If it’s not true in the real world, it’s not true on the test

Established fact: a statement can be true in the real world but still be an incorrect answer on the SAT or ACT.

Pretty much every test-prep book you’ll ever read will tell you this. So, for example, you see an answer that says that Shakespeare is one of the greatest dramatists in the English language, you shouldn’t automatically assume it’s true because that statement might not actually be supported by the passage. I’m not about to disagree with that.

What no one talks about, however, is the fact that statements are NOT true in the real world are, for all intents and purposes, NOT correct answers to SAT questions.

So, for example, an answer choice that reads “scientists have made no progress in solving problems,” or “scientific and artistic achievement are fundamentally incompatible” is more or less guaranteed not to be correct. Those answers aren’t just extreme — they’re blatantly at odds with reality. And it’s fair to say that the SAT is biased in favor of reality.

Now, theoretically there could be an exception, but the chances of one occurring are pretty darn slim. (Maybe on a “which of the following would most undermine the assertion in lines 25-37?” question. But otherwise, it’s a very big stretch).

Yet I consistently see students — even high-scoring one — pick answers like these. When I point out that these answers have no basis in the real world, they’re surprised; it never even occurred to them to look at the test that way. I suspect that at some level they’ve been so brainwashed by the whole “the SAT is trying to trick you” and “the only thing that the SAT tests is how well you take the SAT” mentality that they don’t quite realize just what the test will and will not do. This is part of why I hate the whole “tricky” thing so much — it tends to make people jettison their common sense, and much of doing well on the SAT is simply about pushing common sense to its absolute extreme.

As a side note, that’s the other thing I keep telling my students: the test is set up so that you can figure things out, even if you don’t know 100% what you’re doing. Your job is to focus on what you do know and use that to get to what you don’t.

But back to the issue at hand — why couldn’t the test just be trying to trick you by making the correct answer some bizarre thing has nothing to do with reality?

Here’s why:

One of the things no one ever seems to mention about the SAT and ACT is that they are designed to mimic the kind of academic and journalist “conversation” that happens in the real, adult world beyond high school. You know, the sorts of things you’ll tend to encounter in college (if you bother to do your reading, that is). On the reading side, at least, it’s partly a test of how familiar you are with the sort of language and ideas you find in publications like, say, The New York Times. So if you know who Angela Merkel is and what her economic policies are doing to Greece, chances are you won’t get weirded out if a sentence completion requires you to know what “fiscal austerity” is.

Standardized-test reading might feel very fake, and in many ways it is very inauthentic, but given the unavoidable limits of the standardized-testing format, it actually does a pretty good job of doing what it’s intended to do. (Passage 1/Passage 2 is based on the same principle as NYT’s “Room for Debate” series — and interestingly, commenters often exhibit the same comprehension errors that many test-takers fall prey to, most often ascribing much more extreme positions to writers than those that they actually espouse.)

It’s important to keep that real-world framework in mind in terms of “reading the test,” and it’s something I now go way out of my way to remind my students about. So as you’re reading through those answers tomorrow, trying to figure out which ones you can truly eliminate, ask yourself whether they make sense… like, for real.

Does reading help raise Critical Reading scores?

On the surface, the answer to that question might seem pretty simple. If Critical Reading is a reading test, then wouldn’t the obvious way to raise one’s score be to read more? Well… maybe. But also maybe not. Like most thing involving the SAT, it depends where you’re starting from, what you know, and where you want to get to. And if you’re looking for a summer study plan, then you need to think about what you can realistically accomplish in the space of a few months.

If you’re not one of the “lucky” people who’s read so much since childhood that you can simply intuit the answers to Critical Reading questions, then spending your summer trying to slog your way through Dickens or Dostoevsky probably won’t miraculously allow you to acquire that skill — especially if you don’t actually enjoy reading five-hundred page nineteenth-century novels and will spend most of your time trying not to tune out while reading them. You might pick up some vocabulary, especially if you keep a list of unfamiliar words, look up every single one, and go out of your way to learn how they’re actually used, but if you’re not a strong reader in the first place, a Great Work or two won’t suddenly compensate for years of just reading things like Harry Potter or Twilight (or nothing at all). As a matter of fact, reading fiction will most likely have limited value in terms of helping you recognize and summarize arguments, understand rhetorical strategies, and make inferences in the precise way that the SAT requires.

