Parallel Questions

User avatar
Anaconda
Posts: 610
Joined: Sun Jun 06, 2010 3:51 pm

Parallel Questions

Postby Anaconda » Sat Jul 31, 2010 2:35 pm

The LRB says to stay away from assigning variables to parts of the stimulus since it will only lead to further confusion since you're making abstract comparisons even more abstract. Does anyone else think this is a totally ridiculous assertion? If anything converting the stimulus into a variable relationship turns abstract words into a simple formula, at least for me.

User avatar
yzero1
Posts: 185
Joined: Thu May 27, 2010 4:33 pm

Re: Parallel Questions

Postby yzero1 » Sat Jul 31, 2010 3:11 pm

I got the impression that the LRB was talking about the assignment of a set of uniform variables to represent both the stimulus and the answers. Like assigning X, Y, and Z to depict the stimulus and then writing out answers using the same variables X, Y and Z. This is unhelpful and confusing because your mind has to do an extra step to convert argument elements into the uniform variables. However, if you use CUSTOMIZED acronyms to symbolize the relationships in the stimulus and answers (depending on the specific words used), I think it helps tremendously because it simplifies the logical relationships among the elements.

User avatar
Anaconda
Posts: 610
Joined: Sun Jun 06, 2010 3:51 pm

Re: Parallel Questions

Postby Anaconda » Sat Jul 31, 2010 3:26 pm

yzero1 wrote:I got the impression that the LRB was talking about the assignment of a set of uniform variables to represent both the stimulus and the answers. Like assigning X, Y, and Z to depict the stimulus and then writing out answers using the same variables X, Y and Z. This is unhelpful and confusing because your mind has to do an extra step to convert argument elements into the uniform variables. However, if you use CUSTOMIZED acronyms to symbolize the relationships in the stimulus and answers (depending on the specific words used), I think it helps tremendously because it simplifies the logical relationships among the elements.


I totally disagree. By using standardized variables, you can simply plug in parts from the answer choices into the structure of the stimulus. Isn't the whole point to understand the structure and relationships? When you use customized variables, how can you easily compare between the stimulus variables and answer choice variables?

Here's my customized example of a flaw parallel question:

Stimulus: All bears are wild animals. Some wild animals are lizards. Therefore, some bears are lizards.
A -(all)-> B, B -(some)-> C : A -(some)-> C

example of one incorrect answer choice, and one correct answer choice:

Wrong: All smartphones are cellphones. Some smartphones are PDAs. Therefore, some cellphones must be PDAs.
A -(all)-> B, A -(some)-> C : B -(some)-> C

Right: All colleges have law schools. Some law schools are ranked in the top 50. Therefore some colleges are ranked in the top 50.
A -(all)-> B, B -(some)-> C : A -(some)-> C


By making a simple diagram, you can spot why the wrong answer choice is wrong, and the right one is right.

When you have a 6 sentence stimulus, how is one supposed to keep track of the structure of the argument without making a diagram that can be applied to all the answer choices?

How would diagramming the stimulus as: B -> WA, WA -> L : B -> L instead of using standardized variables make things any easier?

User avatar
yzero1
Posts: 185
Joined: Thu May 27, 2010 4:33 pm

Re: Parallel Questions

Postby yzero1 » Sat Jul 31, 2010 4:54 pm

Anaconda wrote:
yzero1 wrote:I got the impression that the LRB was talking about the assignment of a set of uniform variables to represent both the stimulus and the answers. Like assigning X, Y, and Z to depict the stimulus and then writing out answers using the same variables X, Y and Z. This is unhelpful and confusing because your mind has to do an extra step to convert argument elements into the uniform variables. However, if you use CUSTOMIZED acronyms to symbolize the relationships in the stimulus and answers (depending on the specific words used), I think it helps tremendously because it simplifies the logical relationships among the elements.


I totally disagree. By using standardized variables, you can simply plug in parts from the answer choices into the structure of the stimulus. Isn't the whole point to understand the structure and relationships? When you use customized variables, how can you easily compare between the stimulus variables and answer choice variables?

Here's my customized example of a flaw parallel question:

Stimulus: All bears are wild animals. Some wild animals are lizards. Therefore, some bears are lizards.
A -(all)-> B, B -(some)-> C : A -(some)-> C

example of one incorrect answer choice, and one correct answer choice:

Wrong: All smartphones are cellphones. Some smartphones are PDAs. Therefore, some cellphones must be PDAs.
A -(all)-> B, A -(some)-> C : B -(some)-> C

Right: All colleges have law schools. Some law schools are ranked in the top 50. Therefore some colleges are ranked in the top 50.
A -(all)-> B, B -(some)-> C : A -(some)-> C


By making a simple diagram, you can spot why the wrong answer choice is wrong, and the right one is right.

