Can someone share their knowledge about the ACT versus the SAT? Here's what happened:

DS took the ACT in mid-grade 6 when he was 11, and had a
26 composite, 84%ile of high schoolers, 96%ile of Numats 6th-graders.
He took it again this year as a HS freshman, had a
34 composite, 99%ile of high schoolers and Numats freshmen.
So that's all fine. He also just took the SAT, and had a
1310 combined math/critical reading, 89%ile of high schoolers and 87%ile of Numats freshmen.

He thought that the SAT was so easy, but this score does not reflect that! He typically scores in the 99%ile on CoGat and our state assessments and 94-99%ile on the NWEA MAP, was subject-accelerated to 4th-grade math as a 1st grader, and was in our district replacement LA and Math classes through 8th grade. While chronologically a freshman, he is taking mostly soph honors classes.

It seems that this score difference is far more than the "different day" score difference one might see, and certainly doesn't jibe with act/sat conversion charts. Are the two tests that different, that a highly abstract thinker would not score as high? Or are they scored so differently? Or is the SAT more high school curriculum based than the ACT?

To be clear, I'm not asking how he can raise his score, he did not "prep" for any of the tests beyond taking one sample test each, spread out over a few weekends. In fact, before the SAT he practiced only 2 sample sections. I just wonder what might be behind this great variance.

Does anyone have experience with such SAT/ACT score comparisons?

No idea about systematic score comparisons, but my DS also thought SAT was really easy and ACT was much harder (but he scored really well on ACT as well). The way I see it (by looking at old tests) is that SAT is easier to prep.

DD took both-- SAT at 13, and ACT just after turning 14, six months later. Because of her acceleration, she did this between junior and senior years in high school, however.

Her scores on them both were (overall) 99th percentile, but she found the format of the ACT much MUCH easier.

The SAT? It was too much down time, too often. It really put her off her game to have it so choppy-- the short sections were awful.

One thing worth noting is that her practice scores for the ACT were about what they were for the real thing, and with the SAT, her practice scores were at the ceilings, and the real thing was 50-60 points below practice scores. In the case of the math sections, that meant a sub-700 score for her, which is mind-boggling given what I know of her math ability.

ACT, much better reflection of her abilities.

I had always heard that the SAT was a better for abstract thinkers and ACT better at testing knowledge. Not sure if that is true our school recommends students at least try a sample of both before deciding. My daughter took the ACT and I always assumed my son would do better on the SAT. (Not sure because of the upcoming changes.) But I have never had DS15 take either test yet. The SAT currently penalizes for guessing could that maybe have effected your sons score. Perhaps the problems he did get wrong lost him points?

When I was in school I often did poorer on test that I thought easy. I would get lazy and make more mistakes.

I wondered about the choppiness. Each section was 25 minutes, and DS said he finished in 15 leaving lots of down time...

We found the ACT practice scores were good predictors of the real thing, too, but he didn't take enough of the SAT practice test to predict. We weren't really into the prep thing, just wanted a feel for the test before taking it.

Are you supposed to approach the SAT differently?

Honestly? I have no idea. That's what DD did to 'prepare' for them both. Mostly, for DD, this amounted to finding her "ideal" pace-- and for figuring out a strategy for the math sections to reduce errors in computation, which are the bane of her existence.



DD thought that the SAT math sections in particular were about "trick" questions. By that she means that prepping helps a TON with them, because you have to know the One Right Way of doing the problem-- there's no way to really leverage general math knowledge to do them a less efficient but still valid way. KWIM? Her impression was also that there were weird types of questions on the SAT just in general-- things that if you weren't prepped, you would have almost no shot at, just because you had to understand HOW you were supposed to interpret/see the test item.

The ACT, she definitely felt was less of a "prepping" fest.

The ACT was a matter of MUCH more significant pressure on reading speed, though. DD liked that-- it was much more like the AP test format. Less down time for DD, but she enjoyed it much more because she was able to get in the zone and stay there for longer.


Most students never even finish the sections in the ACT. DD had little time left OVER in the Science section, but the rest, she definitely had lots of time. Strategy-wise, she used that time in the math section to immediately begin working her way BACK through the test items until she ran out of time-- that meant that for the last 2/3rd of the section, she worked the problems twice. The other two sections, she worked them through at her ideal pace and then just waited.

This could be key, since what you are seeing are percentiles, how your DS performed relative to others. It's not an absolute measure of performance. So if lots of his peer group is successfully prepping for the SAT, that will artificially bring down his scores on that test.

My eldest took both her junior year. She did some prep for the SAT, but just took one practice ACT test to get an idea of the questions and timing.

Her SAT CR+M SAT score was equivalent to a 33, actual ACT was 32, and three part SAT was equivalent to a 31 (didn't like the writing section on the SAT). She thought the ACT was all stuff you should have learned in school, especially the math. Math is definitely her weak point, but her highest subscore of 35.

