I know across CC everyone says once your SAT/ACT score is 1540/34 ish you are past the ‘threshold’ for ‘elite’ schools. While I agree with this to some extent, I don’t know how entirely true this is.
The most ‘elite’ schools in the country, HYPSMC + U Chicago, all have the highest average SAT score. By a pretty solid margin. I know the following link is for 75th percentile, but a similar conclusion can be reached with this data: http://www.stateuniversity.com/rank/sat_75pctl_rank.html
HYPSMC + Chicago had a higher 75th percentile SAT score by around 20-40 points compared to other ‘holistic’ and ‘elite’ schools such as WashU, Duke, Penn etc.
My point is that HYPSMC + Chicago go out of their way to admit these top scorers (1580+) and their average/75th percentile SAT scores reflect this.
So the idea of a ‘threshold’ is not correct. Scoring a 1600 vs a 1560 does make a difference. HYPSMC + Chicago have their pick of 1600s and therefore they admit them. It won’t be the difference between an admit and reject, but it is an advantage.
Last point, if the idea of a threshold truly was correct, then all the top 10-20 schools would all have similar average SAT scores and similar 75th percentiles (all within 10-20 points of each other). A threshold would cause adcoms to weigh a 1540 and a 1600 the same, dragging the schools averages down.
Thoughts? I’m interested in hearing if my reasoning is flawed.
I’ve been wondering about the same thing, so I’m curious to hear what other people think. My private counselor told me about the threshold theory. He said 1550+ and 34+ are the numbers that will get you into the mix. I got a 1550 on my first SAT sitting so I probably won’t take it again. I still wonder though if I should keep trying. But I think it might look good to have that one solid score in a single sitting.
OK, your reasoning is flawed. 
No they don’t. The CDS gives the 25th/75th percentiles for SAT EBRW and for SAT M. One cannot simply add those two numbers together and be accurate. But not to worry, sites like prepscholar also make that common mistake.
IMO, 1550 is too high. A 1500, and even slightly lower, will keep you in the mix. But at the end of the day, all of these schools evaluate holistically. Each and every one of these schools routinely rejects 1600/36/4.0 applicants while accepting applicants with lower stats.
While a fun parlor game, trying to make sense of the admissions statistics for colleges with single-digit acceptance rates is just productive as analyzing past Powerball numbers to predict tonight’s numbers.
I wouldn’t know how much truth there is to this, but I would hope that adcoms are smarter than to think student A is higher-quality having gotten four more questions correct on the entire SAT than student B. My sense is that this difference would matter more at some slightly-lower ranked schools than at HYPSM, e.g. Vandy.
At the moment, the 99th percentile starts at 1480.
My argument isn’t that the adcoms think that a student with a 1600 is smarter than a student with a 1550 or whatever. What I’m saying, is that elite schools will accept the 1600 to boost their average SAT score making them seem that much more difficult to get into and more ‘desirable’
I highly doubt that senior administrators at colleges with single digit admissions rates are standing in their war rooms and pondering, “Gee, how can we make our college seem more desirable.”
Also, going back to your original post, which I’ve already expressed as being flawed, the 20-40 point difference in SAT scores between HYPMS and the-however-one-defines-as-the-next-tier is statistically insignificant IMO.
For both points, YMMV, of course.
alright thanks for your input, I was simply interested in what others had to say about this topic
@skieurope
That’s because the tests are way too easy to make meaningful distinctions at the top end. For instance, the math portion of the SAT is very low level, and yet a single silly mistake could knock an 800 to a 770 or 780. Are we testing rote conscientiousness, or true ability? We need different tests, or else we consign ourselves into an ever more crowded madhouse in which the top 5% can make themselves look like the top 1%.
Precalculus is not “very low level” for the vast majority of those that take the tests.
Well beyond the scope of this thread, and not one that will be decided by any of us.
^ Low level in terms of reasoning ability required. Think of the task as being “easy” but complex. That the math test has a very cursory treatment of certain more advanced topics only serves to highlight the inequity of the testing regime generally. Relatively low ability, but privileged, kids can do very well as a result of coaching and quality of schools, and therefore those seeking to maintain the status quo will prefer this sort of testing regime.
By contrast, imagine a testing task that is “hard” but simple. Such a task is more sensitive to different levels of ability, and less sensitive to environmental privileges. It would have more meaning imo for college admissions, especially as colleges already have extensive data in the form of GPA, course rigor, school environment, AP scores, SAT 2 scores, etc. from which to infer academic preparation and motivation. Relatively high ability kids - regardless of privilege - will prefer this sort of regime, as will people seeking to upset the status quo.
(Of course, any test that involves any cognitive task is going to involve basic ability at some level, and that is why we see predictable score variation regardless of preparation effects. The current system is clearly not sensitive enough to measuring basic ability at the top end of college admissions, or else one wouldn’t see the clustering of scores at the extreme top end - e.g, the 75th percentile at a number of schools is right at - or within one silly error - of the highest numerical score possible. It’s difficult to see why making the tests harder would have any deleterious effects at colleges of lower selectivity, as what is always being compared are individuals, not scores per se. One way to see this is to look at the average SAT scores for, say, Yale pre-1995 and then after 1995: they jumped about 100 points, but of course there could not have been any change in the underlying average ability of the admit pool in a few years.)
you spent too much time discussing insignificant details - get your app together and apply dont worry about 50 or even 100 point difference in SAT - there are lots of kids with perfect score who dont get in and just as many with avg stats who get in…