A couple of months back, I stumbled across a paper in which Emory University English professor Mark Bauerlein discusses the difficulties that the Common Core’s emphasis on non-fiction pose for English teachers. Bauerlein makes the very valid point that English teachers are trained to teach literature, not “informational texts,” and that requiring them to shift their focus to non-fiction would not only require them to abandon their area of expertise but would essentially create a curriculum that would place a physics textbook on the same aesthetic footing as Hamlet.

I’m not entirely convinced by Bauerlein’s next claim, however, namely that students who are continuously exposed to a rigorous curriculum consisting primarily of challenging classic works of fiction do not really need to study non-fiction because they will be able to automatically transfer the comprehension skills they’ve developed over to non-fiction texts for tests like the SAT. As evidence, Bauerlein cites Massachusetts pubic schools, which do generally offer a traditional curriculum based on challenging works of fiction and whose students consistently obtain some of the highest reading scores in the country.

As a product of the Massachusetts public school system who studied a curriculum much like the one Bauerlein describes, and who went on to achieve top Verbal scores with minimal formal prep, I think Bauerlein is generally correct in stating that the comprehension skills developed through the study of complex classic work of fiction do carry over to non-fiction.

At the same time, however, there are important differences between the two genres, and it seems like an oversight for schools to focus on developing the former at the expense of the latter (especially since so much of college is based on non-fiction reading). The type of character/plot/theme-based analysis I did in school and the kind of structural/rhetorical/inferential reading required by the SAT required two very different approaches, and the fact that I literally understood what I was reading on the SAT did not make what it was demanding of me any less foreign. I intuited the gist of what it was trying to accomplish well enough to figure out what I needed to give it, but it would have been much, much easier if someone had sat me down with a “complex text” like, say, Samuel Huntington’s The Clash of Civilizations and directly taught me to analyze its arguments rhetorically and logically à la the SAT.

But I digress.

The point I’m trying to make is that unless you fall into the very small minority of people who have somehow automatically absorbed everything the SAT tests just by reading, the best way to improve your Critical Reading score is to practice reading critically — the extent to which you can do that outside the structured format of SAT practice tests depends on you. But if you are going to do some independent reading for the specific purpose of prepping for the SAT, here are some suggestions.

1) Focus on relatively short pieces of non-fiction. They don’t have to be as short as CR passages, but they should be short enough for you to practice looking at how they’re organized. That’s much easier to do in a three-page article than in a twenty-five page one.

I would strongly suggest that you go on Arts & Letters Daily and pick an article or a couple of articles to read every day; pretty much everything on there is written at or above the level of the SAT. The New York Times Opinionator is also great.

2) Look out for pieces that discuss some of its most common topics and themes: string theory, the effects of technology on the reading/writing and the humanities, animal cognition, the body/mind problem, immigrant/minority experience. (There are LOTS of articles that touch on these subjects on Arts & Letters Daily because these are hot topics in the real world.) After a while, you’ll start to get familiar with the conventional arguments surrounding these debates, which means you’ll have to waste a lot less energy just trying to figure out what they’re literally saying.

3) Look up every unfamiliar reference, not just vocabulary words — names, places, concepts. Never heard of de Tocqueville or Hegel or Stanislavsky? Go find out who they were and why people care about them. Critical Reading does not exist in a box; it’s designed to reflect the Common Core, and passages are deliberately drawn from a wide range of topics in the humanities, sciences, and social sciences. The more you know about the world, the easier it will be to literally comprehend readings about an incredibly wide range of topics (it’s much harder to appreciate a passage about an anthropologist if you don’t know what an anthropologist does.) It’ll also give you lots of fodder for the essay.

4) Treat everything you read like an SAT passage. Pay particular attention to the introduction and the conclusion when looking for the point, and see how quickly you can figure it out. Make sure you’re clear on when an author is expressing their own ideas vs. someone else’s ideas, and look at the words and phrases they use to indicate or suggest agreement vs. disagreement. Notice when an author is supporting their point with personal anecdotes vs. hard facts vs. broad generalizations, using extreme language (expressing “the strength of a conviction”), and using common words in alternate meanings.

Provided you understand what you’re reading and can accurately identify the elements discussed above, pending even thirty minutes a day reading this way will most likely help you go just as far — if not farther — toward increasing your Critical Reading score as simply sitting with a Princeton Review book and taking practice test after practice test. You’re also a lot more likely to learn something in the process.

