@ I honestly can't mentally separate LR from the 20s from the 30s or even 40s right now. But in general, the main thing I gained from practice in those earlier LR sections was improved habits. Doing lots of early LR questions and sections allowed me to make certain core LR strategies second nature. So while I am having to make some adjustments now that I'm taking newer tests---altering my expectations about question type frequency, about where I'll find the toughest questions, and even about how difficult many wrong answer choices will be---I'm able to do so with a strong foundation of good habits.

Tl;Dr: It was really helpful to get lots of practice.

I have found that the frequency of formal logic questions characteristic of the 20s is NOT found in the 70s! (Haven't done the 80s yet so can't speak to that, but I've done lots between 20s and 40s and now am in 70s).

I think you were right, and that PT7 S4 Q13 was indeed an R.A. question!

Conclusion: The fact that the percentage of homes with smoke detectors has gone from 30% to 45% over the last ten years has not made early detection of house fires more likely.

Support: Over half of domestic smoke detectors are inoperative.

Flaw: Takes for granted that just because half of domestic smoke detectors are inoperative, an increase in the percentage of homes with smoke detectors has contributed nothing to early house fire detection. (Fails to consider that the proportion of detectors that are faulty may have been as high or higher ten years ago--in which case a rise in smoke detector use would indeed seem to help.)

D, the correct answer, addresses the flaw exactly, stating it in its positive form: there was an increase in the detectors' defectiveness rate.

I do not think that this answer would be sufficient to guarantee the conclusion. Given only that this 50% defectiveness rate is up from 10 years ago, we can't be sure that going from 30% of homes with detectors to 45% of homes with detectors was a useless shift for fire detection. To guarantee that, you'd have to have to show not only that the rate of defectiveness has increased, but that it has increased enough to offset the gains to be had from an additional 15% of homes getting detectors. Seems like that would require something with numbers that my mind can't comprehend lol.

But I think you were right! D has to be necessary. Try the negation test: Given its opposite, "there has been no increase in the defectiveness rate of detectors," how could the current high rate of defectiveness ever show that increasing the proportion of homes with detectors is useless? That would be like saying, "Sure more people get vaccinated these days than 10 years ago, but the vaccine's success rate has plateaued over the past ten years. So there's clearly been no benefit to increasing the proportion of people who are vaccinated." No way!

Possibly more important than all this junk ^^ though (sorry, it's helpful for my own studying to write this stuff out) is that you can tell from the stem that it's an R.A. It does say "properly drawn," but what it's asking is NOT which answer, if plugged in, would "allow the conclusion to be properly drawn." Rather, it's asking which answer MUST be true, if the conclusion is EVER to "be properly drawn" from those premises. This is tricky question wording, but your original instinct was definitely right.

For weaken questions, it really helps me to transform them into RRE questions. I think I got this from something JY posted one time.

For example, to answer a weaken question with this stimulus, "Johnny loves pasta, so he must love all Italian food," you would (1) negate the conclusion (Johnny doesn't necessarily love Italian food), and (2) ask it to yourself as a "resolve the discrepancy" question: "How could it be that Johnny doesn't necessarily love all Italian food even though he loves pasta?" or "How could it be that Johnny's loving pasta doesn't prove/guarantee that he loves all Italian food?" The correct answer will be the one that helps to answer your question.

Any idea about the feasibility of them setting up “remote testing”?

I just realized I think another critical thing is whether or not it says “all other things being equal...” first. If it does, then definitely not necessarily a linear relationship. But if it doesn’t, I don’t see how it could be anything other than linear.

@ Yes, that's what I'm thinking!

I think a big issue here is the difficulty of putting together the conditional "true/false" with the matters of degree indicated by "the more" and "the less." If you say, "the more you know about history, the less you love history," there certainly is a positive if/then that you can take from it: if you know about it more, then you love it less. But as I read it, it's not only that relationship that you can take away; you can also get a bunch more relationships of degree: If you know about it a lot more, then you love it a lot less, right? I think that's how it'd work if you graphed it. (When I think about "the more bananas you eat -> the fewer oranges you eat," I imagine a huge increase of bananas eaten meaning a huge decrease in oranges eaten, whereas an much smaller increase in bananas eaten would mean a correspondingly small decrease in oranges. But so wouldn't that also translate to statements of degree here, in this qualitative matter?)

If so, then the initial steps would seem easy enough, as "a moderate amount of additional knowledge" would seem to be imply "a moderate amount less love" and perhaps even the same with "a bit more knowledge" and "a bit less love." But what happens when you get decreasing knowledge? Would that imply more love? I am really not sure. If you were to graph this as a linear relationship, and increasing knowledge meant decreasing love, then this would indeed be true--as you moved backwards on that graph, every bit less knowledge would mean a bit more love. If so, then I think what you'd be looking at here indeed not only a biconditional, but a zillion little biconditionals of degree--a little less knowledge means a little more love, a ton less knowledge means you are absolutely infatuated with it, etc.

I think I disagree with @'s argument. Sure you could conceivably imagine someone liking history less NOT because they study more but instead because they dislike their teacher, but does a statement such as "the more you know about history, the less you like history" allow for that possibility? Perhaps if it were just "an increase in knowledge is sufficient for a decrease in love of history," but I think that "the more... the less" is far more expansive than that because of its implied relationships of degree. To my mind, "the more... the less" would indicate that you simply cannot dislike history less without seeing a corresponding increase in knowledge. And that would indicate something like a billion little biconditionals...

But I'm afraid I might be totally wrong. Perhaps we need some #help?

Did the LR section seem more difficult then normal?

Yes, I thought so. Probably the hardest of the three for me, but I already see someone else thought it was the easiest, so it just really varies.

