Standardized Test Opportunity?

Applying to College
21

So the CDS unfortunately does not contain good data on test submission for applicants as opposed to enrolled students.

But I think it is a good bet that at least, say, 60% of the applicants to a college like Yale submitted tests (with enrolled students ending up higher).

So for the Class of 2028, Yale had 57517 applications, which implies to me something like at least 34000 applicants submitted a test. Possibly more.

OK, so back in the 2019-20 CDS, Yale was still test required, and it got . . . 36,844 applicants. Given how rough all this is, I would agree those are more or less the same number within the margin of error. Meaning it does indeed look like at Yale at least, going test optional did not in fact significantly cut the number of applicants who submitted test scores.

And I think there are obviously going to be cases of people who would have applied to Yale either way, and so some of the recent test optional applicants very likely would have applied to Yale with test scores if they had to.

So yes, all this helps explain why MIT saw only mildly fewer applications when it went back to test required, and it didn’t go all the way back down to the levels of the earlier test required period. Lots of applicants were submitting tests to MIT anyway, others were not deterred when they had to submit, and so it was only a modest minority fraction that actually did not apply due to the change back in policies.

1 Like
22

I note before the demographic cliff there was the demographic plateau, where what had once been a rapidly increasing population of college-bound kids in the US leveled off. This actually happened circa 2010.

And overall, lots of colleges did in fact see a leveling off, or an actual decline, in applications.

But some colleges continued to see applications go up, in some cases by a lot.

There are at least a few different contributing factors. There is an ongoing trend of certain applicants applying to more colleges. There is also an ongoing trend of certain applicants being more likely to apply to colleges outside of their state/region. And finally there is an ongoing trend of an increasingly large population of reasonably well-qualified International applicants.

All this has meant that at most of the more “national”/“international” colleges, applications continued to go up even as the domestic US population plateaued. Meanwhile, some of the more “local” colleges, at least in many states, experienced declines, sometimes really rapid declines.

OK, so now the domestic US population might start heading down, and this will likely be not-so-good news for the colleges already struggling just with the plateau.

But personally, I would have modest expectations for what will happen at the colleges which still saw rapidly increasing applications despite the plateau. Because I think all those offsetting factors still have plenty of room to run.

1 Like
23

Not sure what you consider a minor fraction but the applications to MIT peaked in 2022 with 33,767.
The year after it went down to 26,914 which in my view is rather a large minority. It’s gone up to 28,232 in the last round. However, before test optional it was around 21k so even the 2023 numbers are quite a bit higher. I would link the site for these numbers but it may not be allowed though it’s easy to get with a search.

24

Yes, which is why I am interested. I don’t expect it will affect the T20 much if at all but I wouldn’t be surprised if their numbers do dip this time because a plateau is different than a decline. You might increase application numbers during a plateau but those same numbers could end up going negative if the decline is greater than the increase which is possible. However, if certain schools start to close, that will also likely increase applications for the remaining schools. By the way, the plateau I think ended around the 2003 or 2004 cohort, which is when the housing market started to pick up so the plateau was followed by an increase in birth rate that should hit its height this year.

25

One thing that has dramatically changed in the last few years is outreach. While the population as a whole may decline, T20 schools are actively engaging a group of kids who, in the past, would not have applied.

1 Like
26

Yes, that is what I had in mind. The immediate year over year decline was about 20%, and then by the next year it had recovered up to about 16% down. That is what I meant by a “modest minority fraction” but obviously the semantics do not really matter.

That is true, but I am not sure they will get there.

The data is noisy because of COVID and this test optional issue, but off hand it looks to me like Yale applications (I am using Yale because they have easily accessed summary data) have been going up around 4% per year on average (compounding), controlling for TO/COVID stuff.

The demographic cliff, in contrast, is more on the scale of about 1% a year. Like this is one commonly-cited projection, and between 2025 and 2037, they are projecting US total HS graduates to decline from 3,929,290 to 3,518,412, which is right around 1% compounded annually:

Obviously this is too crude to be precise, but that sort of shift seems unlikely to me to actually produce a decline at colleges like Yale if the other factors leading to that 4% increase keep rolling.

Yes, targeted outreach seems to have been working to increase application rates to these more “national” colleges among various domestic subpopulations who previously might have more stuck with local colleges.

And in fact that also goes along with various shifts in aid policies. Like, a lot of the wealthier colleges have been doing things like shifting to no-loan aid. This has helped them market themselves to populations where they can rightly claim attending their college might actually be less of a financial burden then choosing a more local college.

Of course their gains among such populations are losses for those local colleges. But I do think this sort of thing has not run its course, meaning I think there are still plenty more people who could potentially benefit from shifting their focus who they have not yet reached.

27

When D19 was applying, there was a lot of talk then about how the demographic cliff was about to lead to a drop off in numbers. Of course that didn’t happen, and the year she started college was when Covid hit, so possibly the TO move was responsible for some of the phenomenon of students applying to increasingly more colleges.

28

It seems very clear it was part of it, but not all of it.

Just to throw in another issue, some of the increasing applications per unique applicant seems to be related to the Common App making that easier, which makes sense. But the Common App system is not nearly at “full capacity” yet. Of course a few kids are bouncing up against the 20-app limit, but they are still much more the exception than the rule. So, there is plenty more room for Common App applications to increase even if unique applicants are not increasing, or even going down.

And for that matter, the 20-app limit is not really for practical reasons, it is just a policy. So if there was actually enough member college demand to increase it, they could.

1 Like
29

But any applications outside the common app are extra, right? So theoretically a student could apply to 20 schools on the common app, all the UCs on the UC app, all the CSUs on the CalState app and I assume other states may also have their own state apps? Obviously when you need unique essays etc it’s a lot more work but for some of those it’s just ticking more boxes.

