Thanks merc81!
Yes, at the top schools we always want to simplify it to scores and GPAâs because that is the only concrete thing we can compare, but it really isnât that useful. Having said that the higher your SAT/GPA the better your chances.
Yes, the usual mistake on these forums is to exaggerate the effect of test scores, GPAs, and certain yes/no attributes seen as âhooksâ (e.g. legacy, URM), since they do not involve opaque subjective grading that those outside the admission office of a highly selective college have no visibility into. But because highly selective colleges have so many applicants at the top of the ranges, the other opaquely subjectively graded attributes like essays, recommendations, extracurriculars, interview, etc. become much more important than at less selective colleges where test scores and GPAs differentiate the applicants more. Note that this also makes the admission process at highly selective colleges appear like a lottery from the outside, even though is not.
@ucbalumnus, agree but we canât discount the ever increasing influence of the search for diversity (URM and First-Gen) at the most selective LACâs as it has made the funnel for non-hooked applicants the narrowest that it has ever been.
There is so much broken and wrong with this post that I almost donât know where to start, however, I will start with one particular quote from the original post:
âI think prospective freshmen should attend the best school they can possibly afford.â
The term âbest schoolâ is a fundamentally broken term. There is no such thing as a âbest schoolâ.
For any one perspective student, in general the academically most challenging school that they can get into is definitely NOT the best match for that student. There are long lists of reasons why this is the case.
As examples: The academically most demanding school that a student can get into might have a lousy program in whatever that student wants to major in, or might not even offer the major that the student wants. Going to a very challenging university might make it very difficult for a student to get high grades, which in turn might limit that studentâs ability to get into graduate school or medical school. Some students are brilliantly intelligent, but donât deal well with stress. Some students are more comfortable in a small school, but the academically most challenging school that they can get into is very large. Some students might want to go to a school near home, but the academically most challenging school that they can get into is on the opposite side of the country or even in a different country.
We have seen a few cases on CC where a student has for some reason gotten accepted into a school where their GPA and SAT scores puts them near the bottom for that school (perhaps in the bottom 10%). Someone has to be the student with the lowest incoming GPA to enter the freshman class at a school. However, do you want to be that student? I know that I do not want to be that student (and I wasnât).
Our education system and our students would be much better off if the term âbest schoolâ was retired permanently and never used again. The term âbest match for a particular studentâ is far more important.
@ucbalumnus, in addition at my DDâs very competitive high achieving HS, the majority of the students that get accepted to Ivies and Ivy equivalents are those that are either recruited athletes with much lower academic profiles and/or legacies that do profile, so it feels much less opaque than you outlined.
@DadTwoGirls, I donât think the CC crowd is representative of most students, and agree that over stretching isnât a good place to be. That said, many of the students on CC that are seeking admission to the top schools are already in the top 25% so its a different audience for sure.
ucbalumnus, I do not think HS GPA is the best indication of admission difficulty. I think admissions officers know this. Yes, they consider HS GPA but they must know it is subject to interpretation according to such things as academic standards at the high school and the difficulty of the HS courses taken. Average HS GPA is generally not available anyway and it is not standardized like the SAT.\
Chemblodad, I had to use 2002 data for Bates because that is the last time they reported SATs to the US Department of Education. The official data for a college is that which is reported to the US Department of Ed, not the Common Data Set. The definitions and computational methods are clearly defined by the US Department of Education. If Bates is SAT optional, then their SATs reported in the Common Data Set would not meet the USDE standards. Furthermore, I am not sure that attending a âreachâ school would necessarily be more stressful or more difficult to get a high GPA. Students at âreachâ schools are generally more successful as indicated by their higher freshman retention and graduation rates.
DadTwoGirls, students should not fear being a âsmall fish in a big pondâ. Donât be intimidated. Go for it. The âbig pondâ means more room for growth. If you are admitted, you will probably do fine. If you are admitted, you almost certainly have all the skills you need to succeed. And, you will almost certainly be surrounded by more intellectual peers, better role models, better culture, greater resources, and professors who can better guide and support your growth.
