of about 17,000 students that means at least an additional 5,600 students or 29.7 % of the total 2200+ scorers re absorbed by this group. </p>

<p>In short, over 48% or nearly half of the SAT 99th scoring percentile are absorbed by only 15 schools. </p>

<p>3. Percentage of high scorers admitted to top schools
We should first consider that many of the top scoring applicants do not even apply to the top 15 private universities or enroll in top LACs, accept scholarships at second tier universities or enroll at prestigious public universities such as UCLA, Berkeley, UVA, UNC or Michigan. </p>

<p>From the Brown U. 2006 statistics, as example we find that less than only around 4,000 students with SAT scores greater than 750 in either math or verbal applied in 2006 out of around 18,000 applicants or about 22% of the total. Even Harvard only claims around 10,000 applications with either math or verbal score over 700. Even if generously estimating that half of these or 5,000 are 750+ scores, that is less than 25% of the applicant base. And it is significantly easier to score a single 750+ score than achieve a 2200+ single sitting, so in effect the true number of 2200+ applicants is probably no more than around 20% of the applicant base at any top 15 schools. </p>

<p>So what conclusions can be drawn from these statistics:</p>

<p>1. Top 15 private universities enroll a disproportionate amount of 99th percentile SAT scorers. Out of a total group of 18,900 they enroll 9,100 individuals or close to half of the entire group.</p>

<p>2. Top scorers are admitted at a much higher rate than other applicants. Even in the top schools top scorers rarely represent more than 20% of total applicants. On the other hand they account for close to 50% of enrolled students.</p>

186

<p>xxxxxxxxxxxx</p>

187

<p>"1. Top 15 private universities enroll a disproportionate amount of 99th percentile SAT scorers. Out of a total group of 18,900 they enroll 9,100 individuals or close to half of the entire group."</p>

<p>Just to keep the record straight, you would need to take account of the fact that, unlike the College Board stats,virtually all the elite private schools take the best scores from combined sittings. That is, more than the 18,900 would have 2200+</p>

188

<p>You don't need to take that into account to see that Cellardweller's statements are absolutely correct (if anything they understate the case), as they have been throughout the whole thread. The ranking of applicants by this or that version of the SAT metric will be substantially the same as long as it gives comparable weight to both math and verbal. Universities may report maximum scores but they see the whole test profile including SAT I, II, AP, etc, and the SAT data are really just a proxy for "level of objective test credentials".</p>

189

<p>sisrune is correct. The 99th percentile 2200+ single sitting is approximately equivalent to a 2250 multiple sitting combined score or 750 average math, verbal and writing. Either of these metrics would yield largely identical results. There is little data on how 2300+ students would fare compared to 2200+ students. There is some evidence that the extreme scorers get a boost, but probably not as pronounced as a 98th to 99th percentile. Rank and GPA at that level may be the greater discriminant. HYPSM enroll nearly exclusively out of the top 5% (and mostly top 2%). Top 15 schools enroll virtually all their students from the top decile (and mostly top 5%).</p>

190

<p>
[quote]
In short, over 48% or nearly half of the SAT 99th scoring percentile are absorbed by only 15 schools.

[/quote]
</p>

<p>Or, to put it another way, over half of the 99th percentile kids do not go to these schools -- and very often, it's because they were rejected.</p>

191

<p>That half of them are rejected confirms Cellardweller's analysis. One expects that for about half of the scorers at the 99th or any other percentile, their true level is (a little or a lot) higher than measured, and for the other half their true level is lower. Which side a given applicant is on can be guessed from further measurements such as SAT-II and APs. The ones whose data shows they are clearly above the line, tend to get in. The ones whose data is ambiguous or negative tend to be rejected. The split of about 50-50 at this range shows that Cellardweller calibrated the selectivity quite realistically.</p>

192

<p>The fact that 50% of top scorers are enrolled in the top 15 private Us does not mean the other 50% are rejected. Many never apply to these schools or choose to go elsewhere even if accepted. A very substantial number of high scoring candidates choose an alternative to a top private university for a variety of reasons: full ride at lower tier school, better program at other school, LAC environment, geography... Top publics also attract a substantial number of high scorers. It follows that the percentage of high scoring students that are ultimately rejected from all their top university choices will be quite low. Anecdotal evidence suggests it happens, but not often. Jian Li still got into Yale.</p>

193

<p>Sorry, cellardweller, I didn't mean to imply that all of the other 50% got rejected.</p>

<p>Of course, some of them chose to attend LACs, top publics, schools that excel in particular specialty fields, or lower tier schools that offered them merit scholarships. </p>

<p>However, based on what I see around me, quite a few of the top scorers did get rejected. My daughter has several friends with scores in the range you're talking about who will be attending our state university because they were rejected from all of the top private universities that they applied to. They will have huge merit scholarships at State U. But that's only a consolation prize. </p>

<p>I am absolutely astounded at some of the kids I see who have no alternative to State U. National Merit Finalists in some instances. Kids with excellent GPAs and interesting ECs. I would have expected them to at least get into colleges at the Duke/Cornell/Northwestern level of selectivity or the level just below that (Tufts/Georgetown/Chicago), but in some cases it didn't happen. This was a tough year for admissions.</p>

194

<p>jazzymom, thanks, I would not have predicted it either! I have worked very hard in high school and thought I would get accepted. As for Georgetown, I do not see me ever applying there again but maybe my perspective will change in four years. We had a kid from our school last year who had a perfect ACT and SAT, Valedictorian, involved in everything and he had almost the same results. He ended up going to Yale off the waitlist. I have had admissions officers tell me that being from an urban area which is highly educated and wealthy is definitely a disadvantage. Maybe I should have moved to Arkansas afer my junior year :) I hope the OP's son has better results is that is what HE wants.</p>

