@wraith985-4026 said:

Well, if prior work isn't something you should have to be using anyway, and the better you get the less you need it, then it suggests that as you get better at the test the disadvantage to pen/highlighter (no prior work to crib off of) fades, while the disadvantage to redrawing (time taken to redraw) does not. So at the very end of the rainbow, I'd argue that pen/highlighter is unequivocally better even without jotting down boards (plus,when you're that fluent, you probably won't need to use them much anyway). But since most of us will not be there when we take the test, it's kind of a useless statement. What's best for everyone at any given moment varies with the circumstance. And the fact that you can mix and match makes it sort of a moot point regardless.

Practically speaking, I will always teach my students to work on their theory first. Theory forms your foundation and sets the floor; strategy is a multiplier. So regardless of pen/highlighter or redraw, I will always have my students treat each question as an isolated entity to start. We build their foundations first, and then multiply their baseline with efficiency tricks later. Perhaps that should be the major takeaway here.

I do not personally like local-then-global, you are correct, though I acknowledge once again that it can be a useful tool at times and I don't hesitate to have students utilize it when necessary (usually when they're utterly lost in a setup). Main issue there is that the LSAC constructs their games in such a way that big important stuff you need to see/understand tends to be tested first and utilized/iterated on later, so skipping over an early global question just because it's global risks missing out on an opportunity to discover a critical inference or pressure point. See, for example PT72 G4 (if you're not there yet, minor spoiler coming, but not a huge one) - the major core inference is asked about early on in a global question, and knowing it makes the rest of the set a lot easier. I'm not that adamant on this one though, because as stated above I approach questions as isolated and independent by default. Since I can do them in any order, it doesn't make sense for me to object too hard to a specific order, and it doesn't really harm you nearly as much to just do questions in a different order as it does to fundamentally try to skip over parts of the process entirely.

Just because one doesn't need to use prior work to solve a game doesn't mean that it's not a significant timesaver in certain situations. For example, rule equivalence questions. yeah, sure, you can do it without prior work. But I can't really accept the idea that using prior work to help eliminate answers isn't extremely helpful on this type of question. Not necessary. But helpful.

https://www.theregister.co.uk/2008/11/30/malcolm_gladwell_no/

Check out the world chess rankings for classical vs. blitz. The difference in performance is...overblown. Fabiano is #2 in "classical" and #8 in blitz. Almost every person in the top of classical is also somewhere in the top of blitz, it's just the ordering is different. The LSAT certainly is not measuring people as finely as the ELO system is, so the effects of highly speeded vs. less speeded are bound to be much more muted on the LSAT.

Just think about how people do on practice tests on timed vs. BR. It's almost inevitable that someone's BR score is higher than their timed score. Gladwell seems to think that there are a meaningful number of people who, though lower timed, would significantly outperform people with higher timed scores when it comes to their BR scores. This doesn't seem to play out in reality.

Finally, I love how Gladwell uses a single anecdotal example to make his point at the end. That's why I can respond using my own anecdotal experience as a tutor. Let's see some more robust statistical analysis please!

Second finally, I completely disagree with the Noodle advice and Gladwell's whole perspective on how he should be doing the test. If finishing every question requires an "uncomfortable" pace, then he doesn't have to finish everything. He just has to recognize when to move on from a question. There's absolutely nothing about the actual reading/understanding that should be uncomfortably rushed.

https://heleo.com/conversation-malcolm-gladwell-on-why-we-shouldnt-value-speed-over-power/13687/

He has no good responses to Adam Grant, who is skeptical of Gladwell's complaint.

I don't understand his argument - halfway into the podcast and all he says is "why do we value speed?" Well, why not? He hasn't said anything about why speed shouldn't be valued. Obviously we value getting the correct answer, but isn't expertise going to allow someone to get to the correct answer faster? So the speed is an indicator of mastery.