Ask yourself what you’re missing

When I’m working through Critical Reading questions with a student, I regularly encounter the following scenario:

The student reads and understands the question without a problem.

The student goes back to the passage, re-reads and accurately summarizes the section in question, then formulates a general idea about what information the correct answer should contain.

The student looks at the answers but fails to see one that clearly fits.

The student’s eyes start to glaze over with panic as he stares at the page.

At this point, I usually interject nonchalantly, “So what are you going to do now?”

Student (looking sheepish): Uhhh… I don’t know?

Me: What happens when you work through the whole question carefully and then nothing seems to work?

Student: Is it (C)?

Me: Don’t guess. What do you do when you think you know what the right answer is going to say but none of the answers say it?

Student: No, wait, I think it might actually be (B).

Me: I said don’t guess. What’s the question you need to ask yourself when this happens?

Student: Uhh… I don’t remember

Me: I somehow feel that we’ve had this conversation already. Like, oh I don’t know, two or three or five hundred times. C’mon, this is probably going to happen when you take the test, and you need to know how to handle it.

Student stares blankly.

Me: What am I missing? What am I not seeing? That’s the question I always ask myself. If I’ve worked through the whole thing carefully but still don’t see the answer, that’s a sign that I’ve missed something. I’m either focusing on the wrong thing, or I’m just plain thinking in the wrong direction, and that means I need to go back and reconsider my original assumption. See how I’m turning it back on myself and taking responsibility for not knowing the answer? The people at ETS didn’t mess up; the answer is there, it’s just not something I’m not expecting. If I’m not getting it it means I’ve overlooked the necessary information.

Look, I did the exact same steps you did, and I wasn’t sure about the answer either. That happens to me too. But the difference is that I didn’t just decide to guess when I didn’t see the answer right away — I went back and tried to figure out where I went wrong. And if you seriously want to get every question right, you have to be willing to do that as well.


Perhaps unsurprisingly, I tend to have this conversation most frequently with students scoring 700+ — they’re so accustomed to spotting the answer immediately that they’re simply flummoxed when they don’t, and then go into all-out panicking/guessing mode. Needless to say, this is not a particularly helpful strategy for getting to 800.

So for the record, when you think you’ve figured everything out but actually haven’t, here’s what to do:

1) Assuming that you’ve correctly understood both the question and the passage, start by crossing out the answers that absolutely, clearly do not make sense. Make sure you cross out the whole answer, not just the letter; when your visual field isn’t being cluttered by extraneous information, it’s easier to focus on what’s left.

2) Now, carefully read the answers that remain. You might have simply misread something or overlooked a key word the first time around. If that’s the case, you’re done. But if not…

3) Go back to the passage and look for specific textual elements that usually indicate importance: if you have a colon or dashes or italics or strong language (e.g. “most”) or a major transition like “however” or “therefore,” the information around it is probably going to be key, and the correct answer will probably restate it in some form.

Make sure also that you have the necessary context for the lines in question; sometimes the right answer won’t make sense if you only consider the lines referenced. Remember that “function” or “purpose” questions regularly go either way: sometimes you need to focus on the lines given, and sometimes you need to focus on what comes before or after, and there’s absolutely no way to tell upfront which one it will be. If you haven’t read both places and can’t figure out the answer, chances are you’ve been focusing on the wrong place.

4) If you’re still stumped, start with the most specific (usually the longest) answers and pick a concrete aspect to check out. If an answer mentions that an author was criticized on “moral grounds” but the passage only indicates that she was criticized because her work was challenging aesthetically (i.e. it didn’t conform to traditional notions of beauty), you can eliminate that answer.

Remember that the correct answer might be phrased in much more neutral or general terms than the passage itself; if an answer accurately describes what’s going on in the passage but does so neutrally while the passage is fairly negative or positive, it’s probably the answer.

Remember also that you shouldn’t eliminate answers simply because you find them confusing; your understanding has no bearing on whether they’re wrong or right.

5) If you still truly have no idea, skip the question and come back to it if you have time. It’s not worth wasting five minutes on if there are other things you can answer more easily.

“If I can do it, anyone can do it,” and other favorite SAT fallacies

It’s relatively common knowledge that “extreme” answers (ones that contain words like “always” or “never”) tend to be incorrect, but what doesn’t often get discussed is just why those sorts of answers are often wrong and how that reason relates to the overall goal of what the SAT is trying to test.