195

<p>We probably terrifed the poor OP....</p>

196

<p>Cellardweller, unless you have access to some secret stash of information, you are selling a bill of goods. Knowing how many people got 2200+ on a single SAT test date gives you very little information about how many of those people are "absorbed" by the top 15 schools. The top 15 schools generally report only (a) CR and M scores, separately (b) on the basis of best component score out of multiple test dates, (c) with SAT-equivalent ACT scores mixed in if they are better or if no SAT scores are available. Every one of those differences expands the pool of kids you are talking about by somewhere between 10% and 100%.</p>

<p>I'm glad that you get comfort from your analysis, but it doesn't add up. It doesn't add up in theory, it doesn't add up based on what the colleges report (Brown, which gives more information than anyone, and which is about smack in the middle of the top 15, only accepts 25% of applicants scoring 750+ on either SAT component), and it doesn't add up on the ground where one can see that nowhere near half of that group goes to "top 15" schools.</p>

197

<p>Marian--an application strategy that includes only top 10-20 and a state U. is, in my opinion, very smart for this population IF they can keep emotion out of it (and who can). If you are searching for VALUE and you live in a state with an excellent state U. then why pay up for a second or third tier private school. If you can honestly say "I'd be happy to attend state U, and a school would have to be pretty spectacular to be worth the extra money" then you can see how kids might have a list that includes a few ivies and the state university. </p>

<p>I suspect most kids who end up in that boat didn't do it on purpose though, but some may have.</p>

198

<p>JHS:</p>

<p>A 2200+ single sitting is largely equivalent to a 1500 combined M + CR or a 2250 combined. I am still speaking of the 99th percentile cutoff. However you cut it or slice you are only talking about a group of less than 20,000 kids annually. A 10% difference is insignificant. Under no circumstance is it 100% or even 20% off. ACT numbers are largely insignificant at those schools. </p>

<p>An yes, you can estimate how many of these kids are enrolled from the college 25/75 reports. When the 25% percentile is at or near the 99th percentile you that around 75% have better scores. So, yes it easily follows that well over half of HYPSM admits are in the 99th percentile of SAT scorers. For other top schools, if the combined 25% percentiles for above the 99th percentile then at least 25% of students are enrolled. That is pretty basic and there is no possible overlap as we are talking strictly about enrolled students not admits which can overlap. Even in cases where you only have the scorers over 700, you can estimate quite closely how many are between 700 and 750 and how many are between 750 and 800. The Brown data is a good proxy. Collegeboard distributions are also good proxies yielding no measurable differences. </p>

<p>The other point you seemed to miss is that whatever cutoff you use, the high scorers enroll at least at 2.5 times their proportion of in the applicant pool. </p>

<p>In the end it does not matter what exact cutoff you use. Top colleges do enroll a very large number of top scorers. I estimated about half at the 99th percent cutoff and I believe that number to be conservative.</p>

199

<p>Mombot:</p>

<p>I agree with you that such a plan can and probably will work if you take the emotions out of it. If you are willing to forgo some preferences such as urban vs rural or cold v warm climate or too narrow a limit on size, you should get some acceptances at top colleges assuming you fit in the highly qualified pool. Most applicants do not and shortchange their list by removing excellent choices that they either do not like based on a single visit or comments from friends or relatives or simply for the purpose of cutting back on their list. Some students may not apply to Penn because it is too urban or Cornell because it is too remote or Dartmouth because of its D-plan or Duke because it is in the South etc.. It is quite clear though that any of these schools would provide an exceptional education and if the choice is in the end between one of them and a State U., many would forgo their a priori preferences. One also never knows which will provide the better financial package even if they are supposedly all need blind. Dartmouth with a great package with mostly grants may beat Yale with a package of loans. </p>

<p>You may not be able to get your first or second choice, but if you don't apply to enough schools you may not have any good choices left at all.</p>

200

<p>Cellardweller: </p>

<p>The points YOU are missing are:</p>

<ol>
<li><p>Yes, half of HYPSM enrollees -- about 7,000 kids -- have scores as measured by HYPSM that would be in the 99th percentile of SAT scores if they had been received on a single test taken on a single date. You have no idea if they would have been in the 99th percentile of the best single whole test taken by a cohort from multiple testing dates, and you have no idea where the 99th percentile of mix-and-match component scores is. I would bet anything that it's much, much higher than the 25% level at any of these schools.</p></li>
<li><p>ACT: What makes you think it is irrelevant? Only about 10% of applicants only submit an ACT, but substantially more submit both, and the college data is predicated on converting the ACT to an SAT score and reporting the best one only.</p></li>
<li><p>
[quote]
For other top schools, if the combined 25% percentiles for above the 99th percentile then at least 25% of students are enrolled.

[/quote]

I'm not sure what you meant to say. I think you meant that if the combined 75th percentiles are above the 99th percentile of SAT scores, then at least 25% of the enrolled students have combined SAT scores in the 99th percentile. Except that isn't true, since the top 25% at a school in each category is not necessarily comprised of the same people, and even for the subgroup that is included in both top 25%s, there is no guarantee that they would have been so included solely based on their last SAT test.</p></li>
<li><p>I'm not missing at all the fact that high scorers are admitted disproportionately to their presence in the applicant pool. That's clear. The only thing we're fighting about is whether the probability of admission is about 50% or a lot less than that, and how many schools you have to look at before you have absorbed the population of high scorers who have applied to those schools.</p></li>
</ol>