Portion of Matriculating Students that Are Test Optional at Selective Colleges

Applying to College
1

I initially posted this in the switching to test required thread, but was flagged as off topic; so starting a new thread instead.

The federal database lists % submitting SAT and % submitting ACT, but does not list % submitting scores in total. Approximating total submitting as % SAT + % ACT double counts students that take both, leading to some colleges with >100% submitting. Roughly approximating as 0.88 * (% SAT and % ACT), I get the following. I am only including academic-focused selective colleges for which admit rate is <20% . I also excluded test blind colleges in Fall 2023 including Caltech, UCs, and Pitzer.

Note that it is difficult to draw conclusions from this percentage in isolation. Different colleges on the list had wildly different % of applicants submitting scores, ranging from under 40% of applicants to over 70% of applicants. This partially relates to different rates of test score participation in different states ranging from 24% of HS students taking SAT in CA to near 100% in states where taking SAT/ACT is mandatory for graduation. Different colleges also use different languages on their website about whether scores are preferred or not. Different colleges also have different test optional history. Applicants submitting and not submitting scores often also different average qualifications in other portions of application. For example, students who choose not to submit scores because their scores are subpar for the college often also do subpar in other non-score aspects of application.

Average = 52% Submitting Scores
Median = 48% Submitting Scores

Fall 2023: Estimated Highest % Of Matriculating Students With Scores (IPEDS)

  1. MIT – 100%
  2. GeorgiaTech – 99%
  3. Webb – 96%
  4. Georgetown – 95%
  5. Ozarks – 90%
  6. Yale – 72%
  7. Duke – 71%
  8. Northwestern – 70%
  9. Rice – 69%
  10. Princeton – 68%
  11. Chicago – 67%
  12. Brown – 67%
  13. Penn – 67%
  14. Harvard – 65%
  15. Dartmouth – 64%
  16. Michigan – 62%
  17. Stanford – 61%
  18. CMU – 60%
  19. Harvey Mudd --59%
  20. U Virginia – 58%
  21. UNC: CH – 55%
  22. Emory – 54%
  23. Columbia – 54%
  24. Williams – 53%
  25. Bowdoin – 52%

Fall 2023: Estimated Highest % Of Matriculating Students Without Scores (IPEDS)

  1. Bates – 81%
  2. Northeastern – 69%
  3. Babson – 68%
  4. Colorado College – 67%
  5. Boston U – 67%
  6. Vassar – 66%
  7. Colgate – 66%
  8. NYU – 66%
  9. Middlebury – 63%
  10. Wesleyan – 63%
  11. Boston College – 61%
  12. Pomona – 61%
  13. CMC – 60%
  14. Tulane – 60%
  15. USC – 60%
  16. Smith – 58%
  17. Cornell – 57%
  18. Barnard – 56%
  19. Denison – 55%
  20. Vanderbilt – 55%
  21. Cooper Union – 54%
  22. Washington and Lee – 54%
  23. Miami – 53%
  24. Colby – 52%
  25. Swarthmore – 52%
5 Likes
2

Based on cases from the past, this may not represent the situation. For example, prior to becoming test optional some colleges would report scores such that the total for the SAT plus ACT percentages would consistently conform to 100%. By inference, then, when students had submitted scores for both exams such colleges were choosing to report information for just one or the other based on an internal policy. However, in the test optional era any such policy would be difficult or impossible to detect by inference alone.

3

Should “not” appear between “does” and “list”?

4

Good catch. I’ve edited the post. I initially typed quickly, without double checking

1 Like
5

I’m not sure that it’s difficult to draw conclusions.

Those schools - the Vassars, Bates of the world - would either have fewer applicants (thus a higher acceptance rate) or lower scores - and they’ve connect the world into thinking - they can attract just as great students - who don’t test as well as the rest of their classes.

It doesn’t mean they don’t test well. Just not as well.

At Pomona, the median SAT is 760 for English and 780 for math. What if someone had a 720 and 730. Still outstanding, but not Pomona outstanding.

Same with the 34 ACT - what if someone had a 31 or 32. Great - but it might keep you out.

We don’t know, though, whether that student had a 25 or 32.

Wesleyan shows a 27 math ACT at the 25th %. The other year Chicago showed a 20 composite. I found out later it’s because even TO students have to submit before enrollment.

