How would GRE score readers react to an analytical writing essay that's 20,000 words long?

[InquilineKea]

I think both of my responses turned out to be that long (20 minutes at 100 WPM => 20,000 words).

Obviously, it's not the length that matters - it's the arguments that do. But I very thoroughly addressed and overanalyzed all the necessary things. My main weakness was really the organization since I didn't do much revising.

[wasistdas]

i don't know about the analytical writing section, but I think your quantitative score is what you have to worry about.....

[Hopeful SLP]

i don't know about the analytical writing section, but I think your quantitative score is what you have to worry about.....

lol.

2,000, mocha. not 20,000.

[contretemps]

my undergraduate thesis is 18,500 words. LOL.

[eomentl]

I think both of my responses turned out to be that long (20 minutes at 100 WPM => 20,000 words).

Obviously, it's not the length that matters - it's the arguments that do. But I very thoroughly addressed and overanalyzed all the necessary things. My main weakness was really the organization since I didn't do much revising.

The review guides I read said there is a strong correlation between length and score. Longer essays typically get higher scores.

[Eigen]

I think revision and strong organization is more important than length, but that's just me.

As I've asked in other threads, I'd really like to see some of the data showing strong correlations between length and score- it's something that crops up often, but doesn't seem at all correct to me based on the essays/scores I've seen. Also, if you look over the grading guidelines for the essays, length isn't even mentioned. Heck, they even have sample essays and grades to look over. I think it's about the grammatical correctness, good word choice, organization and structure.

[eomentl]

I think the review guide I read which said that was the Princeton Review book (might have been a Kaplan, idr). They gave a few sample essays for the different scores, and then explicitly stated that length is one of the key identifiers of the higher scored essays (which is true given their samples). I'm not sure where they got their data/samples, but the way I see it, people who write good essays also happen to write longer essays. So length is a by-product of quality, not vice-versa. If it means anything, my essays were each about 7 paragraphs long and I got a 5.0.

Where have you seen essays/scores Eigen?

[Eigen]

From ETS, as well as the sources you mentioned.

I think you hit on what is to me, the key part- better essays are often longer, but longer essays are not necessarily better. A lot of times I hear it repeated on these boards, it's that length is a benefit in and of itself, and I don't think that's correct- all of the ETS publications on scoring, etc. that I've read don't even mention length in their discussions.

[sciencegirl]

I think if its longer, you have a greater chance of scoring of higher, but not necessarily a perfect 6. The Princeton review book mentions that longer essays receive higher scores - but you should be careful here since they don't specify what they mean by "higher scores." Higher than what? Do longer essays score more 6's? 5's? More 4's than shorter essays that score 2s and 3s?

My guess is that since its the Princeton Review, their research probably found that longer essays had a higher tendency to score 4 and above - since when you get to the 5 and 6 range, you do need really good writing and organization (both of which my essays had including length, I got a 5.5). Princeton Review generally tends to target for the mean, which is why their advice to write a lot probably will help those trying to score above 50 percentile.

[Starlajane]

I definitely think that length matters, but only because scorers are looking mostly for fluency; people who are more fluent, verbose, etc. tend to write longer essays displaying fluency of thought as well as writing, which is key.

I took the test in the beginning of November and got a 6.0 for AW. I didn't think that my essays were particularly good and I was very surprised at the score. My essays were not very long (3-5 paragraphs of about ten lines each--intro and conclusion were 5 lines each) but I had almost no typographical errors--I am in the habit of editing as I go along rather than going back.

I also used examples in my essays, which was something that our review instructor advised us to do; not abstract examples but concrete examples and at least three relating to your argument, in each essay.

I'm not sure how much the actual argument counts but I think mine were strong and nuanced, which was another thing that our instructor advised.

[InquilineKea]

I think if its longer, you have a greater chance of scoring of higher, but not necessarily a perfect 6. The Princeton review book mentions that longer essays receive higher scores - but you should be careful here since they don't specify what they mean by "higher scores." Higher than what? Do longer essays score more 6's? 5's? More 4's than shorter essays that score 2s and 3s?

My guess is that since its the Princeton Review, their research probably found that longer essays had a higher tendency to score 4 and above - since when you get to the 5 and 6 range, you do need really good writing and organization (both of which my essays had including length, I got a 5.5). Princeton Review generally tends to target for the mean, which is why their advice to write a lot probably will help those trying to score above 50 percentile.

Interesting - how do you know that PR tends to target to the mean? Even strong students often get much higher scores when they use PR

[Eigen]

Interesting - how do you know that PR tends to target to the mean? Even strong students often get much higher scores when they use PR

Actually, that's not usually true. I'm not sure of the exact data for the GRE with Kaplan and PR, but I'm pretty familiar with their MCAT reviews. For the MCAT, they target people primarily with scores below a 22 or 23 (this is out of 45), and if you're careful reading the fine print they say that they can't guarantee/show improvement if your starting score is already above average.

From what I understand, the GRE works the same way. Everyone can gain a bit from reviewing how the test is put together and practicing, but where it really shows benefits is people trying to bring their scores up to average. People who're already scoring well don't usually see the same type of improvements.

And, as with some of your other posts, please support the bolded statement. Where do you see that strong students get "much higher scores" when they use PR?

Edited December 13, 2011 by Eigen