rspence

SAT - Wikipedia, the free encyclopedia

According to Wikipedia, an MIT writing professor found a 90% correlation (r = .949) between essay length and score. This was taken from a sample of 50 graded essays. What do you guys think about this?

MandieJ

Well, the fact that there's at least some correlation is obvious; a one-paragraph essay will not get a 12, and a full two pages typically won't get a 2 (unless it's off-topic or something). That said, there is also an unfortunate, more minute correlation: a longer essay will score higher than a shorter essay of exactly the same caliber or even possibly better caliber. I write fast, but I also write pretty small; my practice essays were always ~1.5 pages, even though I used every minute. Knowing this correlation, I decided to purposefully write with larger handwriting on last week's SAT; I ended up writing on every line. I guess in two weeks we'll see whether my longer, more-largely-written essay scores higher than my first SAT essay did ... that may confirm some hunches.

rspence

Yeah, well it makes sense that there is a correlation, but a 90% correlation?

Both my essays (I took two SATs) were about 1 - 1 1/2 pages and scored 8's, which just leads to more evidence.

MandieJ

Yeah. I'm definitely not disagreeing with the correlation, and I do think it's unfortunately high.

DrGoogle

Yeap, this is true. The Princeton review guy told us that and my oldest daughter got a 12. This is why just write furiously when you take the SAT writing.

MandieJ

^ and write with large(r) handwriting if your writing is typically small!

DrGoogle

Write legible, fill up the pages.

Wood5440

That MIT professor's article was so full of holes I'm sure his colleagues were embarrassed by it. I've written a point by point refutation that I think proves that. I will send it in a PM if you're interested.
Long story short, Duh. A long essay usually contains more ideas and explains them in more detail. Multilayered idea development is a grading criterion. Better idea development is also indicative of a more active and observant student, who can be expected to be more advanced in an entire constellation of writing-related skills, ie sentence structure, vocab, grammar, coherence.
BTW Bigger or smaller handwriting doesn't change content and content is what counts. But I can tell you that microscopic handwriting, particularly when done with a dull pencil, will cause a reader to groan out loud before they start reading. Just a note.

rspence

They want it at least legible, and use good indents for your paragraphs.

Stew17

I wrote about a page and a quarter in May and got a 7. In June I just dove right in and started writing the second I was done reading the prompt. I used every second of the 25 minutes on writing the paper and filled every single line. We'll see what happens.

Wood5440

Have you ever listened to a kid with reading problems read out loud? If their problem is decoding, the excruciating process of trying and failing and then trying again to decode every other word can completely destroy the flow of ideas. Since working memory only lasts about 30 seconds or so, by the time such a reader gets to the end of a sentence, much less a paragraph, the writer's point is long gone.

The result is identical when a reader has to go through the same excruciating decoding process as a result of miserable handwriting. It seems that most often, when a word simply cannot be read and the reader has to guess and go on, that word is the one key word that carries the entire meaning of the sentence.

SATcrammers

Yes, I totally agree with everyone here. There's definitely a correlation. Writing 3-4 paragraphs is perfect, but writing anything less or anything brings down your score (statistically)

JuniorMint

This is an opinion that I have heard....

The oggle theory: the essay score is a fraction. 1 page = 6/12, 2 pages = 12/12, 1.5 pages = 9/12
If you get an odd number score, then you either had less or more quality compared to length.


This actually makes some logical sense.
Unless you are brilliant, twenty-five minutes is not going to produce a terrribly well-developed or flowing essay. If you are smart, then it is also not going to produce a bad essay. Through handwriting, spelling, grammar (technically, these do not affect your score) and examples, the grader can determine quality in thirty seconds. With this obligation fulfilled, the oggle test can begin.

So length versus score may not be causation, but along with quality, it is a correlation.
Would we really have it any other way? It is much easier to write a long essay than an impressive one.

SwaggaLikeANerd

As my AP US teacher says "A long essay isn't neccasarily a good essay, but a short essay usually isn't a good one."

So, yes i do agree.

SATcrammers

SwaggaLikeANerd, I really like what your teachers says!!