Help on Inference Questions Forum

[lifeprincess13]

I consistently get many inference questions wrong; Kaplan categorizes the inference questions as Must Be True, Most Supported and Fill in the Blank Conclusion or Main Point questions. I tend to get the most inference questions wrong though. The only good advice I have gotten on them is not to focus on the assumptions and focus on the implications. Is that right? And does anyone have anymore solid advice to improving on the inference questions?

Thanks!

[CardozoLaw09]

Work within the stimulus and discern how the pieces in the stimulus interact with one another. There'll usually be connections you can make between the premises and conclusion; for these questions though it's best not to try and prephrase your answer before looking at the answer choices because the answer could be something you don't really expect. Just bear in mind the relationship between the elements of the stimulus and look for an answer choice that conceivably connects one piece to another.

[ptittle]

I second Cardozo's reply. Suppose someone says "My dog is cute because she's brown." The only way that holds is if "All brown dogs are cute." That's missing premise, the connection between "My dog is cute" and "My dog is brown". You have to figure out how the presented bits are presumed to fit together, and then fill in the gap.

[gobosox]

I consistently get many inference questions wrong; Kaplan categorizes the inference questions as Must Be True, Most Supported and Fill in the Blank Conclusion or Main Point questions. I tend to get the most inference questions wrong though. The only good advice I have gotten on them is not to focus on the assumptions and focus on the implications. Is that right? And does anyone have anymore solid advice to improving on the inference questions?

Thanks!

98% of the time, the answer can be inferred using logical equivalencies, which make them very very easy if you get conditional logic, since it's like doing a simple math problem. i.e. if somewhere in the prompt it says that humans are green and if you're not green you're blue, what are some logical conclusions we can draw? Yes, we can say that humans are blue, but a more likely LSAT answer is that if you're not blue you are not human. (the contrapositive). Or maybe they will nobody who is not blue is not green. (convoluted wording of contrapositive of one of the statements.) Sometimes they don't ask you to synthesize both.

If that confused you, I'd advise reading a lot about conditional logic and taking a look at Velocity's free videos on the subject. Hope that helps a bit!

[BlaqBella]

You need to dump Kaplan's material. All those categories for inference questions are ridiculous. Main Point questions are inference questions in their books? Wow. Please do yourself a favor and either use Manhattan LSAT or Powerscore books to better dissect inference questions.

[Mik Ekim]

Hi - this is Mike, co-author of the Manhattan LR book -

it seems that you've gotten a lot of info about inference q's, some of which is useful and accurate, some of which is not, and if I were in your shoes, the challenge, for me, would be in prioritizing all of this information and advice --

to that end, here are some big picture thoughts about inference q's -- i think if you hit your "reset button" and try to implement some of these strategies with a clean slate, it will make the questions feel much easier --

First of all, it's important to note that some LR q's are designed for you to be able to predict the answer (for example, ID the flaw q's) and some q's are designed for you to not be able to predict the answer (for example, inference q's).

That's a big difference in the design of a question, and should be factored into your process --

To give an analogy, imagine the difference between these two math q's --

1.
1 + 3 =
(A) 2 (B) 4 (C) 6

2.
Which of the following is divisible by 5?
(A) 9 (B) 10 (C) 12

For the first one, you should expect to come up with the answer, then find that answer within the answer choices. For the second q, you need to understand what the question is asking for, then evaluate each answer choice relative to criteria.

Coming back to the LSAT -- different types of inference questions, such as must be true, most supported, fill in the blank, etc, do have different types of characteristics (unique qualities that mostly show up in tendencies for right answers) -- however, it's important to note that for all of the various types of questions, you should not expect to predict the answer, but, rather, expect to check the five answer choices against the stimulus to see which one answer is most provable, much as you might for that second sample math question.

More specifically, the best way to do this is to focus on why eighty percent of the answer choices will not be justifiable based on the text -- if u are zoned in on why answers are wrong (as opposed to clever ways in which they could be right), and if you keep practicing looking for wrong answers, they will really start jumping out at you. The right answer will be the one that has the least amount of flaws.

For the toughest of these questions, right answers don't feel as "clean" as right answers, for, say, sufficient assumption questions might. This can be really stressful during the course of the exam. Wrong answers are much more black and white and absolute -- so --

to summarize and repeat --

Don't waste too much energy trying to come up with all of the various ways that you can think about the info in the stimulus and don't try to predict an answer - just try to understand the stimulus the best you can --

Focus your energy on looking for reasons why answers are not justifiable based on the text. Most wrong answers will have obvious tells that show you that they cannot be proven by the text.

Confirm the answer that remains.


Hope that helps. I hope you give the strategy a try, and if you do, let me know how it goes. Good luck.

[lifeprincess13]

Thank you for everyone's input. It helped a lot, especially Mike's. Reformatting my thinking to figure out which ones are wrong instead of right definitely helped me improve. Unfortunately, I am still getting too many wrong in the LR section. And the ones I am getting wrong are all over the place; there is no pattern or type that I get wrong most often. I literally get one question wrong in about half the types, with no pattern or frequency as to the types.

I am hoping I will continue to get better, but I definitely would not have improve without everyone's help. Thank you so much everyone.

Thanks Mike for taking the time to provide such a thought out answer and being dedicated to success, and not just business.

Thanks for all your good luck, I am sure I will need it!