This is a really bad idea. Please do not raise this issue to LSAC or law schools. Think about it from the law school's perspective. There is a red flag in your record for having worked past the time allotted. That's not a dealbreaker. I'm sure they've dealt with plenty of people who have had that and if you explain what happened it can be managed. But if they see that a applicant with this red flag also notes that a proctor knocked a pencil out of his hand, it makes it seem like the student is still aggrieved by having been written up for going over the allotted time. Even if the school believes that the proctor did in fact knock the pencil out of your hand, the fact that the applicant feels like it's worth mentioning is very odd and concerning. Frankly, it seems extremely weird and would make me concerned about this person's behavior in school.

Thank you. Could you share your diagnostic breakdown in each section (if you remember)?

@leahbeuk911 said:

@anonclsstudent104 We may have to agree to disagree, but I do in fact disagree with a lot of your post here.

First, do you have the question reference from the example you gave of the LSAT's use of many? I would like to read the original. From what you summed up here, I don't think that that example disproves the fact that many is equivalent to some for the purposes of the LSAT. I emphasize that because, as I mentioned above, sometimes the LSAT uses words differently than we do in spoken/written standard English. I agree with you that as I am speaking with someone, I would never use "many" to mean only 1. However, I don't think that says anything about how the LSAT uses the term. The LSAT intentionally uses our own biases against us at times.

Whether or not "many" can equal 1, I absolutely disagree with you that many specifically means a large number. For instance, an example 7sage uses is: say there are 3 mice that are living in your house, uninvited. It would depend on your opinion of mice how you would describe that. Personally, I would say there are many mice in my house. Way too many, in fact. But is 3 a large number? No, it is not. It is a pretty small number, all things considered. What even is the definition of a "large number"? Over 50? Over 5000? We don't know. That is part of the reason that, again for the purposes of the LSAT, many is equivalent to some. It is an unknown number that is larger than 0. I am not 100% sure that, even on the LSAT, many can include 1. I feel like it wouldn't be used that way often, and would largely depend on context. I can definitely see that if we're discussing numbers like 1 out of 2, the LSAT might possibly use many. But since it is an ambiguous word, I think that it likely doesn't use it in that context frequently, if at all. They would be more likely to use "some" in that scenario.

A 'large' number is undefined. It's not meant to be a specific range. It's also subjective, to a point. So the mice example does not counter this definition of 'many'. If 'substantial' number is more acceptable as a definition, that's what it means.

PT1, Section 4, #21 is the problem I was referring to. It imo definitively proves that the LSAT does not consider "many" to include 1. In addition, the problem that OP was asking about - the lavender and stress problem in PT76 also definitively proves that the LSAT does not consider 'many' to include one. This is because if 'many' meant 'some', then one of the wrong answers would be necessary to the argument. We can only eliminate that wrong answer if 'many' does not include 1. Notice that there is no LSAT problem where the correct answer depends on interpreting 'many' as including 1. So I don't think the insistence that 'many' includes one is serving any purpose. PT1, Section 4, #21 proves that 'many' and 'some' have different meanings, since there are two answers that are indistinguishable except that one uses 'some' and the other uses 'many' and only one of those answers is correct.

Did you have much experience with writing analytical papers before law school? What was your undergrad major?

I would like to re-iterate that 'many' does not mean 'some'. The reason it is said to mean some is that in the context of drawing inferences from multiple quantified statements, it allows the same kind of inferences that "some" allows. In addition, people confuse "many" with "most", which is why it's easier to just say that "many" is the same as "some".

"Many" is subjective in that it means a 'large' number, but it cannot mean only 1, regardless of the context. Proportion has nothing to do with it - "many" does not imply a small or large proportion, and a large proportion does not automatically justify the use of "many". "Are many of your grandparents serial killers? It doesn't matter that 1 out of 4 of your grandparents are serial killers, it would not be linguistically correct to answer "yes" to that question.

Consider the following set of statements:

Some As are B.

All As are C.

We could conclude that some Bs are C. But we cannot conclude that many Bs are C. This idea has definitely been tested on the LSAT (there's a problem about whether a society in which "many" crimes are committed must have "many" laws - it doesn't, because many people could be violating only a single law.)

