to the top

"Difficulty Level 3" is consistently my worst level

superbutros
LSAT Novice
 
Posts: 2
Joined: Fri Apr 14, 2017 8:41 pm
Points: 2

I am routinely scoring lower on Difficulty Level 3 Logical Reasoning questions than any other Difficulty Level. On the October 2013 LSAT I scored:

88% on Difficulty Level 1 questions
82% on Difficulty Level 2 questions
46% on Difficulty Level 3 questions
100% on Difficulty Level 4 questions

Granted, level 4 is the smallest sample size, but I have repeated this pattern over many practice tests. Is this a common problem, and do you have any advice for breaking this pattern?
Luke Haqq
PowerScore Staff
PowerScore Staff
 
Posts: 169
Joined: Thu Apr 26, 2012 2:28 pm
Points: 129

Hi superbutros!

That's a great question, and it's good to see that you're giving such a keen eye to analyzing your strengths and weaknesses on the test.

Ultimately, I can only take a stab at suggesting what might be the cause of scoring worse on those types of questions. This PowerScore article, "How and Why Difficulty Varies" (see: https://www.powerscore.com/lsat/help/ls ... sition.cfm) discusses question difficulties for a handful of tests (Feb. 1996, 1997, 1999, 2000) that LSAC released with question difficulty levels and their explanations.

As you can see from the table in that article, it seems like the average question is around level 3. And that seems to make some sense to me--for example, the first two levels are less likely to help differentiate test takers (i.e., for purposes of plotting them along the 120 to 180 scale) because they're easier, and a level 4 question similarly might not help differentiate test takers if most people get them wrong. By contrast, level 3 seems to be a middle ground, a challenging type of question that enough people are getting right, and enough people getting wrong, that it makes for a good type of question.

So at least from the figures suggested in that post, perhaps one reason that you might be getting the most wrong in that category is because its the most common difficulty type. If that's right, then there could be a variety of explanations. For instance, you note that level 4 questions have a small sample size, so maybe the 100% isn't indicative--but if you're noticing a pattern that you're still getting proportionally more level 4 questions right versus level 3, maybe that's because when you spot hard questions, you focus your energy on diagraming/thinking them through to ensure you get them right? Whereas perhaps on a level 3 question you might spend less time on it? Again, in the end, I can only speculate, but my best guess would be that they make up the bulk of questions, they might not initially appear difficult, but it's that fact that might cause the LSAT test taker to spend less time on such questions.

As for the best way to try to break that type of pattern, one thing you can do is feel free to tackle the questions out of order. Some students find parallel-the-reasoning questions to be the most challenging, for example, and therefore choose to skip over them, only returning to them if they have time. I'm not sure how useful that will be in the case of level-3 questions, as I don't know the average level of parallel the reasoning questions, and it also seems to be a question type that it's probably hard to detect. Another thing you can try is to use the various drills in PowerScore's lessons, focusing on questions they've generally labelled as difficult, and also focusing on drills that are specifically difficult for you. They won't have these labeled as level-3 or -4, but it's at least a rough proxy for drilling yourself on those types of questions.

Finally, in working on those questions, keep in mind the importance of (1) diagramming out conditional reasoning, and (2) sorting answers into contenders and losers. If you have your work diagrammed, you have something solid to compare potential answer choices against--this helps reduce the risk that the test writers will trip you up with an answer choice that makes a slight change to the conditional reasoning. And sorting the answers into contenders and losers is an important tool to use for the sake of being time efficient, which seems especially important for level-3 questions, while also being able to make a reasoned choice in selecting an answer. If you can narrow down the answer choices to just two, hopefully that'll help you be able to spot a problem in one of them.

Hope that helps!
superbutros
LSAT Novice
 
Posts: 2
Joined: Fri Apr 14, 2017 8:41 pm
Points: 2

Luke,

Thanks for the detailed reply! I think your advice about taking questions out of order is the key. On-demand Lesson 12 also stressed skipping a question that is giving you trouble so you can come back to it later. After watching this lesson I took the December 2013 LSAT and broke the pattern: Level 1 100%, Level 2 100%, Level 3 76%, Level 4 60%. Sure, my Level 4 accuracy dropped, but I answered 7 more LR questions correctly than on the previous practice test, and got my best score ever.

Until recently, I was focused and determined on finishing every question in the section and making sure I had time left over for the "hard" questions at the end. I would never work out of order, which resulted in me panicking and hurrying around questions 16-20 as time wound down. After accepting that I can't just mow every question down in order, I actually ended up with more time left over than before. On LR section one I finished with something crazy like 7 minutes extra and was able to extensively review my answers. I only missed two in that entire section due to "silly" errors--as one of the on-demand courses put it.
Lawyered
LSAT Apprentice
 
Posts: 23
Joined: Tue Jun 13, 2017 12:27 pm
Points: 25

How are you able to tell the difficulty levels of the Qs?!
Jon Denning
PowerScore Staff
PowerScore Staff
 
Posts: 684
Joined: Mon Apr 11, 2011 4:36 pm
Points: 948

Hey Lawyered - that's a good question! Most test scoring systems out there will let you see your performance not just by question type but also by difficulty. For instance, when you score a test via our test scoring platform (either in your Online Student Center if you're a student of ours, or for free at our Self-Study Site: http://students.powerscore.com/self-study/index.cfm) you see how you did on questions grouped into Difficulty Levels 1-4, with 1 being the easiest (most people get them right, at 81-100%) and 4 being the hardest (fewest people answer correctly, at 0-40%).

So that's what superbutros is referring to :)
Jon Denning
PowerScore Test Preparation

Follow me on Twitter at https://twitter.com/jonmdenning
My LSAT Articles: http://blog.powerscore.com/lsat/author/jon-denning