June 2010 Exam, Fourth AR Game Forum

[user101]

After hearing about peoples' various reactions to fourth game on the June 2010 exam, I feel that there are merits to both sides of argument being discussed. I can't speculate on how LSAC is actually going to respond, but I'm bothered with the proposed solution of dropping the entire fourth game of the analytical reasoning section.

People taking this exam have different strengths and weaknesses. Some people cruise through the logical reasoning sections, some people excel on the reading comprehension section, and some people shine on the analytical reasoning section. The test taker's score derives from his or her performance on these three section types, and their contributions to the overall raw score (for the June 2010 exam, with 100 questions) are as follows:

LR: 50 questions, 50.0%
RC: 27 questions, 27.0%
AR: 23 questions, 23.0%

It is worth pointing out that analytical reasoning currently receives the least amount of love; that is, it makes the smallest contribution to the overall raw score. I don't recall how many questions were in the fourth game, but let's assume that there were six questions for the sake of argument. Consider how the raw score would be impacted if the fourth game is removed from scoring (resulting in 94 scored questions):

LR: 50 questions, 53.2%
RC: 27 questions, 28.7%
AR: 17 questions, 18.1%

If you're one of those people that are strong in one section but weak in another, you typically rely on a solid performance in the former to compensate for any deficiencies in the latter. If the proposed solution is implemented, it presents rather disheartening implications for people whose strength lies in the analytical reasoning section.

I believe the fairest solution would be to have the fourth game's difficulty reflected in the score conversion chart, instead of merely reducing the amount of scored questions in the analytical reasoning section. I don't know too much about the equating process, but my lay understanding is that the intent is to strongly correlate raw scores with scaled scores (and thus provide scaled scores that are consistent and meaningful) while accounting for the variance in difficulty in each test.

For what it's worth, I do not believe that the fourth game was particularly difficult. However, if LSAC deems that the fourth game was atypical of most games' difficulty, I would definitely feel more comfortable with the solution that I've brought forth rather than the proposed solution of entirely dropping the fourth game from scoring.

[quishiclocus]

I don't think there's a chance in hell they're going to toss the fourth game. If lots of people missed on it, it might change the curve a smidge, in which case you'll be in better shape if you did okay there but didn't do so well elsewhere. The scale doesn't, so far as I'm aware, care what kind of questions. It just cares how many people got right in total.

[cardinalandgold]

No effing way they drop an entire game from scoring. Even dropping a single question is pretty rare.

[3|ink]

There's no way they'll toss it. Let's just accept our crappy scores and be done with it.

[FuManChusco]

There is a 0% chance they are dropping a game. There is also a 0% chance they are changing the curve. Curves are pre-determined. I see a -11 coming. This was the same reaction that we saw after the dinos. The section wasn't that bad people.

[RPK34]

There is a 0% chance they are dropping a game. There is also a 0% chance they are changing the curve. Curves are pre-determined. I see a -11 coming. This was the same reaction that we saw after the dinos. The section wasn't that bad people.

Game 4 was actually pretty damn easy. I had to read the rules and explanation twice to understand how to set it up, but after that, the rules were limiting enough that you could just power through it.

I don't think Dinos was that hard either, but that one had at least a lot of moving parts and a lot of variables to keep track of in your head. Game 4 was pretty straight forward; I didn't get the "big inference" some people keep mentioning, but nonetheless, I thought there were a couple inferences that could be made that made the game very easy.

[user101]

I don't think there's a chance in hell they're going to toss the fourth game. If lots of people missed on it, it might change the curve a smidge, in which case you'll be in better shape if you did okay there but didn't do so well elsewhere. The scale doesn't, so far as I'm aware, care what kind of questions. It just cares how many people got right in total.

I've added emphasis to your reply, to further highlight the implications that dropping the fourth game would have on the test takers' scaled scores. Dropping the fourth game would have an effect that essentially "dilutes" the contribution that AR provides... which is bad news for people that rely on analytical reasoning to compensate for weak performance in other sections.

[LSATNightmares]

Echoing other people's thoughts, there's no way they will drop an entire logic game. I wish they would, but it would be unprecedented. And too many people seemed to have done okay on it.