As I wrote about recently, one of the great themes of Critical Reading is that there’s “always a however” — that is, arguments are not black-and-white, that they contain nuances. Simply saying “well, x is obviously always right, and everyone who disagrees is an idiot” is not a particularly effective mode of argumentation. That’s part of the reason Passage 1/Passage 2 exists.

Ever notice that when one author presents one side of an argument in black-and-white terms, the other author will come back arguing exactly the opposite side? That’s not a coincidence. The point that the test is trying to make is that stating that something is true very strongly does not actually make it true, nor does it negate that fact that there are plenty of arguments against it. If you want to make a solid argument, you have to seriously consider the opposite side and think about how it fits into your argument — that goes a degree beyond what the SAT tests, but the skills that P1/P2 tests are the basis for that ability.

Note, by the way, that considering degrees of nuance is exactly the OPPOSITE of what most people are taught when it comes to the SAT essay (and, might I add, in school). That’s not to say you can’t earn a very high score arguing in extreme, black-and-white terms (indeed, it’s usually much easier to do so in the space of 25 minutes), simply that training yourself to look at only one side of an argument is a stellar way to blunt your understanding of how to approach Critical Reading.

So that’s point one.

The second, closely related point is that is individual experiences cannot necessarily be generalized, and that different people will hold different positions shaped by their background, social circumstances, etc. The assumptions you hold about the world might not be true for the kid who sits next to you in Bio, or who lives down the street or on another continent — even though you can all be classified as members of the category “teenager” or perhaps “teenagers studying for the SAT. In other words, what is true about a particular person or type of person (teenager, artist, scientist, anthropologist) cannot automatically be extended to every single other person or type of person — and the failure to fully grasp that idea is one of the fallacies on which many wrong Critical Reading answers are based. This is why answers that are “too broad” or “too extreme” are frequently (but not always) incorrect. It’s also why so many SAT rhetorical strategy questions ask you to notice personal anecdotes; relying on your own experience as your sole support for a claim does not a particularly solid argument make.

One of the things that I’ve noticed while tutoring is that my students have tendency to conflate the specific and the general — so, for example, if a passage discusses a particular painter and i ask them what the passage is about, I’ll often get a response like “painters.” At some level, they don’t seem to completely grasp the importance of the distinction between the singular and the plural; I know that because they often roll their eyes when I ask them how many painters the passage actually talked about. “Ok, fine,” I can hear them thinking, “it was only one painter, but who cares? Isn’t that, like, basically the same thing?”

Actually no, it’s completely different.

What’s really interesting to me, however, is the way in which this type of faulty reasoning replicates itself in discussions about SAT prep — the irony is really quite impressive. On one hand, this isn’t at all surprising: if people have trouble with the SAT because their logical reasoning skills are lacking, that weakness is going to manifest itself equally outside the test. But just for grins, let’s look at some of the most common. Notice the extreme language common to all of them:

-If I can do it, anyone can do it

Umm… Let’s examine that claim. Maybe you studied for a couple of years and raised your score 500 points to a 2350 — it certainly happens — but presumably you didn’t have a serious decoding problem (that is, you could actually read the words on the page and weren’t constantly guessing what unfamiliar words meant), were at some point capable of perceiving the underlying patterns in the test, had a good enough memory to retain all the vocabulary you were learning, and already knew many of the elementary and middle-school level words that the SAT tests (e.g. permanent, compromise, tendency. And that’s just a handful of skills for Critical Reading.

Having worked with kids who had serious trouble with all of the above, I can state pretty confidently that those are not things that can be taken for granted for every student. If a student has trouble with all of them, it’s a pretty safe bet that a 2350 — or even a 2000 — is not a realistic goal. An 1850 is actually a pretty solid score to start with; it’s usually an indicator that someone primarily needs to learn to take the test and that the underlying skills are more or less ok. In that context, 500 is points is a lot, but given the steepness of the curves on Writing and Math, it’s a difference of not all that many questions.

But there is a whole huge category of kids whose basic skills cannot be taken for granted and for whom raising their score even 100 points is a massive struggle. Kids with high-achieving peers and tutors with high-achieving students tend not be aware of their existence and therefore fall prey to another common logical misstep: if I haven’t seen it, it doesn’t exist (the underlying assumption being that they’ve already been exposed to the full range of achievement levels, or that their particular experience is somehow representative of the high school population as a whole).

