RSS Feeds Feeds: Articles | Issues
Articles About TAP Subscribe Donate
TAPPED  |  Beat the Press

Remember Me
Forgot your password?

The symbol identifies content for paid subscribers only.


 



The group blog of The American Prospect

HOW WELL DO OUR KIDS WRITE?

While high school grades remain the single best indicator of how successful a student will be in college, a new study finds that of all the sections on the SAT, the writing section is the best predictor of academic success. The College Board decided in 2002 to roll what used to be the SAT II writing subject test into the SAT I, which now contains both an essay and a multiple choice grammar review.

I am a total writing triumphalist, but I'm a bit surprised the SAT essay section has proven to be so predictive. The topics students are asked to write about on the exam do not at all reflect the typical college assignment. The SAT prompts personal essays on broad, amorphous topics, not exercises in building an argument through carefully engaging with competing evidence. That's why I've always been a fan of the "Document Based Question," which New York State uses on its Regents examinations. Those essays give students a number of primary sources around which to build an argument. For comparison's sake, here's an example of an SAT writing prompt:

Being loyal—faithful or dedicated to someone or something—is not always easy. People often have conflicting loyalties, and there are no guidelines that help them decide to what or whom they should be loyal. Moreover, people are often loyal to something bad. Still, loyalty is one of the essential attributes a person must have and must demand of others.

Adapted from James Carville, Stickin': The Case for Loyalty

Assignment: Should people always be loyal? Plan and write an essay in which you develop your point of view on this issue. Support your position with reasoning and examples taken from your reading, studies, experience, or observations.

Now a really engaged (and privileged) high school kid, one who might even know who James Carville is, could use this prompt to write about the presidential race. But most students will write about friendships, relationships on athletic teams, and other examples of loyalty in their personal lives. If they do so grammatically, include an introduction and conclusion, and begin their paragraphs with topic sentences, they will potentially ace this section of the exam. The sad fact is, most American high school students can't do even that. And that's not a problem, of course, that can be solved at the college level.

--Dana Goldstein



COMMENTS

Only privileged kids have CNN? Under privileged engaged kids don't have the ability to follow the election?

Have you ever noticed that the five paragraph essay, with topic sentences and everything, is used by nobody in real life?

Political journalism works like this:

Pun headline.
Subhead that ends in a question mark.
Starts with and illustrative anecdote.
Blah blah blah.
Final paragraph gets to the fucking point (option for concern trolls: ends with another question).

An essay question on a standardized test is risky business, because the topic can sometimes skew the grade regardless of how objective the graders are supposed to be. The GRE is less stringent about which kinds of essays are assigned, and I know from personal experience, that the topic can have a direct reflection on the score you'll receive, however well you write. I took the general GRE twice and had such a wild variation in my writing score that it's the only conclusion I can make about it. So, perhaps assigning an essay question requiring the test-taker to build an argument based on actual evidence is not in the best interests for the SAT board.

I think the SAT really ought to test just what it tests today - a person's ability to write. An informed, argumentative essay is but one of the myriad of writing assignments a typical college student would have, and it just happens to be one of the least objective kinds of essays on which to grade a student's aptitude.

The SAT prompts personal essays on broad, amorphous topics, not exercises in building an argument through carefully engaging with competing evidence.

Maybe I'm missing something, but I'm not sure one would answer the prompt that was given without building an argument that brings up at least one counter and then dismisses it with a counter of your own. Using a presidential race or a friendship is meaningless--it's whether they use those examples well as pieces of evidence. I guess the graders could grade very easily, but that's not the same thing as being alleged.

Now a really engaged (and privileged) high school kid, one who might even know who James Carville is, could use this prompt to write about the presidential race. But most students will write about friendships, relationships on athletic teams, and other examples of loyalty in their personal lives.

Another example of the false meme that one must be priviledged in order to do well in school or know current events.

It's all luck, you know....money and knowledge...

The SAT test is, as noted, a horrible predictor of collegiate success and a ridiculous waste of time and money for all involved.

Perhaps, instead of giving our students "content neutral" evaluations like the SAT, we could come up with a test that tells us something about what students are actually learning. We have a silly bias against measuring "knowledge" in this country because it is at the bottom Bloom's Taxonomy. (Is there any other discipline besides education where jokers like Bloom or Maslow can pull wacky theories out of their respective keysters and have them received as gospel?)

Here's my idea: Have a general topic such as, say, The Civil War and Reconstruction which all high school juniors will study for an entire year leading up to the exam. Who knows, the students might learn something interesting instead of the drivel they are getting in their SAT prep classes. (Monkey is to yogurt as woodpecker is to apple sauce.)

I'm not a huge fan of the SAT or anything, but these comments - and to an extent the post as well - don't make sense given the (apparent) fact that "a new study finds that of all the sections on the SAT, the writing section is the best predictor of academic success."

If the study is right, then the writing part of the SAT is already doing its job, and thinking about what it's doing right will be a lot more fruitful than dismissing it as afactual, or disconnected from the real world, or somehow biased.

I would chalk it up to a simple concept: any kind of prose "essay" - whether reflective or analytical or argumentative - will reveal the writer's ability to compose her thoughts and express them in a way that makes a complete stranger think that she knows something. That skill is rare enough - and critical enough - that it is a pretty good proxy for general success at college, even in (relatively) non-writing disciplines.

