Yesterday, the College Board rolled out the new, much-debated version of the SAT. The test now contains a writing section, and to make room for that section it no longer features those notorious analogies. Writing in The New York Times, Adam Cohen argues that the new test is not only inferior to the old one, but that it marks our collective failure to recognize how crucially important it is for the members of a free society be able to make and to recognize reasonable comparisons.

When Grover Norquist, a leading conservative activist, was on the NPR program "Fresh Air" a while back, he casually made a comparison that left the host, Terry Gross, sputtering in disbelief. "Excuse me," she said. "Did you just ... compare the estate tax with the Holocaust?" Yes, he did.

We are living in the age of the false, and often shameless, analogy. A slick advertising campaign compares the politicians working to dismantle Social Security to Franklin D. Roosevelt. In a new documentary, "Enron: The Smartest Guys in the Room," Kenneth Lay compares attacks on his company to the terrorist attacks on the United States.

Intentionally misleading comparisons are becoming the dominant mode of public discourse. The ability to tell true analogies from false ones has never been more important. But to make room for the new essay portion of the SAT that was rolled out this weekend with much fanfare, the College Board has unceremoniously dropped the test's analogy questions, saying blandly that analogical reasoning will still be assessed "in the short and long reading passages."

Replacing logic questions with writing is perfectly in keeping with these instant-messaging, 500-cable-channel times, when the emphasis is on communicating for the sake of communicating rather than on having something meaningful to say. Obviously, every American should be able to write, and write well. But if forced to choose between a citizenry that can produce a good 25-minute writing sample or spot a bad analogy, we would be better off with a nation of analogists.

My own preference would be for an SAT that asked students to demonstrate both analogical reasoning skills and writing ability (having scrutinized a number of SATs lately while helping juniors prepare for the exam, I can say that the reading comprehension sections do not, despite the College Board's assurances, do much to test analogical thinking). This could be done readily enough by asking slightly fewer questions about vocabulary, comprehension, and usage (usage questions are new as of this test as well, though that aspect of the new exam has received far less attention than the essay), and so creating room for analogies.

Readers' thoughts are, as ever, welcome.

My sixteen year old daughter took the SAT yesterday, and her comment was that it was "easy." The added essay question she also regarded as easy.

The entire concept of the standardized, multiple guess question test is a bit silly. They can measure certain things, but the things measured have only a partial bearing on the question at hand: how well will a particular student do in college (a general question), and whether a student ought to be admitted to a particular college (a more specific, and misused, thing).

Curious that Cohen's two examples of sound, or at least possibly sound, analogies come from Karl Marx and Franklin Roosevelt, while every single example of bad analogizing he offers involves political conservatives. Does he think that "intentionally misleading comparisons are becoming the dominant mode of public discourse" only on one side of the political aisle?

If you were going to write about how dubious analogies are damaging public discourse, why dig around for a looney like Grover Norquist? You could just scan the Congressional Record or the comments at Moveon.org and find countless comparisons of George Bush to Hitler and Republicans to Nazis. Those analogies have been staples of recent political discourse, rather than one-off comments by marginal characters, yet Cohen shows no concern about them. Perhaps he thinks they're reasonable -- in which case maybe he should dig out that old Miller Analogies workbook. . . .

Speaking of comparisons, aren't you making a rather crude one between the defunct analogies items and the art of tropes?

And, speaking of tropes, I remember Howard Becker warning social scientists away from the easy analogy in Writing for Social Scientists (Chicago: University of Chicago Press, 1987).

I sounds as if the new and improved SAT has incorporated the writing and usage elements of the old SAT II writing test. While I was teaching high school (gifted programs), the only perfect SAT scores our students received (which one exception on the math portion of the SAT) were on the SAT II writing exam.

Perhaps the need for reexamining the art of analogy should contain a mention of professor Ward Churchill's tempest in a teapot essay analogy comparing victims of 9/11's terror to Nazis. Ignited and fanned by media coverage, (another institution fraught with ugly analogy,) this is why analogy examination should be encouraged rather than its importance minimized.

As a lover of analogies, I agree that they are very important to the thought process and, if you grant that there should be such a thing as an SAT, then analogies should probably be on it.

But Cohen seems to think that knowing how to use analogies automatically means that one will draw the same political conclusions that he does. I will match my SAT analogies score with his any day, and I don't agree with any of the political points he made.

These people are truly becoming boring monmaniacs..whether the subject is the SAT, the state of literature, or whatever, they will find a way to bring it back to GWB.

My 13-year-old daughter just loves writing essays, so that portion of the test doesn't worry me.