In other words, you are using SAT scores to compare simply because they are convenient to compare, not because they are more important than high school record in admissions or predicting student performance in college.
That the more selective schools have higher graduation rates (because they have stronger students to begin with) does not necessarily mean that any given student will have a higher chance of graduating from a more selective school than a less selective school.
This kind of analysis may help identify schools where a studentâs chances are slim because their scores are too low. But I donât think itâs particularly helpful for those trying to project odds at the most selective schools. Those schools could have classes with averages of 1550/35 if they wanted, but they choose not to for all sorts of reasons that are difficult to evaluate on a student-by-student basis.
Maybe they show which schools prioritize test scores but that doesnât tell you whether a particular student with those test scores will be accepted.
ucbalumnus, yes, convenience is part of the reason for using SAT scores but I also like the fact that they are standardized, objective, and measure both math and verbal skills/achievement. My motive is simply to help prospective students make the best choices and to reach their full potential. The SAT percentiles were intended as a rough guideline for finding reasonable reaches and avoiding applications where there is almost no chance of acceptance. I think we both share the motivation to help prospective college students. Can you think of a way to improve on the process of finding reaches? The method I proposed is much better than guessing about which schools are reasonable reaches. It gives parents and students a tool they can use as a rough guideline. I know that other admissions factors are important. I would not be surprised if SATs are correlated with the admissions standards applied to those other factors such as HS GPA and essays and extracurricular activities (with possible exception of athletics).
With regard to the higher retention and graduation rates at more selective schools, yes, they reflect probabilities and do not guarantee success for any particular student. I would not be surprised if SATs were less highly correlated with success of individual students than with the overall graduation rates at colleges. Individual student outcomes are more difficult to predict from SATs than the overall graduation rate because there are so many factors that can cause a student to drop out such as finances, substance abuse problems, mental health issues, altered life goals, and so on. SATs help predict overall grad rates but not necessarily individual student outcomes. SATs are excellent at predicting the proportion of students who will graduate but not as good at predicting the outcomes for individual students. But, I stand by my point: students in the bottom 5% of a collegeâs SAT range still have a good chance of graduating because most students who drop out do so for reasons other than motivation and ability.
First of all, the data on the table is obviously outdated. Add 30-60 points for the new SAT score scale. Second, test score is just one of the factor and may be not even the most important factor. What is more crucial is the admission rate. Or more precisely, the admission rate at a particular test score. With a 25th percentile at two different schools, the admission rate would vary a lot. Not to mention, the variation is even larger for public school with different pools of applicants from in state and OOS. Why would one go through a list of all the schools after all? Student should rank their reach schools by location, majors offered, cost, etc, etc rather than test score. It is a crap shot for the top schools regardless your test scores anyway. IMO, this table is pretty useless.
@pantha33m, the most selective schools (top 20 universities and LACâs) do have median 34/35 ACT scores for non-hooked acceptances. In addition, just like hooked applicants getting a lift, the same can be true for those that meet the holistic needs of the schoolâs overall student population - too many people perceive holistic only as it pertains to the individual applicant and not to both despite admissions never distinguishing when they use that term.
Case in point, at my DDâs HS, 8students that academically profiled the same applied, 7 were female and 1 was male; given that Brown is heavily overweighted with female applicants, the male got in - is that holistic, I think so as they are looking for a balanced student population (male/female, geography, in-state or OSS for highly ranked flagships, academic interests, and in many cases full pay - the dirty little secret at need blind schools as the Common Application gives plenty of information for a school to profile the probability of need). Was I happy about it no as my DDâs were part of the other group, but as many have said âschools choose you, you donât choose themâ and fit means many things - all good.
So, my advice is that if an applicant doesnât meet the hooked or holistic needs of a most selective schools, they should expect to have to be in the top-25% in order to assume they have a reasonable chance, and then chances go down from there as acceptances rates get cut in half, then another half, and so onâŠ
Excellent point. E.g., Vassar⊠Males have a higher admission rate than females by a not insignificant percent.