This study is new, but we've actually known for a while that the writing test best predicts student performance: if you look at this old report, you'll find that the SAT II writing test was a better predictor than either the old SAT I verbal or math sections. There was little reason to suspect that this result would change when the SAT II writing was folded into the main SAT, although I suppose that the larger test population might have different dynamics.

Dana's "writing triumphialism" seems quite appropriate: the obvious flaws with the SAT's evaluation of writing only reinforce how critical writing is to academic success. When such a noisy measure of writing ability still produces such a strong correlation with grades, the underlying ability must be very important indeed!

This study is new, but we've actually known for a while that the writing test best predicts student performance: if you look at this old report, you'll find that the SAT II writing test was a better predictor than either the old SAT I verbal or math sections. There was little reason to suspect that this result would change when the SAT II writing was folded into the main SAT, although I suppose that the larger test population might have different dynamics.

From what I remember back from my high school days, the subject tests were voluntary and only really required by "higher-end" schools. Might there be some self-selection there then? If you're taking the test so you can apply to Harvard and Columbia, you probably already have good grades and are a bright kid...

"Only privileged kids have CNN? Under privileged engaged kids don't have the ability to follow the election?"

I didn't have CNN growing up because cable was too expensive. Luckily my parents got the paper, but that's actually pretty expensive too.

This whole topic would be so much more enlightening if any of you knew what you were talking about.

If they do so grammatically, include an introduction and conclusion, and begin their paragraphs with topic sentences, they will potentially ace this section of the exam.

This is just astonishingly wrong. Grammar is hands down the least important element of any standardized essay score. There is absolutely no requirement to start a paragraph with a topic sentence. Introductions and conclusions are completely unnecessary, although I always advise people to include them.

The criteria for any standardized essay are pretty much the same, regardless of the test. In order of importance: answer the question, support answer with required criteria, organization, and mechanics.

They are ranked in order of impact on grade, which is usually 1-6. However, the most common problem with essays is the second criteria--supporting with the required criteria.

The scoring is "holistic", which means that the reader is supposed to grade on overall impression. But in practice, grading goes something like this:


  1. Answer the question clearly--3 points. Without any support, it can't get higher. It can lose a point for mechanics. So if a writer just answers the question clearly in a well-written paragraph or two, he or she would probably get a 3. Answer the question with mechanics, 2. Scribble an answer with nothing else, 1.
  2. Answer the question in a standard 4-5 paragraph essay with good organization but the wrong kind of support, the highest grade given usually is a 4. This is hands down the most common problem students have. The SAT mandates specific examples. No specific examples, no higher score than a 4. Lose one point for mechanics errors.
  3. Answer the question in a well organized question with required support, starting grade is a 5. If the quality of writing or the support is unusually good, it will get bumped up to a 6. Mechanics get a point either way.

A kid walked straight off the plane from Korea into my class last year, after a 3 month English immersion course. He had next to no English, little knowledge of grammar. I told him it didn't matter. Just answer the question, come up with examples, and organize by example. His writing was barely readable, but what you could read was clearly an answer and support.

By the end of the class, he was getting a 4, and that's what he got on the SAT as well.

I've also taught test prep to low income Hispanics and African Americans, and the vast majority of them get their grades to a 3 or a 4.

Sixes are exceptionally rare; out of some 200 students in the past few years, I've seen just one 6.

And that, dear readers, is why the writing scores is the most predictive of grades. Kids with lousy test scores often get good grades in college, not because they know all that much but because they are taking remedial courses and working hard, getting good grades. The writing test routinely provides low income, low performing kids with their strongest scores--thus the test does a better job of predicting "academic" performance in college.

The sad fact is, most American high school students can't do even that.

Again, you have no idea what you're talking about. The essay section of the test has the highest average score of any of the sections. In my experience, most kids can write fairly well, once they get a clear idea of the criteria. And I am considerably more familiar with the entire range of student quality than you are, I'd wager.

The essay section of the test has the highest average score of any of the sections. In my experience, most kids can write fairly well, once they get a clear idea of the criteria.

But fewer than half of them actually graduate from highschool and even less even go to college.

UGG Boots is one of most popular boots in cold winter, UGGso.com have 100% sheepskin UGGS for you to pass the cold winter

Post a comment


Search TAPPED for:

Archives

About TAPPED

TAPPED, the Prospect's award-winning group blog, is a link-intensive collection of musings, ramblings, opinions and other assorted writing on the political developments of the day. See a list of our contributors.

| RSS | Twitter


Renew your print subscription or e-subscription.
Get an e-subscription for $14.95.
Give the gift of political insight. Send The American Prospect to a friend.
Change your email address or street address.
YES! I want to receive The American Prospect
— the essential source for progressive ideas.
Explore The American Prospect's award-winning investigative journalism and provocative essays in a free trial issue. Continue receiving The American Prospect at only $19.95 for a one-year subscription - a savings of 60% off the newsstand price!
First Name
Last Name
Address 1
Address 2
City
State
ZIP     
Email

Should you decide not to continue receiving the magazine after the initial free issue, simply write "cancel" on the invoice and you will not be billed.

© 2010 by The American Prospect, Inc.  |  Privacy Policy  |  Permissions and Reprints