<p>I am trying to understand how to interpret the data provided in the Common Data Set for various colleges and universities.</p>
<p>Mostly, I am trying to understand where my child's stats put him in relation to the college for admission purposes (ie -- reach, match, safety.)</p>
<p>What is confusing to me, is that when I look at the actual stats of enrolled freshmen, these stats are generally lower than what is published elsewhere (the average SAT score for example). Is this because, for many colleges, the average SAT of admitted students is higher than the average of those who actually attend?</p>
<p>Any clarification would be appreciated, including whether looking at CDS is even a valid strategy.</p>
<p>IIRC CDS data is for admitted students (as opposed to enrolled) which is what you want to pay attention to as far as the admissions part of the decision. Pretty much if your child is below the 25th percentile (test score wise) consider it a reach school (the usual exceptions apply for all Ivies et. al. which are reaches for everyone). If you child is above the 75th percentile, consider it a safety (same exceptions). Anywhere in between, consider it a match. Look at the admit rate as a reasonable estimate of likelyhood of admission if you hit the 50th percentile.</p>
<p>I think your post mentioned average which is different than median. Pay no attention to that number as depending upon the school, special admits can make that number off mark as an indicator of typical admit.</p>
<p>And most of all CDS is a common set of metrics across a lot of schools.</p>
<p>And if the enrolled score numbers are lower than the admitted numbers, it says the school is used as a safety by a significant number (who don’t enroll). It may also say that their FA merit money isn’t enough to keep those top 25% of admitted students.</p>
<p>Of course the flip side (the school that’s enrolled student have higher scores than admitted) is a school that buys students. It is probably trying to raise its profile to attract higher profile “match” applicants who will hopefully pay.</p>
<p>All of the Common Data Set GPA, rank, and test score data refers to enrolled students, not admitted students. If you are finding discrepancies, it probably relates to data from different years, not enrolled vs. admitted.</p>
<p>Not many colleges publish data on their admitted classes. Yes, of course, the stats on the enrolled class will tend to be lower than those of the admitted class. Students with high numbers tend to have lots of choices, so they are less likely to attend any particular institution that admits them. Students accepted to their dream-school-ultra-reach tend to enroll.</p>
<p>data in section c of the cds tells you the stats for ENROLLED freshman.</p>
<p>yes, if a school gives you stats for those ADMITTED often those stats will be higher than those of the students enrolled – if you think about it, it makes sense – those at the top of the acceptance pool are the ones most likely to have other attractive options.</p>
<p>the biggest failing of the common data set – they don’t tell you anything about the stats of those rejected. just because your stats compare favorably to the stats of those freshmen in the top 75 percentile, doesn’t tell you how many of those with those stats were rejected – at top schools there can be a LOT of students with those same stats who were rejected. relying on having stats in the top 75th percentile can give you a false sense of security. unfortunately most schools don’t make that information readily available. </p>
<p>for example – see this info from upenn – it shows you what percentage of people with a given stat were accepted. [Penn</a> Admissions: Incoming Class Profile](<a href=“http://www.admissions.upenn.edu/profile/]Penn”>http://www.admissions.upenn.edu/profile/) i think that gives you a much better sense of your chances than the common data set info. according to collegeboard, the mid 50% for sat cr for upenn was 650-740. so if you have over 740, you’re in the 75 percentile which sounds great. but if you look at the page from upenn on profile you see that only 30% of those with a 750 or above on the sat cr were admitted – a lot more sobering than the 75 percentile stat.</p>
<p>unfortunately, this type of information just isn’t universally provided – i think only this type of info really lets an applicant assess what their chances are – especially at the more competitive schools.</p>
<p>No. The only college I know that does it systematically is Amherst (and they only do it for test scores). For selective colleges, though, I tend to think that the Amherst ratios are going to be not that different from any other school’s. The mean SATs for the admitted class are 10-15 points higher than those for the enrolled class. The top 25% of the enrolled class represents about 40% of the accepted students, and the bottom 25% represents about 20% of the admitted class. </p>
<p>Interestingly, though, except for applicants with the highest and lowest test scores, applicants in each score range are admitted in roughly the same proportions that they apply. E.g., Applicants with Math SAT Is in the low 700s, and the low 600s all seem to have about a 14-15% acceptance rate. (There’s a little bit more of a curve for CR and W, but not much.) Applicants with scores in the high 700s have a 25% acceptance rate, and below 600 the acceptance rate goes from 10% to 5% pretty quickly. The point is that test scores seem to matter, but not as much as one might think.</p>
<p>The CDS is extremely useful and is the best tool to compare schools head to head.
