Can/Will LSAC do anything about it?

hazara
Posts: 43
Joined: Thu Sep 05, 2013 6:42 pm

Can/Will LSAC do anything about it?

Postby hazara » Sun Feb 09, 2014 1:34 pm

So far there seems to be mixed emotions regarding the RC and LR but there is no doubt a unanimous agreement by students in Canada and the US regarding that 4th game. Lets say 90% of test takers miss most of their answers in that game. Can/Will LSAT do anything about it? I am also wondering how they picked that game, there must have been allot of people who go those questions right in a previous experimental section?...

User avatar
Attax
Posts: 3589
Joined: Fri Oct 04, 2013 10:59 am

Re: Can/Will LSAC do anything about it?

Postby Attax » Sun Feb 09, 2014 1:35 pm

hazara wrote:So far there seems to be mixed emotions regarding the RC and LR but there is no doubt a unanimous agreement by students in Canada and the US regarding that 4th game. Lets say 90% of test takers miss most of their answers in that game. Can/Will LSAT do anything about it? I am also wondering how they picked that game, there must have been allot of people who go those questions right in a previous experimental section?...


They may withdraw a question, but there is a curve for a reason. Sometimes it doesn't fit like they want, but that's the nature of standardized testing. Retake. Better luck next go.

User avatar
bombaysippin
Posts: 1977
Joined: Fri Mar 22, 2013 3:11 pm

Re: Can/Will LSAC do anything about it?

Postby bombaysippin » Sun Feb 09, 2014 1:43 pm

hazara wrote:So far there seems to be mixed emotions regarding the RC and LR but there is no doubt a unanimous agreement by students in Canada and the US regarding that 4th game. Lets say 90% of test takers miss most of their answers in that game. Can/Will LSAT do anything about it? I am also wondering how they picked that game, there must have been allot of people who go those questions right in a previous experimental section?...


Bolded seems a bit exaggerated.

User avatar
PattyCake
Posts: 440
Joined: Sun Oct 06, 2013 9:02 pm

Re: Can/Will LSAC do anything about it?

Postby PattyCake » Sun Feb 09, 2014 1:44 pm

Attax wrote:
hazara wrote:So far there seems to be mixed emotions regarding the RC and LR but there is no doubt a unanimous agreement by students in Canada and the US regarding that 4th game. Lets say 90% of test takers miss most of their answers in that game. Can/Will LSAT do anything about it? I am also wondering how they picked that game, there must have been allot of people who go those questions right in a previous experimental section?...


They may withdraw a question, but there is a curve for a reason. Sometimes it doesn't fit like they want, but that's the nature of standardized testing. Retake. Better luck next go.


And some people were fine with it. I likely blew a lot of that section, but if you diagrammed it well that particular game was pretty easy. It was just overwhelming at first glance, and a number of the questions were very obviously designed to trick you based on your diagram. I still agree that the section overall was much harder than average, but I doubt they'll eliminate a game or anything. This is what they do, they're very good at creating a test that will place the right number of people into each score band, or whatever they're called.

random number
Posts: 231
Joined: Tue Jan 07, 2014 3:26 pm

Re: Can/Will LSAC do anything about it?

Postby random number » Sun Feb 09, 2014 1:45 pm

It also has been pointed out in the waiters' thread that the LSAT needs to have a groups of markedly harder questions in order to make distinctions within the top group of scorers. A 170 is somewhere around the 97th/98th percentile, which means in order to distinguish between scorers in the top 3% and spread their scores over a 10 pt range, the test needs to include questions that few people will answer correctly.

When this happens on LR/RC, it is seen as no big deal because the hardest questions are usually far apart or distributed between sections. For obvious reasons this is not the case for LG because the difficult questions are grouped around a single game. It is also easy to mentally justify selecting an incorrect answer is RC/LR, whereas in LG you know at the time whether or not you found the correct response.

In short, I think that the test is exactly as it was supposed to be. I may very well be wrong, but LSAC knows what it is doing.

User avatar
PepperJack
Posts: 646
Joined: Sun Sep 29, 2013 1:23 pm

Re: Can/Will LSAC do anything about it?

Postby PepperJack » Sun Feb 09, 2014 1:53 pm

Off topic, but at one of my interviewers the bro told me his buddy wrote LSAT questions and therefore was unable to take the test and had to go to a school that didn't require the LSAT. He still did well in the end. I can't imagine how smart someone would have to be to actually write the test.