Unless the grader isn't following the rubrik.

Or her paper gets graded last.... or first...

These are a father's worries...(for the not-too-distant future.

I have similar worries about the essay portion. How is this going to even be close to objective? At least with multiple choice aptitude tests, you knew that it was objective. Sure, it measured multiple-choice-test-taking-under-time-constraints, but if you thought that was a useful skill, the score meant something.

There was a feature, I think in the Atlantic, in one of their college issues, grading various snippets of what we call good writing (by Shakespeare, Hemmingway, Fitzgerald, &c) on the proposed SAT Essay rubrick. Needless to day, they did badly.

And if anything, usage questions are even easier to game. I gained about 150 points on the SAT II Writing by following this rule: If it sounds like it came out of W's mouth, it's grammatically incorrect. (This dates me, I know)

As for analogies, yes, I'm all for putting them back in the test. Doesn't the research show that underprivledged kids do better on analogies compared to other parts like reading comprehension?

Yep, I would agree with your thoughts. I think that the SAT is no longer objective, and it has become subject to the marking scheme of the examiners interpretation. It may be ok this year, but it can only go down hill, IMHO. I only hope that the SAT 2s don't go the same way, except the ones that obviously have to have a written element of course. I also wonder if the exam will now be any harder for homeschooled kids, who perhaps don't have as much inside information? I dunno.

With things like this, I'm glad I only evaluate grad students (at least so far, that's changing soon). We've found the GREs & TOEFLs to be "substandard" metrics for what makes a successful researcher (not to mention effective english speaker & tech writer). The standardized tests are a usable tool, but for us they rank *WELL* below the certified transcript, the tone & content of recommendation letters & phone checks, direct corespondance and plain old headhunting. Given all I've heard of the undergrad exams, I'm dreading the coming committee meetings.

I would agree with the points made by others here that it is improbable that an ability to recognize an appropriate analogous pair automatically means one can tell good from bad analogies (of course, we could make another section of the exam that uncovers whether students can discern conjecture from proof). However, I also don't think that the essay is a comforting replacement for the analogies. What little I know about test grading centers and the rubrics and employees and training they use, I wouldn't want to have my kids' futures resting on those - especially right now as they are getting the kinks worked out.

A revisit.

I agree with you on the use of comparisons. I have become increasingly concerned about the "word games" that politicians have been using the last few years.

What used to be an "estate tax" is now referred to as a "death tax."

"The Patriot Act"

"The Defense of Marriage Act"

"The No Child Left Behind Act"

"The Social Security Reform Act"

The list goes on... What all these have in common is a not-so-subtle word-play designed to tar anyone that opposes these as somehow being outside the mainstream. It's a shameless device to suppress opposition.

How can anyone be against their country, marriage, children, or seniors?

I think that the word games will continue....

Marshall McLuhan, obviously, a big believer in the underlying symbolism of advertising (comparative examples). I predict further re-vamping of the test, in the immediate future.

It occurs to me that Mr. Cohen is the perfect piece of evidence to support his own argument about the importance of analogies in society. Consider this line:

'Today, the federal tax system - which through much of the nation's history kept government income and expenditures in rough balance - is being compared to "theft" and recklessly dismantled.'

Without dwelling too much on the hysterical Leftist persuasion of Mr. Cohen's writing, it's necessary to point out that the ability to recognize analogies as true or false also depends on some knowledge of history. The income tax, far from existing for most of the history of this country, wasn't formally instituted as part of the Constitution until 1913. The argument that the income tax kept things in balance AFTER it was formally instituted is rather like saying, "The young man who lived off his parents kept his spending and income in balance by demanding more money every time his coffers ran low." And to use Mr. Cohen's own method, the essential thing about theft is that it is coercive. So is the income tax.

So, yes. I agree with Mr. Cohen's point. Someone ought to remind him, however, that logic starts at home.

who grades the essays? are they trying to employ more unemployable english majors?

would i really want any of the english majors i know grading my essay.

not sure, but am leaning against any subjective part of test.

who grades the essays? are they trying to employ more unemployable english majors?

Hopefully not the same people who grade the TOEFLs. (Sorry that was mean.)

I have been thinking about the best way to handle the SAT question. It seems to me that the subjective quality of the essay portion does not lend itself to valid quantifying as the rest of the test does. I would really rather see the essay portion passed on to the schools as is along with the test scores by request when the student applies. Then the schools can make of the essay portion what they will and will still have the objective results of the other parts of the test. I do think that the analogies should be put back in there.