Having worked with kids at pretty much the full spectrum (mid-300s to mid-700s), I can say pretty securely — and perhaps coldly — that not every kid has what it takes to improve by hundreds of points. Sometimes the deficits are just too extreme.

-SAT prep classes/tutors are completely worthless because I did really well just studying on my own

I’ve worked with lots of very smart kids who did fabulously once they were actually taught just what the test was asking them to do but who might not have figured it out by themselves. It’s nice that you did really well on your own, but you can’t conclude anything other than that tutors and/or prep classes would have been completely worthless for you. (Besides, even if you did amazingly, you might have also learned something from a tutor — for example, plenty of people manage to do very well on CR without really understanding what it’s testing.)

All SAT prep classes/tutors are completely worthless because they just teach you all the same strategies you could read in the books

That has an element of truth when it comes to strategy-based prep (at least in terms of classes), but if you need work on the actual skills, you’re probably better off sitting with someone who can actually teach them to you.

-Since I wouldn’t have done well on the SAT without a tutor, everyone who doesn’t have one must be at a huge disadvantage. Therefore, the only thing the SAT tests is how much money you have to prep.

Again, there’s a kernel of truth — as everyone knows, there is a direct correlation between SAT scores and income. Are kids who grow up in seriously educationally deficient environments at a huge disadvantage? Of course. Are wealthy kids whose parents can afford thousands of dollars for private tutoring at an advantage? Of course. Is that fair? Of course not.

But those are generalizations, and correlation does not equal causation: there are kids who can’t afford tutoring but who can sit down independently with a prep books and figure out everything they need to know; likewise, there are kids who get tutored for a year or more and don’t increase their scores at all (or worse, see them go down). Both are outliers, but they do exist.

There are also kids whose parents can afford tutoring but who are perfectly happy to sit down on their own with a prep book. If some of them score very well, it’s usually in large part because they’ve grown up in an enriched environment and got all the skills they needed just from their parents and school.

The fact that a tutor helped one particular person do well can by no means be generalized into the assumption that every person needs a tutor to do well, or that it’s impossible to do well unless you come from a well-off background.

And my personal fave:

-The only thing the SAT tests is how well you take the SAT

Does anyone truly believe that a kid who doesn’t know what “permanent” means can be expected to perform at the same level as one who’s studied three languages, reads Dickens in her spare time, and holds a position in the National Classical League? Or that a kid who struggles to get B’s in Algebra II (like I did) is seriously competitive with a national Math Olympian? Somehow I don’t think so. Yes, everybody has their strengths and weaknesses, but at the extreme ends of ability, the differences are so gaping that they’re pretty much impossible ignore.

It’s also possible to be very smart and still be missing important skills, and it’s a lot easier to blame the test for your shortcomings rather than own up to the possibility that you might not know as much as you think.

The perils of overconfidence (or: you won’t always recognize the right answer when you see it)

The distance between a high CR score and a truly outstanding one rarely runs along a linear path. Unlike Math and Writing, which are essentially based on a number of fixed rules and formulas and which can therefore be improved by the mastery of discrete concepts, Critical Reading cannot necessarily be improved by memorizing a few more rhetorical terms or vocabulary words. On the contrary, for someone stuck in the high 600s/low 700s on CR, raising that score into the 750+ range frequently involves completely rethinking their approach.

Given two students with identical solid comprehension skills and 650-ish scores at the beginning of junior year, the one who is willing to try to understand exactly how the SAT is asking them to think and adapt to that requirement will see rapid and dramatic improvement (often 100+ points). The other one will flounder, maybe raising their score 30 of 50 points, but probably not much higher. Occasionally, their score won’t budge at all or will even drop. They’ll get stuck and get frustrated because they just know that they deserve that 750+ score, but the one thing they will absolutely not do is change their approach. And by change their approach I mean assume that their ability to recognize correct answers without thoroughly working through the questions is considerably weaker than they imagine it to be. In other words, they have to take a step back and assume that they know a lot less than they actually do.

Let me explain: one of the things I continually find fascinating is that people can spout on for extended periods of time about the supposed “trickiness” of the SAT, yet when it comes down to it, they won’t actually take concrete steps to prevent themselves from falling for “trick” answers (i.e. answers that contain mistakes someone who is rushing or can’t bothered to fully read the question would likely make).