30

Those students were likely to be born in 2001 which is when there was a decrease in births from 2000 following the dotcom recession though was about the 1999 level. The rate picked up shortly with the housing bubble in 2003 so that would not be a demographic cliff but more of a dip which you can see from the linked chart. However, since the 2008 crisis births have declined steadily for at least a decade, hitting a low in 2020 though increasing very slightly since.

31

I understand the demographics, the point was that it wasn’t correlating to college applications. Most colleges in say the top 100-150 were getting more applications and becoming more competitive, not less.even if you were looking for a dip rather than a cliff.

32

Except there was a slight dip during this recession in application numbers using the MIT data I mentioned before which I won’t link to here.

So for instance in 2017 there were 18,989. This was followed by 2018 with 18,357 and then 2019 with 18,306. In 2020, the numbers went up to 19,020. I am not sure that this generalizes to other schools and since the numbers are small, there could be other reasons for it. The demographic cliff is more than a slight dip though so who knows if the numbers will decline or if any decline in population numbers would be outweighed by increased demand.

33

Just to repeat the numerical exercise, if you use births, the decline from the peak in 2007–4.32M–to the trough in 2020–3.61M–amounts to a drop of about 1.4% compounded annually (this is higher than 1% for reasons like immigration also contributing to HS graduations).

At that scale, it is not that hard for other factors at select schools to more than compensate. No guarantees, but I think the language of “demographic cliff” has maybe painted a more dramatic picture than warranted. It is more like a “demographic bunny slope”.

1 Like
34

Interestingly, I just looked up Yale and while it doesn’t follow the two year decrease in applications, there is a year with a decrease which aligns with the MIT data.
Again I can’t link to the site but just look up Yale admission statistics and you can get the site.

In 2017 there were 29,600 followed by an increase in 2018 to 30,932 which is unlike the MIT numbers above. However, in 2019 the numbers actually decreased to 30,236. This was again followed by an increase in 2020 to 31,455. However, the 2019 cohort at both schools was lower which does match the data from the recessionary period of those born in 2001.

Note that the population decline isn’t steady rather more of jumps with some years having larger declines than others. It wouldn’t surprise me if this year will be the height of application numbers for the next few years even with outreach as the latter is more of a steady increase rather than large jumps to offset similar jumps in the other direction - that is unless a school can tap into a new reservoir of applications between one year to the next which is possible. Also more incentives can be added such as no-loan guarantees etc.

35

I highly doubt it will matter at top schools, but it didn’t happen because we are not there yet. As mentioned upthread. it really starts with the kids born in 2008. In fact, my kids '22, '24. '25 had very large classes through their school years, particularly the oldest one, born in 2004.

1 Like
36

After rereading the comments, it seems like a number of people feel that the increase in applicants is likely to offset any decrease in test scores at top-tier schools. Have we considered that additional applicants may actually contribute toward a decline in scores? Or, for test optional schools, with 50/50 submitters - they will no longer be able to increase applications because people will no longer be applying? Who would apply to a school that had a 1480 average last year with an 1170?

1 Like
37

The changes in number of applications are all over the map. There are huge variations between colleges. In general, colleges where applicants were most likely to see a high score as a barrier had the largest change in number of applicants. This includes schools, like MIT, Caltech, and Harvard. Not super selective colleges and/or colleges with lower score ranges saw much smaller changes in applications. Many schools saw little increase.

Breaking down how the number of applicants changed that were test required and test optional, I’ll use Cornell as an example, since they have more detailed numbers. Cornell had 5 test optional colleges and 4 test blind colleges. The portion of applicants applying test optional among the 5 test optional colleges in 2023 by college is below. As listed below, the number of students applying with scores dropped by a factor of between 1/3 and 1/2 after going test optional. The number of students applying without scores exceeded the total number of applications prior to going test optional.

A&S – 76% applied test optional
Engineering – 70% applied test optional
CHE – 72% applied test optional
ILR – 77% applied test optional
Brooks – 77% applied test optional

Application changes from 2019 (test required) → 2021 (test optional) → 2023 (recent) are below. I am ignoring Brooks because it did not exist prior to test optional.

A&S – 20k → 31k → 30k
Engineering – 12k → 17k → 18k
CHE – 2.4k → 3.2k → 3.2k
ILR – 1.4k → 1.3k → 1.1k
Total – 36k → 52k → 52k (44% increase)

Prior to test optional – 36k applied with scores
While test optional – 14k applied with scores, 38k applied without scores

2 Likes
38

Very interesting. Are there score ranges for the separate schools or does Cornell publish the overall range only?

Some have assumed that most students are still submitting scores so basically the original 36k minus the athletes and some others would not be submitting but additional scores would be obtained from gains in new applicants reached through outreach. However, since there are less than half submitting scores compared to the test required period, I would think that score ranges would have increased during TO.

1 Like
39

I am not sure I understand what you mean here by offset, so it is probably worth just making sure we are all on the same page. A few propositions:

(1) I personally think it is likely that at least 25ths will go down some, and maybe 50ths too a bit;

(2) I am skeptical it will get any easier for high score applicants who would have submitted anyway to get admitted;

(3) Finally, I do think a few more high-but-below-the-current-25th applicants who would not have submitted will now submit and get admitted. But I think this will be largely limited to certain types of applicants.

(2) and (3) do not contradict (1), or offset it. Indeed, they would both in their own ways contribute to (1).

1 Like
40

I think that is a very fair assumption. The game hasn’t changed for those kids.

1 Like