Who grades the tests? The same type of people who grade the Advanced Placement exams.

The College Board goes through a rigorous screening process when it comes to its graders. At the grading, there is a strong system of checks and balances in place.

When the teachers (yes, these are curently employed teachers) meet to grade the exams, there are given a grading rubric - a template of sorts as to what to look for and where points should be deducted. The tests are divided into groups (and all students names are kept anonymous during grading) and the teachers are divided into groups with one person being named the "table leader." With the AP tests, table leaders are always veterans of the grading process. Two teachers will then each read the essay and independently score it. If the scores are within a point or two of one another, the score is considered valid and they move on to the next test. If there is a discrepancy, the test is then handed to the table leader to be graded. If the table leader still doesn't come to a concensus with the other teachers, it is passed up the chain of command until an agreement on the grade is reached. Those with the final say are educators (both high school and college) who have been part of the grading process for many years.

The College Board has used this method to grade many different types of exams for 50 years now, and it has worked very well so far.

But I question whether a "rubric" (which, as near as I can tell, is basically an algorithm) can if fact be used to assess quality of writing. Give me your rubric, and I bet I can produce a paragraph which will score very well according to the rubric criteria, but which in fact represents really bad writing...

Is the rubric system even public? (And SHOULD it be public?) Given my experience in not evaluating TOEFLs or GREs but rather *students* who took them as part of the admissions process into the program, not only could we get Foster's good gradin' of bad writin', but I bet you could create a "cramming" system for which writing component could be ... ahem... "prepped."

It's a stretch to say that mastering SAT analogies is going to teach students to spot political demagoguery. The SAT previously only tested simple word relationships. Rather than testing critical analysis and logical reasoning skills, these analogies tested vocabulary recognition and awareness of shades of meaning. Which was great, unless you didn't know one word out of the twelve that you're required to recognize to answer each question.

The essay trade-off was a good decision for the SAT. The analogies mainly served to "trick" students and resulted in verbal scores that didn't really reflect a student's command of language. A writing sample, however, allows the student to exhibit what he or she is capable of producing -- something that cannot be measured by multiple choice.

The essay is also better for the college admissions process. The application essay seems to serve little if any purpose in evaluating a student's writing skills -- when the essays are indeed written by the student, they are endlessly teacher- parent- and peer-reviewed until they become something other than the student's own. The SAT could potentially give colleges a better read on the applicant's true writing ability. Few colleges require the SAT II Writing, so integrating it with the SAT ought to improve the admissions process across the board.

From a timed writing sample, I could get a rough idea of a student's ability to write standard English sentences and paragraphs that are responsive to a given question. I could tell whether or not he could spell correctlly and use proper grammar. I could tell whether or not he had been taught to organize thoughts into a four or five paragraph essay.
I couldn't tell whether or not he was able to write since writing involves editing and rewriting and much thought. If I can write a couple of paragraphs (like the ones I'm writing now) under the pressure of a time limit, does that also mean that I can write meaningful and excellent and thoughtful prose? I think not. And I think students will learn more and more about how to write a quick, decently organized essay and less and less about how to really write something that shows intelligent reflection.

Are the rubrics made public? Yes, I believe they are. I am a high school Calculus teacher, and each year after the AP exam the scoring rubric is released once the tests are graded.

I honestly do not know what they are looking for in the essay, but I think they are grading grammar as well as content. So many teenagers these days spend their time writing in "instant messenger lingo" that they cannot write a cohesive paragraph, and I think this is what the SAT is trying to measure. Wait until materials prepping the kids for the new SAT become available in the bookstores. They should shed more light on what exactly is being assessed in the essay.

Nice blog. I just got mine started today. I hope it will look like this someday. Its just hard to keep finding fresh content

Do you blog? There is currently a research survey out that seeks to know "why bloggers blog." The study is being performed by a graduate student at Appalachian State University in North Carolina. The survey takes less than 5 minutes to complete. Thanks for your time.



Click Here to take the survey

The problem with the new "essay" requirement of the SAT is that there is no way that objective criteria can be developed to "grade" an essay. The SAT people used to provide what were referred to as "achievement tests" (not to be confused with advanced placement tests) some of which included an essay requirement, but the basic SAT did not.

The Week 7 Carnival of Education, opening March 23, is on the road this week, at JennyD. Submissions for the The Carnival of Education Week 7 should be sent to: jdemonte at comcast dot net no later than 10 pm Eastern, or 7 pm Pacific time, on Tuesday, March 22, 2005.

Don’t forget! A great carnival needs lots of sights and publicity.