@CrewDad, itâs actually much greater than that at Vassar: 19 percent admission rate for women, 34 percent for men, a 15-point gap, which is a 80% greater chance.
Others are;
Davidson: 19 percent for women, 26 percent for men, a 7-point gap.
Bates: 23 percent for women, 28 percent for men, a 5-point gap.
Pomona: 10 percent for women, 15 percent for men, a 5-point gap.
Swarthmore: 15 percent for women, 20 percent for men, a 5-point gap.
Bowdoin: 13 percent for women, 17 percent for men, a 4-point gap.
Carleton: 21 percent for women, 25 percent for men, a 4-point gap.
Haverford: 23 percent for women, 26 percent for men, a 3-point gap.
Kenyon: 24 percent for women, 27 percent for men, a 3-point gap.
Middlebury: 16 percent for women, 19 percent for men, a 3-point gap.
Wesleyan: 23 percent for women, 26 percent for men, a 3-point gap.
Williams: 18 percent for women, 21 percent for men, a 3-point gap.
@Chembiodad I think we are saying the same thing - standardized test scores can help (unhooked) students sort out where they are likely to be rejected, but not much help in figuring out where they will get in.
Where did you get those gender-specific stats?
These LACs admit female and male applicants at roughly the same rates:
Amherst
Claremont McKenna
Hamilton
Washington & Lee
Colgate
Grinnell
Macalester
Oberlin
Colorado College
Great than what? I didnât state a percent. I said males have a higher admission rate than females by a ânot insignificant percentâ. Isnât a 15% gap a not insignificant percent? 
Vasssa needs to get a football and hockey team. That should even out the m/f ratio 
The percentiles in the table in post #1 were from 2015 IPEDS data which is the most recent data available to the public. The US Dept of Ed goes through a year of checking data before it is made available to the public.
Here is a table for converting new SAT scores to the old CR+math scores so you can use this table with the new 2016 SATs. You can use this table to convert the percentiles in post #1 to new SAT percentiles.
Iâll say it again: admission rates are not as useful as SAT percentiles in determining whether a school is a reasonable reach.
Converting New Evidenced Based Reading and Writing SAT score (column on left) to Old Critical Reading plus Math SAT.
400 400
410 410
420 410
430 420
440 430
450 430
460 440
470 450
480 450
490 460
500 470
510 470
520 480
530 490
540 490
550 500
560 510
570 510
580 520
590 530
600 540
610 540
620 550
630 560
640 560
650 570
660 580
670 580
680 590
690 600
700 600
710 610
720 620
730 630
740 640
750 660
760 670
770 680
780 690
790 700
800 710
810 720
820 730
830 740
840 750
850 760
860 780
870 790
880 800
890 810
900 820
910 830
920 840
930 850
940 860
950 870
960 880
970 890
980 900
990 910
1000 920
1010 930
1020 940
1030 950
1040 960
1050 970
1060 980
1070 990
1080 1000
1090 1010
1100 1020
1110 1030
1120 1040
1130 1060
1140 1070
1150 1080
1160 1090
1170 1100
1180 1110
1190 1120
1200 1130
1210 1140
1220 1150
1230 1160
1240 1170
1250 1180
1260 1190
1270 1200
1280 1210
1290 1220
1300 1230
1310 1250
1320 1260
1330 1270
1340 1280
1350 1290
1360 1300
1370 1310
1380 1320
1390 1330
1400 1340
1410 1350
1420 1370
1430 1380
1440 1390
1450 1400
1460 1410
1470 1420
1480 1430
1490 1450
1500 1460
1510 1470
1520 1490
1530 1500
1540 1510
1550 1530
1560 1540
1570 1560
1580 1570
1590 1580
1600 1600
Again. It is nonsense to look at scores from public universities with in state and oos students combined. In addition, admission rate is for sure very important. Just look at your table and compare Stanford with Northwestern. It just does not make sense with score alone.
Also, for schools that have higher number of athlete recruits and legacy admissions, the 25th percentile and lower scores are skewed. How can a non hook applicant makes use of these numbers?