Not only can you see acceptance rates and SAT scores and debt load and average grants, etc., but you can see them over time at most schools’ web sites.</p>
<p>Admitted student data is used by certain schools to create the impression that its SAT scores are higher than they really are. Colgate does this and for years it didn’t publish its CDS, so you couldn’t compare its numbers with other schools. There are other schools of this ilk - competitive schools - and I’ve never understood why they do it, as their “real” stats are quite impressive.</p>
<p>However, remember to discount the SAT percentiles somewhat if you are looking at a SAT optional school, since those who submit are likely higher scorers.</p>
<p>Overall, while I think the CDS is the best tool for comparing schools, individual schools do publish class profiles that include other interesting data. You just can’t generally compare them among schools.</p>
You might try the school websites, particularly under “in the news” (or whatever) or the link to the student newspaper. In my experience, schools do issue press releases regarding the stats of their accepted classes, both for the ED round and regular decision.</p>
<p>Many schools post the admitted student stats for the most recent year on their admissions page somewhere. Others will have the data if you ask. Sometimes they provide the info on a handout if you visit and meet with admissions.</p>
<p>The Common Data Sets are standardized information that colleges track and that most of them publish. I think it is what they submit to US News & World Report; I don’t know what role USNWR played in promoting agreement as to what data should be tracked in the Common Data Sets. Anyway, most colleges have at least the past several years of their CDSs somewhere on their websites, although some refuse to do that. They tend to get published in mid-fall for the current entering class.</p>
<p>Anyway, the CDSs contain a lot of information about total enrollment, graduation rates, admissions (broken down by gender and no other category) (including ED or EA numbers), admissions policies, entering class profile (test scores, class rank), class sizes, graduation requirements and special programs, some fairly random information about the student body as a whole, costs, financial aid. The definitions are supposed to be consistent across institutions to promote comparability, and for the most part they probably are.</p>
<p>I strongly disagree with this advice and believe it will result in an overly optimistic college list for many applicants.</p>
<p>You cannot evaluate SAT scores in a vacuum. A low income African American applicant will be evaluated on a different scale than a white applicant from a wealthy suburb who will be evaluated differently than a recruited ice hockey star. If you don’t understand the concept of variable standards in college admissions, read “The Gatekeepers”, especially the paragraph where the admissions officer picks up the folder of a wealthy white applicant and mutters, “This better be good…”</p>
<p>**For the typical white or Asian American College Confidential student applying to elite colleges from an affluent suburban high school or private school, I recommend having 75th percentile SAT scores to consider a college a “match” and 50th percentile scores to consider it a plausible reach. **Unless there is some outstanding hook, affluent white applicants are simply not likely to be accepted to elite colleges with “stats” below the 50% mark. Students being accepted to elite colleges with below 50th percentile stats are bringing something specifically desireable to the party – diversity, impact athlete, child of a big donor, incredible extra-curricular interest, spawn of a famous person, etc. Yes, I know…there are exceptions, but if you are looking for a rule of thumb, there ya go.</p>
<p>[NOTE: with the current economic squeeze, the ability to write a $50,000 check for each of the next four years may tilt the playing field a bit at some points on the admissions exclusivity ladder, but that’s something that would have to be evaluated on a school by school basis.]</p>
<p>Obviously test scores are just part of the puzzle. Determining reaches, matches, and safeties really requires a qualitative assessment of the entire application. Is there something about the application that will stand out?</p>
<p>ID, may I ask you to help me understand something?</p>