User avatar
zhenders
Posts: 741
Joined: Tue Dec 17, 2013 10:21 pm

Re: Can/Will LSAC do anything about it?

Postby zhenders » Sun Feb 09, 2014 1:57 pm

random number wrote:It also has been pointed out in the waiters' thread that the LSAT needs to have a groups of markedly harder questions in order to make distinctions within the top group of scorers. A 170 is somewhere around the 97th/98th percentile, which means in order to distinguish between scorers in the top 3% and spread their scores over a 10 pt range, the test needs to include questions that few people will answer correctly.

When this happens on LR/RC, it is seen as no big deal because the hardest questions are usually far apart or distributed between sections. For obvious reasons this is not the case for LG because the difficult questions are grouped around a single game. It is also easy to mentally justify selecting an incorrect answer is RC/LR, whereas in LG you know at the time whether or not you found the correct response.

In short, I think that the test is exactly as it was supposed to be. I may very well be wrong, but LSAC knows what it is doing.


This. I half-suspect, too, that LG will become increasingly more difficult over time; over the past 5 years, there's been an explosion --starting with Powerscore and leading us to 7Sage/Velocitiy/etc -- of methodologies that can lead to pretty consistent scoring for most people; understanding that the majority of people who take the LSAT don't use these, there's a good chance that a considerably larger proportion of people in the 95%+ range are going -0/-1 on LG and losing their points in other sections, rendering LG not as useful. An uptick in difficulty is to be expected to account for this.

Regardless, this is exactly what the curve is for -- so in asking "can/will LSAC do anything about it", the answer is, "yes -- they already have, by adjusting the curve to compensate for the difficulty."

User avatar
PattyCake
Posts: 440
Joined: Sun Oct 06, 2013 9:02 pm

Re: Can/Will LSAC do anything about it?

Postby PattyCake » Sun Feb 09, 2014 2:03 pm

PepperJack wrote:Off topic, but at one of my interviewers the bro told me his buddy wrote LSAT questions and therefore was unable to take the test and had to go to a school that didn't require the LSAT. He still did well in the end. I can't imagine how smart someone would have to be to actually write the test.


Huh. I wonder if some schools would make an exception and let you apply in that case. Schools routinely let a very small number of individuals in with numbers way below their usual range for a variety of reasons, and that seems like a pretty awesome excuse. How would the addendum go? "I am unfortunately unable to sit the LSAT exam, as I was employed for some time writing the questions for other, little-brained applicants."

random number
Posts: 231
Joined: Tue Jan 07, 2014 3:26 pm

Re: Can/Will LSAC do anything about it?

Postby random number » Sun Feb 09, 2014 2:06 pm

zhenders wrote:This. I half-suspect, too, that LG will become increasingly more difficult over time; over the past 5 years, there's been an explosion --starting with Powerscore and leading us to 7Sage/Velocitiy/etc -- of methodologies that can lead to pretty consistent scoring for most people; understanding that the majority of people who take the LSAT don't use these, there's a good chance that a considerably larger proportion of people in the 95%+ range are going -0/-1 on LG and losing their points in other sections, rendering LG not as useful. An uptick in difficulty is to be expected to account for this.

Regardless, this is exactly what the curve is for -- so in asking "can/will LSAC do anything about it", the answer is, "yes -- they already have, by adjusting the curve to compensate for the difficulty."


I definitely agree. LG can be quickly learned and eventually perfected, meaning that older, easier games are useless for distinguishing between the people who have used the study methods described. Obviously LSAC can't compensate for this by increasing the difficulty of the entire section (this would prevent non-studiers from getting any questions), so LSAC instead inserts a single very difficult game which will challenge the skills of experience testers while using the remaining 3 games to test the abilities of everyone else.

User avatar
PattyCake
Posts: 440
Joined: Sun Oct 06, 2013 9:02 pm

Re: Can/Will LSAC do anything about it?

