What Does the SAT Actually Measure? (Or 'In Cautious Defense of the SAT')

There's a lot of controversy over what the SAT actually measures -- and, in fact, whether it actually measures anything at all. Although the test was originally conceived as a form of IQ test, the notion that it is actually capable of measuring innate scholastic ability has essentially been debunked, as has the notion that it can successfully predict a student's ultimate success or failure in college (the only thing it has been shown to correlate with is freshman college grades). So the question remains then: if the SAT does not actually measure intelligence or academic potential, what on earth does it actually measure?

My response, thoroughly non-empirical and based strictly on personal experience, would be as follows: the SAT (Critical Reading) measures students' ability to understand, summarize, make simple inferences, and compare arguments from relatively sophisticated texts -- in other words, skills that students at competitive colleges must at minimum possess in order to be successful. After all, if you can't truly understand someone else's argument based on a close reading of specific textual elements, how can you possibly formulate a coherent response to it?

Hence, at the most competitive schools, a high score on the SAT is viewed as a baseline criterion for admission rather than as a remarkable achievement. The test favors the kind of rigorous, precise, linear thinking that the academic world is based on. Academics, like everyone in the world, tend to gravitate toward people who think like they do. And that is precisely the point: the SAT is not intended to measure overall intelligence, at least not anymore. On the contrary, it is designed to measure a particular kind of academic intelligence. Even if university professors must often encourage their students to abandon the kind of formulaic thinking they learned in high school, they nevertheless prize the kind of interpretation and logic skills that tend to lead to a high SAT score.

In no way do I mean to suggest that an SAT score indicates a student's ultimate capacity to acquire these skills (I don't think anyone would disagree that many extremely bright students fail to acquire them because of limited education opportunities or socioeconomic factors.) The SAT only serves to measure these particular skills at a certain moment in time.

It is of course possible to be absolutely brilliant out-of-the-box thinker and do horrendously on the SAT. Einstein, so the line of thinking goes, was a terrible student who probably would have flunked the math portion, and certainly a writer like Joyce would have gotten at best a 6 on the essay because of his unconventional style. But thinking like that misses the point: 99.9% of high school juniors and seniors are neither potential Einsteins nor Joyces. In fact, for the vast majority of them, writing a coherent -- if formulaic -- four or five paragraph essay with perfect grammar, clear transitions, examples that clearly support the thesis, varied but correct sentence structure, and an acknowledgment of potential counter-arguments is in and of itself a major achievement. Academic writing prizes clarity and sophistication of ideas and expression -- most professional academic writing is extraordinarily formulaic; it's the ideas that count, not necessarily the framework in which they are presented -- and an essay that includes those two elements will generally score well regardless of any other factors.

Critics of the SAT often invoke comparisons to the kind of high-school leaving tests required in other countries (British A-Levels; the French Baccalauréat; the German Abitur, etc.), tests that require a high level of composition and understanding of complex ideas, and on this point I agree completely -- ideally (idealistically?), the US would have a similar required exam, one more suited to gauging skills far more profoundly and holistically than the SAT can even begin to claim to do. The US, however, is unlikely to implement such an exam anytime soon.

That leaves us with the SAT. And based on my personal experiences, in the majority of cases (note that I say the majority, not all), SAT scores are in fact a generally accurate reflection of a student's mastery of the underlying skill sets being tested. In other words, students who repeatedly score in the low-to-mid 500's on Critical Reading generally do so not only because they fall into the traps set by the test-makers, but also because often they cannot identify the main point of the majority of passages or even figure out how to determine it, never mind sum it up in a few words. Tone questions are difficult because they do not know how to determine whether an author is agreeing or disagreeing with a particular concept, not just because option (B) was half right and half wrong. Asked to write a summary of the argument presented in a passage or formulate a cogent response to it, most would be equally unable to do so. And a low essay score does not result just because a student fails to plug in the right formula, but rather because the argument is incoherent and the writing ungrammatical and sloppy. In other words, the SAT isn't responsible for creating weaknesses -- it simply reflects them where they already exist. Unfortunately, it is often easier -- not to mention more fashionable -- for students to blame the test than to take a hard look at their actual skills and truly attempt to remedy any glaring weaknesses.

All this can lead to a very dangerous mindset that I've observed in a number of my students: The SAT is stupid and pointless and only tests how well you can take it. Therefore, the skills tested on the SAT are stupid and pointless and I shouldn't bother to learn them. And furthermore, since I don't naturally think the way the SAT wants me to think, learning to think that way would be an insult to my individuality.

Nothing could be further from the truth. While it may not be immediately obvious, the SAT tests fundamental skills of analysis and interpretation that are extraordinarily important. The ability to read a text closely and determine with the greatest precision possible what an author is actually attempting to say, as well as the rhetorical strategies (s)he uses to say it, is an extraordinarily powerful tool. This is not merely a question of analyzing "pointless" literary works, but rather one of viewing any text (advertisement, magazine article, etc.) as a deliberate construction designed to make a particular point by triggering a particular response in the reader or consumer.

So, to sum up: Is the SAT a perfect test? Absolutely not. Is it a great test? Not really. As a crude diagnostic of certain extremely important skills, however, it is remarkably accurate, and for that reason and in the absence of a more comprehensive alternative, it deserves to be taken seriously.

