simple question: is each question worth 1 point?

[Guest Gnome Chomsky]

It seems most logical. Scores range from 130-170; there are 40 questions; 170-130 is 40. I'm just curious because I've read that people took practice tests with score discrepancies.

[TakeruK]

From what I understood from the ETS material, if you are doing the computerized General GRE, then they are using CAT (Computerized Adaptive Testing or something). The difficulty (and thus "point value") of each question changes according to your performance. That is, if you are doing well, then your next (set of) questions will be harder. If you are doing poorly, then the next (set of) questions will be easier. This is why you are not allowed to skip any questions.

The idea of this algorithm is for the computer to determine at which level of difficulty are you able to consistently get correct answers. Your score is based on the difficulty of questions you answered as well as the number of correct answers. In the past, people have said things like "it's essential to get the first few questions right so that you can quickly move up the ladder" but I'm not sure that is quite true. But if it is, then the early questions are more important. When I wrote the Quant. part of the test, I figured it wouldn't be too tough since I was pretty confident in my ability. However, it was right down to the wire for me -- I had to guess the very last question with only 5 seconds remaining. I noticed that once I passed question 20 out of 28, all of the remaining questions were of the hardest difficulty level I saw on the practice exams. I might have even gotten one of the last questions wrong, but I still ended up with the maximum score (could be lucky and didn't mess up, or perhaps it was due to the CAT algorithm).

Edited November 14, 2012 by TakeruK

[alf10087]

Not at all. I'm almost sure I just missed 3 questions and had a 164 on quant. Also, in the Powerprep I missed 3 questions on both tests and had 166 on both.

The thing is that ETS levels the tests according to the difficulty of the sets that you're getting, so it basically depends on how you do on the first set.

[edgirl]

From what I understood from the ETS material, if you are doing the computerized General GRE, then they are using CAT (Computerized Adaptive Testing or something). The difficulty (and thus "point value") of each question changes according to your performance. That is, if you are doing well, then your next (set of) questions will be harder. If you are doing poorly, then the next (set of) questions will be easier. This is why you are not allowed to skip any questions.

The idea of this algorithm is for the computer to determine at which level of difficulty are you able to consistently get correct answers. Your score is based on the difficulty of questions you answered as well as the number of correct answers. In the past, people have said things like "it's essential to get the first few questions right so that you can quickly move up the ladder" but I'm not sure that is quite true. But if it is, then the early questions are more important.

This isn't accurate anymore. The old GRE worked like this, but the new one is adaptive by section, not by question -- you can absolutely skip questions, either altogether or coming back to them later. So it's no longer essential to get the first few questions right, but it is key to do as well as you can on the first section of each type. (Not that anyone tries to do poorly, of course.)

[midnight]

This isn't accurate anymore. The old GRE worked like this, but the new one is adaptive by section, not by question -- you can absolutely skip questions, either altogether or coming back to them later. So it's no longer essential to get the first few questions right, but it is key to do as well as you can on the first section of each type. (Not that anyone tries to do poorly, of course.)

Yeah, I just took the new GRE today and strongly prefer the new format. It's more intuitive, less stressful, and easier to tackle.

Quite honestly I didn't see much of a difference in difficulty between my first verbal section and my second, and I ended up scoring 164, two points higher than my verbal equivalent score on the old test. If anything, it seemed like my second verbal section was slightly easier--though I suppose it's possible I missed more on the first verbal section and the second section helped boost my score?

And I'm not a quant person at all, so my score was predictably mediocre, but I still improved A LOT over my first dismal GRE quant score. (I took the old format test last year just before the change and hated the question-by-question adaptive test.)

Also I really felt the time crunch by having only 30 minutes for the argument essay this time around. I forced myself to write as much as I could, but I have no idea how coherent I was since I had essentially no time to proofead or edit. I think they should do away with the AW section; it makes the test much too long and grueling and certainly isn't more valuable or accurate than a writing sample and SOP.

Edited November 15, 2012 by midnight streetlight

[Guest Gnome Chomsky]

Yeah, I just took the new GRE today and strongly prefer the new format. It's more intuitive, less stressful, and easier to tackle.

Quite honestly I didn't see much of a difference in difficulty between my first verbal section and my second, and I ended up scoring 164, two points higher than my verbal equivalent score on the old test. If anything, it seemed like my second verbal section was slightly easier--though I suppose it's possible I missed more on the first verbal section and the second section helped boost my score?

And I'm not a quant person at all, so my score was predictably mediocre, but I still improved A LOT over my first dismal GRE quant score. (I took the old format test last year just before the change and hated the question-by-question adaptive test.)

Also I really felt the time crunch by having only 30 minutes for the argument essay this time around. I forced myself to write as much as I could, but I have no idea how coherent I was since I had essentially no time to proofead or edit. I think they should do away with the AW section; it makes the test much too long and grueling and certainly isn't more valuable or accurate than a writing sample and SOP.

Is there a preferred length limit to the AW section? If not, it makes it hard to know whether they prefer quality or quantity. Does it force you to choose cramming as much info in as you can over writing a shorter, more cohesive piece?

[midnight]

Is there a preferred length limit to the AW section? If not, it makes it hard to know whether they prefer quality or quantity. Does it force you to choose cramming as much info in as you can over writing a shorter, more cohesive piece?

I believe that ETS strongly prefers quantity over quality. That's not to say your writing should be incoherent or illogical, but it's better to write, write, write five or six strong but basic paragraphs than it is to write three perfectly crafted academic ones.

The examples on the ETS site and in almost any practice book show you where you want to be. Don't go in and write the way you did for an A on a college term paper or written exam; the GRE is more like aiming for an A on a generic high school essay, which is not bad considering the time constraints.

Length, specific examples, and a few fancy vocabulary words go far in the AW portion.

Edited November 15, 2012 by midnight streetlight

[TakeruK]

This isn't accurate anymore. The old GRE worked like this, but the new one is adaptive by section, not by question -- you can absolutely skip questions, either altogether or coming back to them later. So it's no longer essential to get the first few questions right, but it is key to do as well as you can on the first section of each type. (Not that anyone tries to do poorly, of course.)

Thanks for the correction I'm glad they changed it, although hopefully I will never have to look at another GRE again

[TeaGirl]

Another note on the adaptive part of the test, it can vary depending what your random extra section is, verbal or quatitative. I did really badly on my first quantitative section (brain freeze) and was sure I was gonna have to take the GRE again. The second came super easy which only depressed me more, but I did my best and I'm sure I scored almost perfectly on it. My "extra" section turned out to be a quantitative section which I also did very well on. I ended up with a 168.

So all is not predictable at the actual exam.

[midnight]

Another note on the adaptive part of the test, it can vary depending what your random extra section is, verbal or quatitative. I did really badly on my first quantitative section (brain freeze) and was sure I was gonna have to take the GRE again. The second came super easy which only depressed me more, but I did my best and I'm sure I scored almost perfectly on it. My "extra" section turned out to be a quantitative section which I also did very well on. I ended up with a 168.

So all is not predictable at the actual exam.

Yeah, I'm 90% sure my second quant section was experimental (which was good because I didn't manage my time well during it and ended up guessing on about five questions and had to rush back to a couple of marked ones as well!).

But in the moment it's easy to forget about the experimental component, and you still need to perform well because you can't be sure....

I'll just say it was a relief to see my results since I met my very modest goal.

Edited November 16, 2012 by midnight streetlight