Also could have been a different test. There were definitely multiple tests yesterday, judging by the multitude of RC passage subjects people named.

@ said:

@ said:

Do you all think that there will be a bigger than usual curve to deal with the issues that people were experiencing? Some of the factors that came about due to proctoru, and just the newer testing environment could lower people's scores and result in a larger than usual curve right? Or do you think it will stay relatively similar due to it being a pretty average difficulty test?

I'm not holding my breath on that. That's not LSAC's style. I felt like they took it easy in the difficulty of the test (though the RC on the one I took was definitely spicier than usual) because of all the uncertainty, so they're not going to throw us any more bones than they already have. I may be wrong, but I guess we'll see.

Would tend to agree. Also a longshot, but my source of hope is this: maybe, the elevation of RC and LG at LR's expense will necessitate a slightly kinder curve. This is mere speculation, and I know I'm biased bc LR is my least bad section, but you do tend to see way more complaining about LG and RC on these boards, right? If people are generally better at LR than LG/LR, then reducing its weighting would have to produce a gentler curve if percentiles were to remain the same... right?

(Or not, idk. Just my personal hope haha.)

@ I don’t think there’s any question that there will be another Flex date to sign up for this summer! Haven’t heard any speculation at all that they would outright cancel July. Has anyone else?

I was on the webinar, and definitely saw an LSAC mod say in response to someone's question that we will have scratch paper.

That’s BS. The only one of those problems I also had was the proctor mic coming on at the start of my test, which was indeed pretty annoying. Losing 10 mins of LG? You deserve a refund at a minimum.

@ for the May 18/19 flex, we were originally supposed to sign up on April 23 but that got pushed to April 27. So I’d assume soon?

Ugh. That really, really sucks.

I also found the LSAT Trainer to be really helpful for RC!

@ said:

Do we think admissions will treat/ weigh the LSAT-flex scores differently? I’ve heard differing opinions.

I emailed Columbia Law Admissions to ask this and they said they will not treat it any differently!

What do y'all think about this (written courtesy of Flex-Stress and morning coffee):

(A) An unweighted three-section test would pose problems for the following types of testers: people (1) whose best section is LR, (2) whose worst section is LG, and (3) whose worst section is RC.

(B) My (spidey) sense is that LR is probably a lot of people's best section, seeing that people have been studying as if it would make up half the test.

-> INTERMEDIATE CONCLUSION: Weighting LR at 33% would disadvantage you, but it'd probably disadvantage a whole lot of other people too! In fact, maybe enough to make the curve a bit more generous...?

(C) Also, my (spidey) sense is that people's LG and RC prowess might sort of be inversely correlated? Relative, I mean, to their other strengths--I'd expect people with LG as their greatest strength to be disproportionately likely to have RC as their greatest weakness, and vice versa.

(D) If that's ^ true, then relatively few people's raw scores would stand to benefit from this unweighted format. Not only would they need to be relatively bad at LR, but they'd also have to be especially good at LG without being relatively bad at RC or vice versa. So their skill ranking would have to be LG/RC, LG/RC, LR. And for the sake of this argument, I'm [perhaps fallaciously] thinking that that's a #relatively rare breed.

-> FINAL CONCLUSION: If reducing LR's weight might in itself be enough to encourage a looser curve, AND if the proportion of people who are likely to really benefit from this kind of weighting is even lower because of the fabled LG/RC tradeoff, perhaps there's good reason to hope for a looser curve in the world of #flex.....?

Good enough for me haha. Cool to know that can work!

@ interesting! If you don't mind my asking, are you a -0/1 LGer yet? (I'm definitely not.) Just wondering if that strategy is compatible with gettin' 'em all right.

Thank you! I definitely think I don’t take advantage of the “switch” tool enough, so that’s great to have as an alternative to actual board splits. I suppose that’s not all that different from the calculus involved in writing an “always together” biconditional in the game board as a block and then not splitting further.

Also, @ I was surprised to see you say that new LGs less frequently invite multiple game boards than old LGs. Is that the consensus? If so, what range of tests would you count as “new”?

(Also I got premium sometime since the initial post so that’s why I wrote that I only have a free account haha)

No problem!!

I think the key lies in here:

Why is “Such songs” taken to reference “songs written for movie soundtracks” rather than referencing “those songs that movie producers (or whoever asks) asked writers to make for movie sound tracks,” because where the former implies they were written for movies, the latter does not.

I think we know that "such songs" has to be referring to songs written for movie soundtracks because of what comes next: "...their writers receive single up-front payments rather than continued revenues from radio airplay." That statement can't be about hit songs in general (which is what it'd have to mean, I think, if it were simply referring to those songs movie producers wanted, regardless of whether or not they got them) because we learned at the beginning of the stimulus that normally, hit songs are played thousands of times and the songwriter is paid for each play.

Rather, it has to be about songs that are written for movies. We're told that hit songwriters sometimes decline to write for movies because of something. The reason they don't accept movie contracts is because the way in which they're paid for such contracts differs from how they're normally paid for hit songs.

Thanks very much! It makes sense that you wouldn't want to create a blanket rule like that, but good to hear that it's probably usually the right call.

For what it's worth @, I had that same experience at first but then pretty quickly saw it normalize. Try experimenting with different font sizes, line spacing settings---I ended up settling on extra large-sized font for LR and LG, and medium + double-spaced for RC. Sounds crazy, I know, but I really felt it helped me accomplish the tasks required for each section... or maybe it was just all mental, Idk haha.

But good luck! Have faith that with familiarity will come normalization; you know how to read and you know the material, and the more you use a screen to do LSAT stuff, the less distracted you'll be by the switch.