Getting from 650 to 750+ on CR: Suggestions for High Scorers

For students who are already in the 650-700 range on Critical Reading, attempts to boost their score into the exceptional (750+ range) can be an exercise in frustration. Since reading comprehension comes easily to them naturally, most have never taken the time to truly analyze their responses and instead rely on instinct, answers that "feel right," to get them through. In my experience, however, there are a couple of factors that typically separate relatively high scorers from exceptionally high scorers, and those factors have absolutely nothing to do with intelligence. The difference between a 700 and an 800 can be as little as five questions, and it's often the student's approach to those questions rather than their content that determines the ultimate score.

So if that describes you, here are my biggest pieces of advice:

1) Study vocabulary 

It's very easy to underestimate the impact that sentence completions can have on your score, but missing just two per section will drag you down by 50+ plus points. If you don't have trouble figuring out the logic of the sentences and simply need to learn words, put in the time and learn the words. You'll be grateful later.

2) Spend 15-30 minutes per day reading SAT-level material...

and look up and write it down every single word you don't know. Good sources are newspapers such as The New York Times op-ed page (it's online if you don't have a subscription), The Wall Street Journal, The Economist (actually a little harder than the SAT), and magazines such as Scientific American, National Geographic, or Humanities. You can also browse through the links on Arts & Letters Daily (www.aldaily.com), one of my favorite procrastination sites.

The truth is that most people who don't read challenging material on a regular basis will find it very difficult to score above 700. The SAT is usually set up to contain about seven level 5 questions (the hardest), which is not coincidentally the number that takes you from an 800 to a 700. Answering many of those questions correctly requires a "feel" for how certain words and phrases are typically used in works written for an educated adult audience -- uses that might be completely different from what an average sixteen year-old with minimal exposure to such works understands them to be. Remember: it's not about what you feel a word should mean; it's about what educated adult readers would pretty much unanimously understand it to mean.

3) Forget that the SAT is a multiple-choice test and just answer the questions

If your reading skills are solid enough for you to be scoring at 650+, they're solid enough for you to answer the questions on your own without consulting the answers first. Will this work for every question? Of course not. There are some that are impossible to narrow down at all without consulting the answer choices, but the vast majority do not fall into that category.

Read the question, look back at the passage, and, as much as possible, sum up in your own words what you think the answer is. Write it down in a couple of words, but only a couple. Any more will take too much time. It's not about being eloquent, it's just about reminding yourself what you need to look for. This step is crucial -- when you have to consider multiple ideas simultaneously, you're a lot less likely to overlook something important if you have everything in writing. Then look for the answer that comes closest to capturing the general idea you've described.

There's a big difference between weighing the pros and cons of each answer choice and actually looking for something. You're a lot less likely to get thrown off by tricky wording or false answers this way because you've already determined the general information that needs to be included in the correct answer.

Plus, if you can answer the question yourself, you won't get it wrong -- the chances of you coming up with a trick answer on your own are virtually zero.

Remember, however, that the SAT is asking you to think functionally rather than literally. So rather than say that a particular paragraph is describing an author's early life, for example, ask yourself what role or function it plays in the passage as whole: is it providing context for a work discussed later on in the passage? It is offering an explanation for a particular feature of the author's work? If you can train yourself to think this way, you'll be amazed at how quickly you start to spot answers.

4) Stop going on instinct and be absolutely systematic 

If you're scoring above a 750 in Math, there's a good chance you're actually solving problems step by step, and, only after you've arrived at the solution, consulting the answer choices rather than simply plugging in numbers and guessing. You need to approach CR questions the same way.

While this goes for the entire test, the area where this is most necessary is on Passage 1/Passage 2 questions. When asked how the author of Passage 1 would respond to a particular idea in Passage 2, for example, you need to deal with each element separately.

First, reiterate for yourself the main point and tone of Passage 1. You should have already written them down, so make sure to look back at them. You need to be certain of what the author of Passage 1 thinks about the topic before you can attempt to infer what (s)he would have to say about someone else's view of it.

Next, determine what function the given line in Passage 2 plays in the overall argument -- chances are it simply supports the main point. When you figure that out, write it down, too.

So now you ask yourself what the relationship is between the main points of the two passages, and once you determine it, surprise, write it down! Is one positive toward a topic while the other is negative? If so, you know that you can automatically eliminate any answer that is positive or that indicates agreement. Likewise, if both authors generally agree in their perception of the topic, you can automatically eliminate any answer that is negative or that indicates disagreement.

The bottom line is: at every step of the way, sum it up and write it down.

Can working like this be tedious? Absolutely. Does it require more upfront than simply looking at the answer choices? Of course. Is there a decent chance you'll get it right even if you just go on instinct? Sure, but there's also a pretty decent chance you'll get fooled sooner or later. The more work you do upfront, the more you reduce your margin of error.

If you've reasoned your way through every step, chances are you'll be able to go right to the answer without getting distracted by any traps. And that's what can get you from making one or two mistakes per section to making none. Take it from someone who knows: the first time I took the SAT, I crashed through the CR purely on instinct and scored a 710; the second time, I reasoned my way through all of the questions, wrote everything down, took my time, and scored an 800.