Postby PattyCake » Sun Feb 09, 2014 2:21 pm

random number wrote:
zhenders wrote:This. I half-suspect, too, that LG will become increasingly more difficult over time; over the past 5 years, there's been an explosion --starting with Powerscore and leading us to 7Sage/Velocitiy/etc -- of methodologies that can lead to pretty consistent scoring for most people; understanding that the majority of people who take the LSAT don't use these, there's a good chance that a considerably larger proportion of people in the 95%+ range are going -0/-1 on LG and losing their points in other sections, rendering LG not as useful. An uptick in difficulty is to be expected to account for this.

Regardless, this is exactly what the curve is for -- so in asking "can/will LSAC do anything about it", the answer is, "yes -- they already have, by adjusting the curve to compensate for the difficulty."


I definitely agree. LG can be quickly learned and eventually perfected, meaning that older, easier games are useless for distinguishing between the people who have used the study methods described. Obviously LSAC can't compensate for this by increasing the difficulty of the entire section (this would prevent non-studiers from getting any questions), so LSAC instead inserts a single very difficult game which will challenge the skills of experience testers while using the remaining 3 games to test the abilities of everyone else.


I agree with the general idea, but again I think the whole section was harder than average. The first game was relatively straightforward but the next two were harder than usual. There are usually one easy, two medium, and one very difficult. This one seemed to have one easy/medium, two hard, and one that could be VERY hard if you missed a few key points in diagramming or got distracted by the circle structure (which barely changed anything if you didn't panic). I personally didn't find the hard one all that difficult, but that may have been dumb luck and the fact that I worked on circular games a lot in the last few weeks before the test.

User avatar
zhenders
Posts: 741
Joined: Tue Dec 17, 2013 10:21 pm

Re: Can/Will LSAC do anything about it?

Postby zhenders » Sun Feb 09, 2014 2:23 pm

RN, I hadn't considered it to that level of specificity, but I think you're completely right -- this being all speculation, still and all I think that makes perfect sense, and seems like the logical course for LSAC. I might go so far as to say, too, that the 60's through the 70's suggest that conclusion strongly; each section had either 1 five-star game and 3 one/two-star games, or 2 three/four-star games and a couple one/two-star games.

delusional
Posts: 1190
Joined: Thu Jul 15, 2010 7:57 pm

Re: Can/Will LSAC do anything about it?

Postby delusional » Sun Feb 09, 2014 2:27 pm

PepperJack wrote:Off topic, but at one of my interviewers the bro told me his buddy wrote LSAT questions and therefore was unable to take the test and had to go to a school that didn't require the LSAT. He still did well in the end. I can't imagine how smart someone would have to be to actually write the test.
Really? I would think that anyone who is in the 170 range would be able to write fairly good questions if they cared to, especially if they had a background in psychometrics and long time to get good at it as a career.

User avatar
zhenders
Posts: 741
Joined: Tue Dec 17, 2013 10:21 pm

Re: Can/Will LSAC do anything about it?

Postby zhenders » Sun Feb 09, 2014 2:30 pm

delusional wrote:
PepperJack wrote:Off topic, but at one of my interviewers the bro told me his buddy wrote LSAT questions and therefore was unable to take the test and had to go to a school that didn't require the LSAT. He still did well in the end. I can't imagine how smart someone would have to be to actually write the test.
Really? I would think that anyone who is in the 170 range would be able to write fairly good questions if they cared to, especially if they had a background in psychometrics and long time to get good at it as a career.


This doesn't sound like a dream job to anyone else?

Can someone pay me $160,000/year to work for BigLSAC? Kthx

delusional
Posts: 1190
Joined: Thu Jul 15, 2010 7:57 pm

Re: Can/Will LSAC do anything about it?

Postby delusional » Sun Feb 09, 2014 2:32 pm

zhenders wrote:
delusional wrote:
PepperJack wrote:Off topic, but at one of my interviewers the bro told me his buddy wrote LSAT questions and therefore was unable to take the test and had to go to a school that didn't require the LSAT. He still did well in the end. I can't imagine how smart someone would have to be to actually write the test.
Really? I would think that anyone who is in the 170 range would be able to write fairly good questions if they cared to, especially if they had a background in psychometrics and long time to get good at it as a career.


This doesn't sound like a dream job to anyone else?

Can someone pay me $160,000/year to work for BigLSAC? Kthx
I totally agree. Just dunno if they're hiring and I don't have a degree in psychometrics.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: 34iplaw, addie1412, cianchetta0, dontsaywhatyoumean, freekick, proteinshake, SweetTort and 12 guests