Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Second time through LR

Zachary_PZachary_P Member

Hi 7sagers... I've noticed that in my recent PT's I have finished the LR sections with approximately 5 - 7 minutes left at which time I will go back to the questions I circled because I was not completely confident with my answer. When I go back through the section to these questions, I get worried about time and tend to spend no more than 45 seconds on them before usually confirming my first answer and going on to the next circled one. I am missing an average of 5 questions per LR section and mostly on these circled (curve breaker) questions.

How do you approach your second time through a section? Would it be better for me to hone in on 2 or 3 circled questions and absolutely get them right at the expense of not double checking all of the circled questions? Any light you can shed would be greatly appreciated.

Comments

  • inactiveinactive Alum Member
    12637 karma

    Bumping so more people can see this!

  • extramediumextramedium Alum Member
    419 karma

    @Zachary_P said:
    How do you approach your second time through a section? Would it be better for me to hone in on 2 or 3 circled questions and absolutely get them right at the expense of not double checking all of the circled questions? Any light you can shed would be greatly appreciated.

    This is basically what I do. I think that's what they recommend on the webinar in fact. Going back to the easier circled questions first and confirming your answer, and then using whatever spare time left on the more difficult circled questions.

    I guess it all comes back to using your time wisely and not sinking an inordinate amount of time into questions that you will most likely get wrong. Better to use your time confirming right answers on questions you know you can get right than to sink time into a question you don't understand.

  • JustDoItJustDoIt Alum Member
    3112 karma

    I skip pretty liberally. So I skip about 4 questions before the last page. I usually get there around 20 mins. Then when I get to the last page then I go back over all my skips before finishing the section. Sometimes a question will wreck me so I'll go to the last page before I do that question. I find this works because I feel I'm maximizing my points. The last few questions are usually the hardest so I try to maximize my points before then. I usually head to the last page at around the 26 minute mark. I keep my times open though because I don't wanna get stressed.

    Anyways this is what works for me. Felt mad awkward at first but now it's a habit and it works so I'll stick with it haha

  • Cant Get RightCant Get Right Yearly + Live Member Sage 🍌 7Sage Tutor
    edited April 2017 27900 karma

    This is going to get way out of hand, lol, so my apologies up front.
    The short version is Game Theory. If that doesn't sound fun, jump to the end!

    So first, I'll explain my general concept, then I'll show how I apply it to this issue specifically.

    I don't think of questions as right or wrong because that is not information I have at my disposal when it really matters. Right and wrong is for the graders. It's not information I can use, and I want information that I can do something with.

    So instead, I think of them in terms of probability.

    If I'm 100% confident in my answer choice, I give myself full credit: 1.0 points.
    If I'm 90% confident in my answer choice, I give myself 90% credit: 0.9 points.
    If I'm taking a blind guess, I've got a one in five chance and allow: 0.2 points.

    At the end, we add it all up, divide by the number of questions, and hope to be as close to 1.0 as possible.

    Assuming we are accurately assessing our confidence (which is a skill in and of itself) we should expect:
    at 0.80 average: 20.00 questions right on a 25 question section.
    at 0.90 average: 22.50 questions right " "
    at 0.95 average: 23.75 questions right " "
    etc.

    So, this is really abstract and there is tons of room for plenty of deviation, but it gives us something that we can functionally apply under time when we can not know which questions we have answered right or wrong.

    We don't get these points for free though. We've got to purchase them. The currency is time and we have 2,100 seconds. The trick is to always be getting the best deal possible in this exchange.

    So the first time through, we want to do a rough once over and get all the best deals. If there is a 0.98 for 30 seconds available, we want to grab that up. This is what we call low-hanging fruit. On the other side of the spectrum is something like a 0.35 in 210 seconds. This is a terrible exchange and we must skip this question--the faster the better. Learning to recognize these questions quickly is a vital skill for top scorers.

    So now to address the question of the second pass, we must take stock of what remains to us after the first. I normally have a few 50/50s, a couple questions I skipped all together, and lots of high confidence answers that I blew through without really confirming or reading all the answers. At this point, my sections usually look something like this:

    0.20 x2
    0.50 x3
    0.90 x10
    0.95 x10
    Average 0.816 for 20.4 correct answers.

    So where do I go from here?

    Well, I need to assess my best values. Where will I pick up the most for my time? The 0.2s and 0.5s are conspicuous starts. The 0.9s are really quick, true, but the return is very minimal. I never go above 0.95, so even if I work all 10 of those, I'm only getting a max cumulative return of 0.5. Not a great investment.

    For me, I typically get my 50/50s pretty quickly once I return to them. Whatever distinction I was missing, I almost always see it once I return. With the 0.2s, I'm going to have to put in a lot of time but there's a greater potential return. What to do?

    Well, I'm great on my 50/50s, so let's say I get two with high confidence and another with fairly high confidence once I return.
    Net gain: 1.25 in 2:15.

    The 0.2s are going to be tougher and take more time. Maybe I will normally get one of them to high confidence but the other one I think I have but just really don't know for sure.
    Net gain: 1.25 in 3:30.

    Obviously, I should go with my 0.5s first. I get the same return for less time. A clear bargain.
    That will give me:

    0.20 x2
    0.90 x11
    0.95 x12
    Average 0.868 for 21.7 correct answers

    Now I want to knock out those 0.2s.

    0.70 x1
    0.90 x11
    0.95 x13
    Average 0.918 for 22.95 points.

    Now I'll confirm my 0.9s until I run out of time. Maybe I go back to my 0.7, but let's say in this scenario that I don't make any more progress on it. For the 0.9s, they should be really quick, so maybe I get to 8 of them before time runs out:

    0.70 x1
    0.90 x3
    0.95 x21
    Final Average 0.934 for 23.35 correct answers.

    In reality, my LR performance tends to average out to about -1.5. So this is spot on to how my real sections play out.

    Skip to here if you don't want to fool with all the math

    So, of course, I'm not really doing all this is my head while I'm taking an LR section. Ain't nobody got time for that. So what does all this nonsense look like in practice?

    As I go through my first pass, I mark questions differently based on my confidence levels. You definitely don't want to mark a 0.9 the same way you mark a 0.2. That's crazy talk. So:

    If I'm 90% confident I've got the right answer, I slash through the question number.
    If I'm 50% confident I've got the right answer, I circle the question number.
    If I'm 20% confident I've got the right answer, I draw a rather dramatic triple circle.

    At the end of the section, this allows me to very quickly take stock of what's left, determine exactly where my best values are, and order my second pass for maximum effectiveness.

  • TheoryandPracticeTheoryandPractice Alum Member
    edited April 2017 1008 karma

    @Zachary_P how many questions do you end up circling?
    Since you have so much time left (5-7 min), I wonder if you are not spending enough time in reading and really UNDERSTANDING the stimulus.

    What I mean by this is
    1) understand literally what the text is saying and be able to summarize ACCURATELY
    2) understand what the author is doing in each sentence (what's the author's purpose for writing each sentence in the way he did?) -
    3) how each sentence relates to another sentence and thinking about what we DO know from the stimulus and what we DON'T know (the parameters of valid inferences)

    Maybe I have a really high standard for "understanding," but I don't move onto answer choices unless I understand all 3 above. I usually do not have much time left to go back and review the circled Qs, but I rarely have any circled Qs... (Maybe you are just a fast reader and I'm jsut slower than you)

    Just my thought!

Sign In or Register to comment.