While we're here, let's not forget that there are a lot of other words that LSAT curricula says means "some" but actually does not. Every single words retains its own meaning - it's just that for the purpose of drawing inferences from multiple quantified statements, they should be treated as "some". Several doesn't mean 1. A few doesn't mean one. Often doesn't mean a single time. But you can think of them as "some" so that you understand that you can't draw "most"-type inferences from them. LSAT curricula just doesn't want to complicate things by having a "several" arrow or a "many" arrow or a "a few" arrow. Instead, they say to translate all of that to a "some" arrow. But this doesn't mean that those words literally mean exactly the same thing that "some" does.

going forward, remember that "many" is defined as large number. It is not defined as "at least one". It is also not defined as "at least two". We can conclude FROM "many" that there is more than one, but the word itself retains more meaning than merely "more than one" - it means a large number. And the negation of "many" is NOT none. It is "not many", which means "not a large number".

OP, the LSAT is not inconsistent in how it uses many. It's just the curricula that say "many" means "some" are wrong. It doesn't mean the exact same thing as "some". Your analysis of this problem is correct - if "many" did equal "some", then (A) would be necessary. But the negation of "many" is just "not many". And that would not destroy the logic of the argument, because the argument just has to assume that lavender reduces illness by reducing stress.

@leahbeuk911 said:

Does "Many" = "Some" or "More than one" ?

7sage says "many" = "some." Thus, "many" can include "one."

But other resources have said "many" means 2 or more. Thus, "many" would NOT include "one."

So does "many" include one or not?

I think this is important because for PT 76.4.12, (I think that) whether answer choice A is right or wrong depends on what definition of "many" you have.

IF we assume that "many" includes "one," and that the negation of "many" is "none," then the way LSAC uses "many" in answer choice A of PT 76.4.12 is INCONSISTENT. In this context, it assumes "many" is "two or more." This is HUGE, because if you assume "many" equals "one," it changes Answer choice A from a WRONG answer choice to a RIGHT answer choice.

So LSAC seems to be inconsistent with how it uses "many."

Any suggestions or advice appreciated. thank you.

Actually, "many" does NOT mean the same thing as "some". It's a serious deficiency in many curricula that they say "many" = "some". Many is subjective, but it does mean more than just one. The reason people say "many" = "some" is that for the purpose of drawing inferences from combining a statement that uses "many" with another statement that uses some other quantifier, "many" is TREATED AS "some". This is because "many" doesn't mean "most". In an effort to emphasize that "many" is TREATED AS "some", many curricula say that "many" MEANS "some". This results in the exact problem you identified in PT76 - if "many" meant "some" the negation of "many" = 0. But that's not true. The negation of "many" = "not many".

I would expect it to take at least 3 years of committed studying.

Bump.

@ramyavemuganti252 said:

@anonclsstudent104 said:

I do not write out conditional logic during LR questions. I think that saves time for me. However, I think not writing the whole thing out came due to practice. How I go about identifying quickly is to have a clear idea of the argument structure. What are the premises and what is the conclusion, what information is context etc. This helps enormously for SA questions because I feel like almost all SA questions are predictable. One needs to be careful about contrapositive answer choices.

Both NA and SA questions have a gap. With SA questions the answer choice closes this gap. For example, in this question, the gap is in between prime purchasers of software and the software being successful, so we have to find an answer choice that bridges it.

For NA questions, although sometimes the answers are predictable, sometimes it is hard to find the necessary assumption they are going for.

Necessary assumptions needn't quite close the gap but involve in finding one premise that if it doesn't hold true, wrecks the argument.

Say for example : Johnny goes to the park when it doesn't rain. Johnny went to the park.

Sufficient : it did not rain.

Necessary: Johnny was alive and/or the park wasn't destroyed in an apocalypse (because if Johnny was dead or if the park was non existent, there wouldn't be going to the park and hence it is necessary for the argument).