When you have a 6 sentence stimulus, how is one supposed to keep track of the structure of the argument without making a diagram that can be applied to all the answer choices?

How would diagramming the stimulus as: B -> WA, WA -> L : B -> L instead of using standardized variables make things any easier?


You can understand the structure and the relationships while still using acronyms. The issue with using standardized variables is that you have to convert elements and arrange them in the correct order. Remember, elements in the answer choices are often given in a different order than in the stimulus (done to make questions more difficult). In your example, you have to convert bears to A, wild animals to B, etc., and do it for each answer choice. What if you mix up a variable in your conversion, perhaps because a certain answer choice gave you the elements in a different order? It's much easier just to not convert and use an acronym or something to symbolize the elements. For ex., using your bears/wild animals stimulus, we can come up with this symbolization:

B->WA<-(some)->L ... B<some>L

The logical relationships are represented concisely and you avoid confusion with converting to standardized variables. You can still compare answers with maximum speed with the acronyms because it's the set of ARROWS that are most important since they represent the relationships.

Additionally, if you use standardized variables, there's an extra disconnect when you're looking through the diagrams because you likely won't remember what B or C or A represents. If, on the other hand, you use acronyms, it's easier because you know exactly what part of the stimulus/answer choice your acronym refers to without having to re-examine your conversions.

Lastly, if you misdiagram something with standardized variables, fixing the relationship will be a lot more difficult than if the variables were acronyms that represented elements.

The key to the LSAT is to be able to diagram effectively and efficiently at the same time. Acronyms are, in my opinion, the best way to represent elements and using them has helped me tremendously; I have about a 95%+ accuracy rate with parallel reasoning questions.

User avatar
Anaconda
Posts: 610
Joined: Sun Jun 06, 2010 3:51 pm

Re: Parallel Questions

Postby Anaconda » Sat Jul 31, 2010 9:09 pm

Anyone else wanna chime in?

User avatar
3|ink
Posts: 7331
Joined: Wed Dec 16, 2009 5:23 pm

Re: Parallel Questions

Postby 3|ink » Sat Jul 31, 2010 9:22 pm

Anaconda wrote:The LRB says to stay away from assigning variables to parts of the stimulus since it will only lead to further confusion since you're making abstract comparisons even more abstract. Does anyone else think this is a totally ridiculous assertion? If anything converting the stimulus into a variable relationship turns abstract words into a simple formula, at least for me.


I'd have to agree with the LRB, but only because I never map out arguments in LR. I personally believe that refraining from mapping has increased my score. When you map out a stimulus, you often take it premise by premise. It's harder to take in the complete argument that way.

If mapping is your thing, disregard the LRB on this point. However, I think you'll find that parallel questions are much easier when you focus on seeing the argument as a whole than when you try and map each individual component.

If you have to map out the stimulus, chances are that you'll have to do the same for each answer choice. I think it's better to train yourself to identify the argument structure or flaw when reading an argument without having to map.

User avatar
AverageTutoring
Posts: 298
Joined: Tue Jul 27, 2010 10:18 pm

Re: Parallel Questions

Postby AverageTutoring » Sat Jul 31, 2010 10:30 pm

Anaconda wrote:The LRB says to stay away from assigning variables to parts of the stimulus since it will only lead to further confusion since you're making abstract comparisons even more abstract. Does anyone else think this is a totally ridiculous assertion? If anything converting the stimulus into a variable relationship turns abstract words into a simple formula, at least for me.


The LRB provides an inadequate explanation for Parallel Reasoning questions in my opinion. If representing a super complex word such as BHAGJA"AAKKAAAHADBHA"LSALSJAHSJ as X helps you draw connections, then do it. In my experience though this is only necessary when the stimulus is conducive to diagramming. In most cases however, this is not possible and this method is not helpful.

PR questions can almost exclusively be answered by two types of logic:

1: Similar conditional statements
2: Similar function

In the case of 1, diagram the stimulus and you're good to go. In the case of 2, you need to understand how the argument works and what is happening with respect to the conclusion. After you get this down PR should be no problem




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: No registered users and 3 guests