Razor Thin Difference between a 33 and 36 on the ACT

Test Preparation
1

This may be old news, but if my math is correct, I find it surprising at how little separation there is between an ACT 33 and an ACT 36.

I stumbled across this 2013 article about the percentage of students that receive a perfect 36 on the ACT.

https://www.■■■■■■■■■■■■■■/blog/2013/12/06/the-odds-of-perfection-too-many-perfect-scores-for-the-acts-good/

I’m from California, so I focused on the numbers for that state, but the same approach can be applied anywhere, or using the overall percentages and you’d get to pretty much the same place.

According to the article, California produces 1 perfect score out of 1,075 test takers. If we round that number down to 1,000 to make the math easier, that’s 1 perfect-36 score out of every 1,000 test takers.

According to the official ACT site, the 99th percentile includes any one of the fourt top scores of 33, 34, 35 and 36:

Score Percentile
36 99
35 99
34 99
33 99
32 98
31 97
Etc.

http://www.actstudent.org/scores/norms1.html

So, here are the mathematical observations: (1) the 99th percentile includes anybody who got a 33 or higher; and (2) using simple math, the 99th percentile of 1,000 test takers only includes the top 10 students in that pool of 1,000.

That means… Of those top 10 out of 1,000, one got a 36, and the other 9 spots are shared by students who earned a 35, 34, or 33. That’s cutting it pretty thin.

If this is accurate, then a 33 is astoundingly close to a 36; separated by a proverbial hair.

2

Yep.

3

Well, if you look at the distribution of ACT scores, you will see that the ACT scores are roughly normally distributed with a mean of 20-21 and standard deviation roughly 5 (I’m just making up numbers, they might be slightly off). So a 33 is already 2 to 2.5 standard deviations above the mean.

If you look at a perfectly normal distribution, a z-score of +2 is already at the top ~2.5% of scores, and a z-score of +2.5 is at the top ~0.7% of scores.

4

I didn’t get what @MITer94‌ just said but that is normal, he is from MIT and I went to a state college ;).

5

@sanwal Miter94 was just talking about statistics. If you took a stats class in college or took AP statistics in highschool, you would have learned about z-scores as well. The college doesn’t matter.

6

I’m missing the point of this post. There are very few people in the top 1% but that doesn’t mean there isn’t a significant difference between a 33 and and a 36. It’s not just luck or a “proverbial hair,” no more than it is for a 28 to a 31. I wouldn’t take your observations and, say, extrapolate that anyone who has scored in the top 1% should take the test three more times and they are bound to get a 36 eventually. Statistics don’t work like that.

7

IMO The difference between a 36 and the 33 is the amount of questions you get wrong. 36 would mean virtually no questions or one question wrong on the sections. 33 allows for some more wiggle room.

8

36 looks cooler on your app.

9

Razor thin until you are the 33 LOL.

10

[QUOTE=""]
Razor thin until you are the 33 LOL.

[/QUOTE]

I guess that’s my point. While a 33 sounds kind of meager when compared with the perfect 36 on the 36-point scale, the context of it meaning that you are both in the top 10 out of every 1,000 test takers seems pretty darned good. In batting averages, that’s a .990 and a .999. Both would be awesome.

11

It matters, to an extent. The difference between like a 31 and a 34 is probably much more prominent and important than 33 and 36 because the general consensus is that 33 is the magic number that you can pass wherein your scores will generally not hold you back at any college in the U.S.

12

I work as a college admissions consultant now and as someone who got into a top 10 college with just a 1750 on my SAT I can tell you that no college will care if you have a 33 or a 36. They look at those scores just to see if you are ready to survive college academics and with a 33 you can survive any college in the nation. Don’t worry focus on other areas such as your essay and ECs. If you want a longer explanation or you need help just let me know.

13

But is there a larger difference for scholarships?

14

@parentmike , I am not sure how long ago you got into a Top 10 college with 1750 but it is now almost impossible, unless you have some other stellar achievements or a major hook. If you look at the data, things have been getting tougher and tougher.

Note: I am just making the above comment so that perspective students have realistic expectations. No offense.

15

The entire right hand side of a bell curve is razor thin, particularly if you are talking about individual numbers, but there are still at least 5 or 6 standard deviations out there.

16

@sanwal‌ A person on CC with a 1750 got into Yale. Not sure about other achievements, but they weren’t “stellar”.
Just saying.

17

You need to think of it a little differently than only tiny percentile differences. There are 3.3m high school seniors graduating. Of these 1.8m took the ACT. 1200 or so got a 36 and 97th percentile would be about the 54,000 student. There are 14,000 Ivy League freshman each year including Cornell which is larger and a bit less selective. If you include MIT and a Stanford, the total number of freshman is around 20,000. You are correct that there isn’t a lot of statistical difference between a 33 and a 36 but small differences matter when you are talking about top colleges. My guess would be an admissions officer sees a 33 and “check” you have the standardized test scores needed so they would look at other aspects of the application. At some schools though the difference between applicants are very minute and a stronger score makes a difference. In states like California where the ACT is not as popular only stronger students would take the tests so it would tend to skew your state averages upward. Same is true for us in Illinois only its the SAT average that’s higher. I think a 36 would probably be something a Harvard admissions officer sees daily but one or two levels down at less selective schools it might be more impactful. By the way 36 is 99.93%, 35 is 99.5%, 34 is 98.8% and 33 is 97.7%

18

33 with the same weight as a 36 is just a myth low test takers say to make themselves feel better, 1 in 1000 people is a lot better than 1 in 100.

19

@qpqpqp and calling 33ers low testers is a pathetic lie that 36ers who really have no self esteem tell themselves to reassure themselves that they’re “better” than other people just because of one test. It’s a difference all right, don’t get me wrong, but not so much that you get to treat me as if I were an idiot, you pretentious, condescending snob.

20

Can’t comment on the ACT because my kids only took the SAT - but on one round of tests omitting 2 on the Math section resulted in a 720! For omitting 2, for Pete’s sake! On a different day, one might easily miss none.

So while 800’s and 36’s sound impressive, I agree with the OP that there is probably little difference between those who score the 33’s and the 36’s (or on the SAT the 750’s and 800’s). Part of it is luck.

Probably, if you were tenacious and talented enough, you could get perfect scores or superscores, but kids shouldn’t be robots. Why return to your own vomit over and over for a perfect score?

Some may think it is worth it for bragging rights, but I personally do not - nor is it healthy!