Hi @anonclsstudent104

Thanks for your response! It's funny, I was just going over NA questions when I decided to check to see if anyone had responded to this (thanks to the admin for linking the question's explanation as well! I'll give that a listen today!). I understand exactly what you're saying! And that's generally the feeling I have about SA questions as well! It's precisely the feeling I had for this question, in fact. I guess to ask my question more clearly:

I'm at the end of an LR section. The question is an SA one--fabulous. Based on my reading of the stimulus, I know what the conclusion and the support/premises are. I know where the gap is. It's between success and this group buying the software. I guess my problem is this: When I eliminated everything unrelated to these two elements, I was left with A and C. Ignoring the "most" in option A, the only way to distinguish between these two answers is to realize that they're reversals of each other. One has success as the sufficient condition, and one has it as the necessary (which as you suggest, you can realize without writing a thing down). My question is about the next step. Knowing which version of this conditional statement I need for my answer. Does that make more sense? Sorry if I wasn't clear before.

The weird explanation I found wasn't in great English, so it was hard for me to decipher and I'm still not sure if what I gleaned from it is what was intended. However, that person said that the sufficient condition in the conclusion (if you break the conclusion down into logic) should be the sufficient condition in your answer when you are in this sort of situation. If that's the case, then I can look at answers A and C, see that they both have the elements I need, check the conclusion, see that its sufficient condition is success, and test my remaining answer choices to for the same. All without writing any thing down. It's a seductive option, but I'm not sure it's logically sound, which is why I wanted to check here.

Yes, that is a reliable way to determine the precise connection you need. Think about why it makes sense:

Premise: A -> B

Conclusion A -> C

The missing link is B -> C. Notice that C is a new term on the necessary side of the conclusion - that means the assumption also needs to be put C on the necessary side. Why doesn't it work if we add C -> B? Because in that case, even though All As are B, and All Cs are B, that doesn't guarantee that every A is a C - some As might be the kind of B that isn't a C.

Premise: G -> D

Conclusion: X -> D

The missing link is X -> G. X is the new term and it's on the sufficient side, so the assumption must put X on the sufficient side. Why doesn't it work if we plugged in G -> X. Because in that case, even though All Gs are D and All Gs are X, there would be plenty of Xs that aren't Ds.

I got a 180 and diagram frequently. Probably ~2-3 problems per LR section. Usually one of the difficult parallel reasoning questions and one MBT or Sufficient.

How have you only gotten 100% on a game a single time??? Am I misunderstanding you? Are you referring to a games section or a single game? Can you do the first game in a section - usually a basic ordering game - perfectly taking 35 minutes??

The text-based lesson materials are actually quite good. The video instructors are not good, but it does make sense for students to read the textual curriculum.

Quick reaction:

Sal Khan is not great at explaining LSAT problems. Doesn't break down problems in a step by step fashion - sometimes goes direclty answer choices without analyzing stimulus well up front. Does not even talk about reversed logic in a diagramming problem...doesn't use conditional arrows in the correct way (doesn't use them at all really)

The other person is hard to listen to :(

Sal reads stimulus first. The other person reads stem first.

@marine4life6798246 said:

@anonclsstudent104 said:

When I took it last June it was a pretty forgiving curve

Really? I just took the June '17 PT and thought the curve was brutal. -5 gets you a 174...

@marine4life6798246 said:

instead, the curve is related to the actual difficulty of the test questions. The idea is that equal scaled scores from any two tests will represent equal ability.

Do you think that this holds up for the upper 170s curve? According to 7sage analytics PT81 has an average section difficulty of 2.5, the same as many other exams, but the curve is absolutely brutal with a -2 equivalent to a 178 and -5 equal to 174. PT77 has the same average difficulty at 2.5 but a -2 on that PT nets you a 180. PT76 actually has a lower average section difficulty at 2.25 but the curve is much more forgiving.

The curve at the upper 170s seems much more like guesswork than an exact science to me, just looking at analytics.

How is 7sage's difficulty rating determined? That could be unreliable if it doesn't distinguish difficulties enough.

So what does everyone think?