An 'Easy' SAT and Terrible Scores

@bluebayou

Back then you didn’t need too either! Just like my FIL paid his way through LSU and Tulane working during the summer 100% not applicable to today’s climate.

The whole process is ultra competitive kids are busier now then ever at some point it has to stop there aren’t enough hours in the day or you have be “hooked” it is crazy!

yes, but there is no guarantee that he would have only missed 2 with a harder test. The classic, if a train leaves Chicago heading east, while a train leaves Cleveland heading west…can take some time to solve.

In other words, he could have missed 4 on a more difficult Math section and still scored a low 700. (That’s how curves work.)

A 1500 is an excellent score and concords to a 34 on the ACT. No shame there. With the two 750 subscores, his confirming index for NMF is 225! Congrats to him and good luck this upcoming year - he’s in very good shape for NMF barring C’s or a snarky essay (neither sounds likely if he’s worried about a 1500 on his SAT . . ).

I understand that. He made a 1510 on his PSAT with a perfect score in math, so he had really high hopes for his SAT score.

@bluebayou I haven’t read the last few comments you’ve made, but did you read the WaPo story? Taylor got one more question right on the June SAT and his score dropped. How does that make sense? Your argument is derived from not even reading the passage fully. And missing one more question doesn’t mean that you drop 90 points in total. I got 4 questions wrong on the June SAT, which is 7 less than what I got wrong on the October 2017 SAT, and I lost 30 points. In what world does cutting your losses by 60% result in losing points?

I am trying to get multiple scholarships because my parents might not be able to scrape together the money for me to go to a decent college for many years, but my parental income has just jumped past 135k so I won’t qualify for enough financial aid. This test actually matters to me in my life. If you honestly think retakes are useless, then that’s just a perspective borne of ignorance about that actual situation that many families experience.

Sorry, I never, ever wrote that retakes are useless. I’m a long-time poster on cc, and my mantra has (r been: higher is (almost) always better. And I have also wrote on cc numerous times that colleges don’t care how many times you retake – it’s in their own best interest to not care.

And yes, for admissions and merit money, the scores count, oftentimes a LOT, particularly for the latter.

But its also important to note what a curve represents. An easier test is just that (and you will find that out in college if you take any STEM courses).

Yes, the June test administration sucked, but that doesn’t mean that it is statistically invalid. As long as it is considered by CB to be statistically valid, the colleges will see it that way.

The only thing you can do is continue to work hard and retake in the fall. Good luck.

Edited to add: just read your other posts areo. With a 35 ACT and 1520 SAT, not sure why you are wasting time on a Saturday morning. Its highly unlikely that a higher score will make a difference with merit money.

“He missed one more question overall in June than in March but his score, he said, dropped precipitously. And in the math portion of the exam, he actually missed fewer questions but scored lower: Taylor said he got a 770 in March after missing five math questions but received a 720 in June after missing just three math questions.”

  • From the WaPo story.

So young Taylor did better on Math and worse on EBRW and on net missed one fewer question. Reading.

But anyway, big deal. Had it been a harder test, with the same number wrong he might have had a higher scaled score. Comparing raw scores will only cause grief and demonstrate that one is ignorant of what scaled scoring even means. For instance, @DynamicAero, you missed seven fewer questions in June than you did in October. Which section? Each one - Reading, Writing, and Math - has it’s own scaled score. Reading and Writing are actually AVERAGED so combined improvements in those sections will always look “better” than what turns out to be the scaled average.

No one says retakes are useless but taking more than two tests or retesting when you are already in the 99th percentile or concorded with a very high ACT shows a fixation with testing. If @bluebayou is correct, you already have a 35 ACT and 1520 SAT, so what exactly are you trying to pursue by additional retakes? For instance, which school is going to deny you scholarship money because your standardized tests are simply too low?

^ Ha. Should take my own advice: on net, Taylor got one fewer question correct; he missed one ADDITIONAL question. Reading. LOL. Thankfully I’m not taking the SAT.

I got 3 wrong on the Math section and 1 wrong on the Writing section.

When it comes to scholarships, having higher test scores gives me more of a chance to get more scholarship money, and using Scholly I can see that I could triple my earnings by applying to as many scholarships as I can afford. There are a lot of private scholarships that use SAT scores to determine if an applicant is awarded the money.

Finally, the reason I’m going to take the SAT for the third time is because I wanted to improve on the 1520. That was the whole point of taking the SAT once more in June. I tried to limit myself to three administrations of both the SAT and the ACT total, but there’s no point in limiting myself because my SAT score dropped from my first to second administration. I can’t leave that out of the equation now because my Writing/Reading score was actually higher in June than in October, so to superscore I’d have to use the June SAT. But I don’t want to show that I’m a variable test taker, and I also want to get an 800 on the Math once and for all to get a good superscore.

Isn’t it clever how they make you confirm your PSAT score with the SAT? An ACT can’t be used. And presumably, he took the ACT and is using it for college admissions. The College Board gets all the money from subject tests and AP tests too, an octopus that covers almost all the tests you have to take in junior year, except the ACT. And then for NMSF, you have to take the SAT too.

I don’t think that it’s a shock to anyone that the CB looks for any way to collect money. Change a test? Pay a fee. Cancel a score? Pay a fee. etc.

NMSC is the organization that decides how the National Merit program works. They call the shots on how to vet your PSAT. They actually call the shots on using PSAT in the first place.