Now - a student at UC can’t submit a test so a 34-36 does nothing for them and the person who might get a 25 is equal in that sense.

But schools are allowed to pick and choose the criteria and how they weight. Some weight ECs, wealth, ability to pay, essays and more things than others do.

They are businesses - and it makes sense.

Thanks for putting this together.

I’ll just say - in today’s day and age vs. covid times, I believe this is an indictment against the colleges themselves - they are “fudging” #s - as this would be a very easy metric to compare vs. other kids in the college. Instead, the schools turn a blind eye so they can enhance their stats.

That’s my belief anyway - I know others disagree.

1 Like
6

Among the 50 colleges listed above, 5 had % SAT + % SAT = 100% in 2019. The highest ranked of those 5 is Webb. If I take Webb at face value, % submitting increases to 96%, which is more reasonable considering that Webb was test required in 2023. I’ll update those 5 colleges in the table.

Scaling by a function of proportion over 100% in 2019 rather than default of 0.88 leads to abnormal results due to large variation in this figure. The table should be considered a rough estimate, rather than a precise figure.

7

I can say for sure that’s the case at Tulane. When S was applying and going to information sessions they said outright "we are test optional and would give guidance that unless your score is in the top 50% of our reported numbers, you don’t submit. If everyone took that guidance to heart, the reported scores are going to creep up.

Do they still give that guidance I don’t know.

3 Likes
8

Cite?

3 Likes
9

@Data10, for the schools that used to require the test but are now test optional with non-reporting rates in the 30-50% range, about how much have average reported scores gone up from, say, 2019? (I realize it will vary depending on the school, but I’m trying to get a rough feel.) Thanks!

10

How many colleges that met these criteria fell between Bowdoin and Swarthmore?

11

An aside:

I wonder how much test-optional policies result in schools admitting and matriculating different kids vs. the same kids but just eliminating so much of the test prep churn (i.e. all the prepping and retesting that some kids used to do.)

2 Likes
12

Bowdoin appears to have made a transition to a required submission policy (for matriculation) for the fall of 2016. By comparing these figures to those of the prior year, it would be feasible to draw inferences about the impact of reporting practices on apparent score distributions:

Fall 2016

SAT Middle Ranges, 54% submitting

CR: 650–750
Math: 640–760
Combined: 1290–1510

ACT Middle Range, 44% submitting

30–34

Fall 2015

SAT Middle Ranges, 42% submitting

CR: 680–765
Math: 685–770
Combined: 1365–1535

ACT Middle Range, 36% submitting

31–34

1 Like
13

So, let me see if I understand this. Bowdoin’s Fall 2016 numbers actually include tests for all students (or nearly all students) but only 54% initially submitted. Or am I misunderstanding?

1 Like
14

That is correct. IIRC Wesleyan does the same…is that correct @circuitrider?

And to be clear…Bowdoin doesn’t ‘require’ a test score for matriculating students, but they do want them to provide one if they have one. They do admit some students who didn’t test, more so in the covid years, but still now too.

1 Like
15

Yes, that is their official policy.

2 Likes
16

I think Bowdoin has stopped including all scores (even those not used for admissions) in their CDS. They’re one of the few schools that’s published the 2024-2025 CDS. For the latest CDS, they report 31% submitting SAT scores and 17% submitting ACT scores.

1 Like
17

Thanks for the update!

18

There is a correlation between different parts of the application. Students who are stronger/weaker in one area are more likely than average to be stronger/weaker in other areas of application. An example publication is below. There are many others. Test optional applicants averaged 120 point lower SAT than test submitters (1124 vs 1243). Their GPAs were also lower – 3.48 GPA for test optional vs 3.60 GPA for test submitters. Many of the 13 colleges composing this average were not especially selective. The effect tends to be magnified as selectivity increases.

This effect contributes to why test optional applicants and test submitter applicants have different admit rates. This makes it difficult to conclude how much of the different admit rate was due to submitting scores vs due to differences in other parts of the application.

image
image779×335 41.4 KB

19

Yes. the table at Admissions | Bowdoin College indicates that in Fall 2023 Bowdoin ended their policy of listing scores for all students and instead started reported scores for only submitters.

image
image666×357 19.7 KB

image

1 Like
20

With all due respect, what am I looking at here? I see a screenshot.