My 10th grader just took the SAT, just to "see her baseline score" (errr, why not prep?). Her three part score was 97th percentile, slightly higher than her sister's three part superscore. It will be interesting to see how she does on the ACT, because she is really good at standardized achievement tests. However, she is clearly more of an abstract thinker than her sister, so I can understand their SAT scores.

Some kids see a big difference in ACT and SAT scores, some don't. The good thing is that all colleges now accept either test, so he can just go with the ACT for college applications.

Bluemagic, funny you should say that about abstract vs knowledge. I really thought DS would do better on the SAT because his teachers describe him as a highly astract thinker. If, as Howlerkarma DD says, the SAT questions are tricky, could that throw even abstract thinkers for a loop, backfire on them in a way? And the ACT was funny because the science portion is more of a reading test than science knowledge, although you should certainly have some school science to approach it.

Maybe DS should have taken more of the practice test sections! Maybe he took the tests too close together, just a month apart, and he was influenced by the one, and maybe even feeling cocky? (He'd found errors in the ACT prep test answer guide, but not in the SAT parts he'd done, maybe that threw him?)

MegMeg, that's a good point about percentages and prep. Though he was 87% nationally, he was only 57% in our state! But ours is more of an ACT state, and we have lots of controlled-admission public and private college prep high schools with kids who want to go to the coasts for university and are prepping for the SAT. So the kids who take the SAT in our state have every reason to prep hard.

I don't know that there is any real evidence to back this up-- or if anyone has actually looked at this effect at all, even-- but I have to wonder if HG+ kids tend to do increasingly worse on some types of achievement tests as they outgrow them. Well, "outgrow" isn't the term that I'm looking for, here, but basically what I mean is that if the SAT is "appropriate" as a tool for measuring a particular 6th-7th grader, then that same student as a high school sophomore is probably taking the SAT as a "below level" multiple choice assessment.

Okay, with that in mind, I feel that the ACT is much better-constructed. It's.... just...

well, the questions that I saw in prep materials were tighter. Less ambiguous or open to interpretation in different levels, if that makes any sense.

That's why I think that my DD probably found the ACT a better reflection of where she is NOW.

At 12, the PSAT was already somewhat below-level for her-- and it showed as she took practice tests. Her responses to looking up wrong answers wasn't; "Ohhhh, so THAT's how you do that..."

It was more along the lines of "Ohhhh-- so THAT is what they wanted me to read into that..."

Which, in my mind, is a mark of a test that is too level-locked to be very good for students below or above that level of cognitive function.

She was simply not reading the questions the way that a sophomore with the assumed background would do; aside from the grammar and math sections (which are very right-wrong and non-ambiguous), the test items were a struggle for her since she had to put herself into the position of "what would my classmates be expected to read into this item?"


For an example of what I'm getting at here, try taking a quick reading practice yourself-- with both ACT and SAT. I do much better on the ACT version, because it doesn't assume that I have any particular scaffolding beneath me to support interpretation, and therefore it's less prone (IMO) to over-thinking. KWIM? In looking at her practice tests for any patterns (particular types of questions, etc.)-- there wasn't a pattern of what she missed, usually-- I often found myself scratching MY head at the SAT items, and pretty much never with the ACT ones.

Could just be my family, too. I'm a zen-master of the standardized test, and DH and DD not so much. I tend to test probably better than I should, and both of them about as they should or a bit worse than that on the day.

I don't know about the ACT, but my impression of the SAT is in line with what HK's dd reports. Looking at SAT prep books, the SAT math is fairly easy theoretically and computationally, but the problems often involve a series of calculations and a twist or trick at the end. So, for example, perhaps the hard part of the question is that you have to factor a quadratic equation and it takes awhile to do so, but the question asked in the end is not the value of x or y but is (due to a twist in the question) actually 1/2 x. Maybe that's not a good example, but what I'm getting at is that I and my dc's tend to see a math question and jump to what's hard and what's (usually) important, which in this case would be doing the factoring right. We might process it so quickly that we don't read the extra sentence at the end saying that the value requested is 1/2x. That's why the SAT gives an advantage to preparation: you will have seen that trick a million times and will watch for it.

The tests have also differed in scoring: The ACT has not traditionally had a penalty for guessing, whereas SAT has.

And by guessing, they mean "guess and be wrong" as opposed to "guess and be lucky"?

DS does not like tricky. He'll make things tricky enough himself, and he "intuits" too much especially in math. This happens in class, too. (His chemistry teacher calls him The Bond Whisperer for intuiting covalent bonds.) Or, he'll read something different into a question than was intended.

