Hacker News Reader

Comment by cameldrv | Hacker News Reader

This is very unfortunate. Yes, you can improve your SAT score by studying somewhat, but not massively. The College board has made a free online prep course with Kahn academy that achieves similar results to professional coaching if the student puts in the time. The fact that a student is willing to put in that time should be an indicator that they're likely to be successful in college.

Where more of the abuse is is in parents getting disability diagnoses to get extra time on the test. This same technique can be used to game grades though too.

View on HN

> Yes, you can improve your SAT score by studying somewhat, but not massively.

This is completely mistaken.

Just taking a couple sample exams will net you almost 100 points as you don't waste time figuring out how to take the test.

Simple prep will net you 200+ points right off the top.

For example, generating a correct answer is almost always way slower on these kinds of tests than quickly removing wrong answers. Knowing where you are on the statistical curve and knowing how the test is catered toward tripping your precise cohort up is extremely important when you have a binary choice remaining.

And that's before you start doing actual systemic improvements like memorizing a couple thousand vocabulary words or drilling basic arithmetic so you don't make simple, stupid mistakes.

> Just taking a couple sample exams will net you almost 100 points as you don't waste time figuring out how to take the test.

You don't go in cold for the SAT. They give you enough free practice resources that anyone serious about taking the test should get those 100 points. Maybe to a practice test every weekend for a month leading up to the test. It's the tutoring where you have to invest a lot more time and money for more marginal gains.

People do go in cold for these tests.

There are many people who come from communities who have sent precisely zero people to elite schools, and they have absolutely no idea how to prepare or how to be a strong candidate despite an abundance of information available on the internet.

Most people assume that straight As and a really good SAT score makes someone Ivy material, and that’s why they are often confused when folks who meet these criteria don’t get in.

> You don't go in cold for the SAT.

A lot of people do.

The mere idea that test taking is a skill which can be honed like any other is an upper middle class social marker. In the case of the SAT, striking out wrong answers is FAR faster than generating correct ones unless your are really good. That's a test taking skill.

My parents were both school teachers and even they didn't know all that much about the SAT--they knew what it was and made sure I ran some practice tests. My guidance counseler (normally a totally useless position at most schools), however, was an old dude who was used to the fact that children of steelworkers had no clue how to prep for college.

He pulled my parents (and others) aside in 9th grade (almost 2 YEARS before you take the SAT for real) and said: "Here is the schedule of tests (PSAT, SAT), etc. and the order in which your child needs to take them. Note that the scholarship score is TWO TIMES THE VERBAL plus the math. Here are the sample test resources your child needs. Here is the vocabulary resource your child needs--that defines your scholarship chances. Your child needs to drill math to not make stupid mistakes. Complete these at an even rate for the next couple years and your child will qualify for most scholarships and have no issue getting admitted to any college below Ivy level. If you really want to go to an Ivy, you need the scores, but we'll need to find you an alumnus to advocate for you in order to boost your chances."

That man created a LOT of first and second generation college graduates.

I am not sure why you are being downvoted (I upvoted you), but your statement is largely correct.

1. 50 points per section (M/V) just for being familiar with the test is common for folks in the middle range of scores — say 400-600 per section. The improvements are less at the extremes.

2. Very basic test taking strategies are worth another 20-30 per section for pretty much everyone.

3. Actually study/review of targeted vocab and math will typically yield another 20-100+.

4. I don’t have as much data on this, but I personally think practicing sitting down on a Saturday morning and taking a high stakes test for x-hours in a sterile environment also increases the score 10 points or so. It reduces anxiety at a minimum.

I’ve seen scores go from 10xx-14xx with a 200+ point improvement in both skills over the course is about two months.

> I am not sure why you are being downvoted (I upvoted you), but your statement is largely correct.

I don't know either, but there is something going on that seems to be time sensitive. Comments that hit quickly after posting sometimes receive significant downvoting while the same comments a couple hours later get significant upvoting.

Something is going on. I just don't know how people benefit from it. Are they trying to create a "history" ahead of the election cycles to whitewash their reputation or something?

I really wish that upvote and downvotes had your username tied to them in a visible way. I couldn't care less if random noname downvotes me if someone important upvotes. And, if someone I respect downvotes me, that's going to certainly get me to think a bit harder.

I'm not convinced more time on tests will actually help somebody who doesn't legitimately have some relevant disability.

Generally tests are structured so that if you know the material, you'll have plenty of time to spare. But if you don't know the material, it's unlikely you'll figure it out even if you're given twice as much time on the exam. It's not like they'll spend that extra hour teaching themselves the trigonometry they neglected to learn in the weeks/months before the exam.

This is not true for the SAT. Having extra time to double check answers or work out harder problems will definitely lead to a better score. Speed is intended to be a factor in the SAT.

Maybe so, it's been many years since I took the SAT and I can't remember much about that experience. But I think with respect to more general academic performance, extra time on tests won't usually help much. In my experience in college, very few of my peers ever used up the full hour or two given for exams. The ones that did seemed to be worrying themselves sick double-guessing themselves, and it never seemed to do them much good.

I had one friend who I suspected of having a bogus disability. He had rich parents and got double-time on exams. But it's a miracle he graduated at all, he seemed to always be on the brink of flunking out. I think that had a lot to do with his video game addiction that stopped him from studying much. He was a reasonably bright guy with a horrible work ethic, and I really don't think the extra time on exams helped him at all.

I have embarrassing personal experience here.

I don't have a learning disability, but I'm a pretty slow reader. I cheated on a section of the SAT by finishing another section with lighter reading and going back to the unfinished one.

I'm not familiar with the SAT, so I might be misunderstanding. But how is returning to an earlier section of the paper after you've completed a later section considered cheating? In all the exams I took over high school/university we could complete the paper in any order we like.

The SAT is divided into sections, with a time limit for each one. If I recall correctly, each one is 10 - 30 minutes, with multiple sections making up reading, writing and math.

The test is also designed with time pressure in mind.

Ah I see, thanks for explaining.

In SAT you're not allowed do this. Each section has fixed time, then pencils down.

I've taken exams in college where the professors intentionally put more questions on the exam than you have time for. They say, "think of each question as an opportunity to increase your score" as opposed to caring about answering them all. The score is then standardized at the end, since most results are below 50% correct. It turns out exams like this usually give a very normally shaped distribution. More time is a massive advantage in these cases.

The examiners did this by mistake in one of the papers in my final exams. I had time to finish two of the three sections. They thought that was fine.

Fortunately, the external examiner didn't, and required them to add points to everyone's marks. I ended up with slightly over 100%.

I guess that such tests are more influenced by general intelligence. That is OK for IQ tests or admission tests, but course exams should test mainly knowledge of subject matter, so they should generally provide enough time.

More time is only an advantage if you get it and others don’t.

The time pressure forces others to not get it. You getting it does depend on your skill to a large degree, but extra time increases your probability of getting it.

Reply on news.ycombinator.com