The best way I know of to reduce the possibility of getting “tricked” is to actually attempt to answer the question before looking at the answers — or at least to determine the general idea that is probably contained in the right answer. Working this way, however, requires you to abandon the assumption that you’ll be able to spot the right answer when you see it, even if you’ve made no attempt to figure it out beforehand.

Now, in case you haven’t noticed, answers to SAT CR questions are deliberately worded in a confusing manner. Unless you really know what you’re looking for, things that aren’t necessarily the case may suddenly sound entirely plausible, and things that are true may sound utterly implausible. You need to approach the answers with that knowledge and consciously be on your guard before you even start to read them. But in order to do that, you need to be willing to admit a few things:

1) Your memory probably isn’t as good as you think it is

Just because you think you remember what the passage said doesn’t mean you actually remember what the passage said — at least not all the time. Even if you remember well enough almost all of the time, it only takes a handful of slips to get you down from 800 to 720. Throw in a missed vocab question or two on each section and bang, you’re back at 680. If you want to get around the memory issue, you need to write down every single step of your process. It doesn’t have to be neat or even legible to someone other than you, but it needs to be there for the times you don’t actually remember.

2) Your thought process probably isn’t as unique as you imagine it to be

The test-writers at ETS are not stupid, and they know exactly how the average eleventh grader thinks — questions and answers are tested out extensively before they show up on the real test, and the wrong answers are there because enough high-scoring students have chosen them enough times. Don’t assume you won’t do the same. I also say this because many of my students are astonished when I trace the precise reasoning that led them to the wrong answer — before they’ve told me anything about why they chose it. They were laboring under the illusion that their thought process was somehow distinct to them. It wasn’t.

3) Sometimes, there is no shortcut

That’s a little secret that most people in the test-prep industry would rather not admit. A lot of students who are accustomed to using common answer patterns (e.g. get rid of anything that’s too extreme) to get to around 650-700 are shocked to discover that this technique won’t get them any further and that they actually just have to understand pretty much everything. Sometimes spotting the “shortcut” also requires very advanced skills that even relatively high-scorers don’t possess. On CR, the ability to determine the function of a paragraph from a single transition in its first sentence is a highly effective shortcut, but it involves a level of sensitivity to phrasing that most sixteen year-olds — especially ones who don’t read non-stop — haven’t yet developed.

4) Getting a very top score is hard

There’s a reason that only about 300 people – out of 1.5 million – get perfect scores each year. If acing the test were just about learning the right “tricks,” there would be a lot more 2400s.

If you really want to get your score up to 750-800 range, you need to respect that the SAT is in fact difficult and that it is your job to conform to it, not the other way around. If you don’t understand why a particular answer is correct, stop before you jump to blame the test for not making it what you think it should be. It doesn’t matter that you take hard classes. It doesn’t matter that your AP English teacher thinks your essays are brilliant. There’s something in your process that went awry, and it’s your job to identify and fix it.

Reading this over, I realize that a lot of what I’ve written in this post may sound fairly harsh. But I also know from experience that overconfidence is one of the biggest problems that can hold you back from attaining the scores you’re capable of achieving. It’s hard — I’m not denying it — but if you can take a step back and start to admit that you might not know everything you think you do, you might just have a fighting chance at an 800.

Finding the point

The single most important strategy you can use to get through SAT Critical Reading is to find the main point of every passage you read. Can this be annoying? Of course. You just want to jump to the questions and get them over with. Unfortunately, if you work this way, there’s always a chance that you’ll get thrown off by a distractor answer, no matter how good you are and no matter how well you think you’ll recognize the right answer when you see it.

The SAT rewards those who work through problems — Critical Reading as well as Math — very, very carefully. Most Critical Reading questions ask about the relationship between various details and the author’s overall point, and if you don’t take 15 seconds and define that point explicitly for yourself, sooner or later you will look right past it when it appears.

But where to find it?

The answer, it turns out, is pretty simple. There are three places it’s likely to be:

1) Last sentence, first paragraph (the classic place for a thesis)

2) First sentence, second paragraph

3) The last sentence of the passage

In general, you should automatically underline the last sentence of every passage you read — don’t think, just underline. It’ll usually sum up the point in some fashion, and if you find yourself totally lost, it gives you something specific to look back at.

Furthermore, do not ever, ever fail to circle/underline/notate in some fashion the word “the point.” It shows up far more often than you’d think. If the author is telling you what the point is, it’s the point. Really.