Aside from that, it makes total sense to vet with the SAT. For one, PSAT and SAT historically have been similar tests with similar content - they have become even more aligned with the revised test. For another, the old conversion tables for SAT and ACT (prior to the revision) were out of whack, and accurate concordance tables for the revised test were jointly published just a few weeks ago; consequently, any “vetting” using ACT would only lead to a massive headache for NMSC. Finally, why should NMSC work with two different testing companies when one takes care of the job? There’s also the issue that regardless of whether ACT or SAT is used, the tester needs to fork over the cash to take the test (barring financial difficulty in which case a waiver might be granted).

College Board has become a behemoth, running everything from standardized testing to applying for financial aid. But ACT isn’t hurting for funds either, and they can always introduce their versions of subject tests etc. Plus, at this point more testers take the ACT than the SAT so they still dominate that particular market.

Still waiting for one of these - or a third testing agency - to conduct a computerized version of a standardized exam that truly measures accurate performance at the high end of the curve. This can be done simply by individualizing the progress through the exam, asking the tester harder and harder questions till a certain number is finally missed. Other psychometric tests are conducted this way so it can’t be that hard to set up. The issue is whether college admissions offices are looking for this type of testing. As they are heading away from using SAT/ACT as an accurate assessment of cognitive or academic ability (via superscoring, Score Choice, or Test Optional) it’s likely that the answer is “No.”

“But I don’t want to show that I’m a variable test taker, and I also want to get an 800 on the Math once and for all to get a good superscore.”

What is your ACT Math subscore? Why not take the Math 2 subject test to show math-ability?

Keep in mind that your chances of getting another set of low(er) subscores might actually be greater than those of achieving a high(er) score in math, especially if your first SAT happened to land on the higher end of your “true” distribution of test performance. There’s a reason why CB provides a range around your total test score along with the following message: “This range shows how much your score can change with repeated testing, even if your skills are the same. It’s a better representation of your true ability.”

Yes, we understand that aero. But what data source are you using that shows a difference in scholarship amounts between say, a 1520 and a 1550? Or, in reality, a ~1560, since that is already what you have in the ACT. I really don’t believe that a few extra points will be of any value in your case. You are already at a the 99th %. (There aren’t any special scholarships that I know of for 1600’s.)

btw: for college admissions, the best way to demonstrate your math ability is on the Math 2 Subject Test since it has higher level math questions.

I’m guessing they look to course rigor and grades and perhaps math ECs to demonstrate high end math ability.

D had a friend who ran through the math sequence at their HS in 8th grade and proceeded to dual-enroll throughout and aced several 400 level courses. Many math ECs.

You think colleges needed a special test score for that kid to know he was advanced-to-gifted in math?

"D had a friend who ran through the math sequence at their HS in 8th grade and proceeded to dual-enroll throughout and aced several 400 level courses. Many math ECs.

You think colleges needed a special test score for that kid to know he was advanced-to-gifted in math?"

  • Not everyone has access to things like dual-enrollment or gifted-ed.

Sure, but why? Why does a college care if the applicant is twice or thrice gifted?

Not likely about it. Colleges have no interest in parsing the 99th percentile. (Well, they can parse the SAT-M with the Math 2 subject Test, but that could also be made even more difficult if CB wanted. They don’t, as their customers – the colleges – don’t as the finer detail is not necessary for their goals…)

“Sure, but why? Why does a college care if the applicant is twice or thrice gifted?”

“Not likely about it. Colleges have no interest in parsing the 99th percentile. (Well, they can parse the SAT-M with the Math 2 subject Test, but that could also be made even more difficult if CB wanted. They don’t, as their customers – the colleges – don’t as the finer detail is not necessary for their goals…)”

  • Agree that colleges may not profess to care - they are trending away from standardized testing with each passing year. Students and families may care, as such a change would take care of any sloppiness such as tests designed to measure the median performance, rather than the ends of the curve. For instance, if CB's equating methodology accounts for differences in average difficulty, as opposed to differences in difficulty at the high end of the performance curve, then occasionally nightmares like the June test will inevitably occur. A test that allows for individualized progression takes care of such problems; scores at the highest, the lowest, or any other part of the distribution are likely to be equally accurate in representing ability. Such a test may also discourage countless retakes since you are far less likely to land at your score by chance (and thus far less likely to improve it, as many who are constantly re-testing are attempting to do). Finally, unless and until colleges reject standardized testing altogether, it's a hard sell to convince anyone that they don't still use these to measure "aptitude" - especially as they publish the metrics and include them in their news releases about the incoming class. The students and families certainly aren't fooled. So as long as the metric is still in use, why not make sure it's done right?

Every year on cc there are a handful of kids/parents that bemoan the fact that the SAT-M does not better parse the tail. But then there are plenty of other tests/competitions to do that.

Philosophically, sure. But practically, not possible. First, how do you determine the “right” test? Think about IQ tests, for example. The basic ones aim for a mean of 100, but they all have different nuances that are better for one kid or another. Then there are numerous other tests that can test the differences between a 132 and a 133 or a 160 and 165+. But those kids are so rare, there is no money to be made developing such a tests on a global scale.

And, and individualized test is even more problematic. The GRE tried it with disastrous results. Miss the first question and sorry, no chance for the highest score, even if the rest were correct.

Discouraging retakes disadvantages those kids with test anxiety, of which there are many with real anxiety.

Kids can show that they are stellar math students by sending a strong SAT 2 Math 2 test (which is through pre calc) and self reporting an AP Calc BC score. That will back up a strong SAT-M score.