<p>I have a lopsided student. In schools that accept around 40% of applicants and are good, but not elite schools, how would you view an applicant who is above the 75th percentile in writing and critical reading and around the 25th in math? Or a student who’s between the 50th and 75th percentiles in critical and writing, but below in math? I just can’t grasp this at all.</p>
<p>You are acting as if URMs, legacies, and recruited athletes make up a large percentage of a school’s population. This is not the case at most schools. </p>
<p>By and large the middle 50% of students are represented by the middle 50% of SAT/ACT test scores. Yes, there is some variation in GPA, ECs, Geography, etc., but as a rule, for those non-elite schools (I did mention the usual exceptions) which make up the vast majority of all schools, your typical student (middle 50%) has an acceptance chance that should be close to the average percent accepted. And for most schools that is a darn good chance of admission. Match does not equal guaranteed admisssion - that is what a safety is. A student should always have more than one match application.</p>
<p>
</p>
<p>You treat elite schools like they are full of students not from “affluent high schools or private schools”, when in fact the greater majority of the students at these schools are from these schools. And their statistics are exactly what the schools report. And the percentage admitted is exactly the percentage admitted.</p>
<p>Yes elite schools have a lower admissions rate, but once you get above a certain test score threshold at these schools, you chances of admissions do NOT go up. These schools graduate a very large percentage of entering students - 50% having SAT/ACT scores below the median - showing that being at the 50% is more than adequate to generate successful alumni. At this point it is the other stuff - ECs, essays, etc. that make the difference in admissions chances. This is what tells them who is likely to achieve distinction at their school and in life (and hopefully donate frequently and generously).</p>
<p>These schools by and large are outliers though. Of course the worry warts on CC who take the SAT 5 more times after getting a 2200 the first time are desperately looking for that reassurance that my superior numbers make me superior find solace in this 75th percentile as a match thinking.</p>
<p>My comments were about the majority of colleges around the country.</p>
<p>I think it really depends on the college, and probably varies quite a bit. Still, I think it’s possible to hazard a few educated guesses:</p>
<ol>
<li><p>For small colleges, that bottom 25% in any category may have very few “normal” students. Amherst, for instance, is a very small college with major historical commitments to sports and to ethnic/cultural diversity. Of course, lots of the athletes or URMs there are NOT in the bottom 25% in any category, but some are, and it doesn’t take many to fill it up. On the other hand, at, say, Bryn Mawr, things may look a little different. But then if you are talking about Cal Berkeley, all the recruited athletes in the university barely affect its statistics – they are a drop in the ocean.</p></li>
<li><p>Apart from the foregoing, I would think that a LOT of the kids in the bottom 25% on math scores are in the top 50% or top 25% on the other test scores. Relatively few kids are going to be accepted anywhere if they are at the bottom in all categories, and those few are going to be completely obvious superstars at SOMETHING. That’s where the bottom 25% of any category comes from – people who are so good at other things that it’s OK to be merely OK at this thing.</p></li>
<li><p>In general, colleges are going to care less about having everything be high than high schools. Ultimately, college students are going to concentrate their studies where they are strongest, so their areas of relative weakness will not be so relevant. Some colleges seem to prefer “lopsided” students; others still like BWRKs; and most seem to want some of each. In this regard, I think a kid with low math SATs but Bs or better in hard math classes looks a lot more attractive to most colleges than a kid with low math SATs and bad math grades in non-honors classes.</p></li>
</ol>
<p>OP, we just threw out the bottom 1/4 and re-centered. We then “figured” a kid with her ethnicity from an unknown high school and an over-represented state would need to be in the top 25% of what remained. For merit aid schools we even cut a thinner slice. Worked like a charm. ;)</p>