Thank goodness the SAT essay didn't matter this go-round, because the question was so odd, it was leading and unclear at the same time...not to mention, they just haven't done that particular kind of writing in school. Still, the scorers gave him an okay mark...and yes, I read the "confessions of an SAT essay scorer". We can view the scanned copy, which is of such poor quality I can't read it.

I confess, I did not take the practice tests with him. He took them, I scored the bubbles, and he went over the answer guides (where he found the publisher errors). Very hands off for me (except for contacting the publisher and the library where we got our book).

I suppose we will wait and see whether he takes the SAT again, next time for reporting purposes. Of course, by that time the new test will be in place. Most students around here take only the ACT.

It's been quite a while since I've taken either one, but this is what I remember feeling after I took both tests in high school. On the SAT there was a lot of "I think this is the answer that they want, but that answer is making assumptions which may or may not be true." I felt that I did much better on the ACT because I didn't spend so much time wondering if the test designer was making logical leaps.

You have lots of good advice and suggestions above. I'll just throw one additional variable into the mix - any one test is just that, one test. Outside factors that have nothing to do with the specific knowledge required to perform well on the tests might come into play on any given day. When I think of SATs the first thing that comes to mind is a vision of the room I was sitting in when I took it.... something like eons ago. Same for the GRE. I don't remember anything about the ACT test setting. The reason I remember the SAT and GRE tests is that the environment was totally not conducive to being able to concentrate - on my GRE, there was a phone ringing in the next room - non-stop - that couldn't be taken off the hook because the door was locked and none of the test proctors had the key. So I wouldn't discount the possibility that something in the environment on test day impacted the score.

polarbear

According to one study SAT scores are higher at test centers with fewer students. A summary "New Research: Taking the SAT in a Crowded Room Means Lower Scores" is at http://www.nurtureshock.com/NEffect.pdf , and here is the paper abstract:

http://pss.sagepub.com/content/20/7/871.abstract
The N-Effect
More Competitors, Less Competition
Stephen M. Garcia and Avishalom Tor
Psychological Science July 2009 vol. 20 no. 7 871-877
Abstract
This article introduces the N-effect—the discovery that increasing the number of competitors (N) can decrease competitive motivation. Studies 1a and 1b found evidence that average test scores (e.g., SAT scores) fall as the average number of test takers at test-taking venues increases. Study 2 found that individuals trying to finish an easy quiz among the top 20% in terms of speed finished significantly faster if they believed they were competing in a pool of 10 rather than 100 other people. Study 3 showed that the N-effect is strong among individuals high in social-comparison orientation and weak among those low in social-comparison orientation. Study 4 directly linked the N-effect to social comparison, ruling out ratio bias as an explanation of our results and finding that social comparison becomes less important as N increases. Finally, Study 5 found that the N-effect is mediated by social comparison. Limitations, future directions, and implications are discussed.

Here is an interesting article regarding test performance (which raises many questions for a follow up research study): APA - 2007 - Seeing red impairs test performance. I'd be curious to know how they controlled for other variables among participants (breakfast, sleep, etc).

LOL, yes! Penalty for wrong answers. At least until biometrics are assessing whether a correct answer was made with certainty or by eliminating some possibilities and then guessing. (Chilling)

Interesting. Here are the possible mitigating factors the day DS took SAT:

-- could not find EITHER of his acceptable calculators. Solution: borrowed the identical calculator from our neighbor.
-- could not find EITHER his school or state IDs. Solution: he used his travel id. BTW, we handle always these things calmly, matter-of-factly, no frantic search or "WHY don't you have them" type drama.
-- test was in a nearby school he'd never been to. But, he did see a friend/classmate.
-- classmate was in his test room. Hmm, is there a competition factor? DS is not competitive, but the friend is.
-- the test was at least an hour and a half longer than the ACT due to the mandatory essay, but the essay came last.
They put about 20 kids to a room, so it's nice and small but you are totally aware of how many finish before you.

On the other hand, the first time he took the ACT in 6th grade, he was self-conscious about being the youngest kid there though he did see a friend, he was one of the few not writing the essay, and was taking it the same time as his HS junior brother, and he did very well.

Shivers!

In my limited experience, some kids test about the same on both the SAT and the ACT while other kids do worse on the SAT. For some reason, the other way around seem less common, although my samples are skewed towards the high end. The ACT is achievement oriented and appears to test basic straightforward knowledge covered in standard high school courses. The SAT is still kind of a hybrid ability/achievement test although the current version isn't accepted as IQ equivalents anymore. The SAT appear to required somewhat more fluid thinking rather than basic knowledge.

This likely doesn't apply anymore but the SAT and the ACT cohort used to be very different. In the 1980's, top schools on either coast did not accept the ACT and many top students viewed the ACT as a basic achievement test.

I know nothing about these tests, but if the SAT has these kinds of "tricks", then it is prudent to at least spend some time to be prepared for them.