<p>Before I begin, I ask that you read the following paragraph. Assume this paragraph came from the body of an SAT essay. Without reading the entire essay, or even knowing the content of the prompt to which the essay was written as a response, please give the paragraph a score from one to six using what you believe are the criteria used by the College Board. If you have a copy of the College Board blue book, take a moment first to skim a few random paragraphs within several of the sample essays to refresh your sense of the CBs grading before you grade this paragraph. Note that you are not grading this essay for accurate historical content, but rather for writing mastery.</p>
<p>"In the 1930s, American businesses were locked in a fierce economic competition with Russian merchants for fear that their communist philosophies would dominate American markets. As a result, American competition drove the country into an economic depression and the only way to pull them out of it was through civil cooperation. American President Franklin Delanor Roosevelt advocated for civil unity despite the communist threat of success by quoting "the only thing we need to fear is fear itself," which desdained competition as an alternative to cooperation for success. In the end, the American economy pulled out of the Depression and succeeded Communism."</p>
<p>This paragraph is a portion of an essay written by a student who was trained to write an essay and submit it for grading as part of a regular SAT test. The student was a mole. A spy. A plant. (The paragraph above was the one paragraph of that essay that appeared in the opening of an article published in Inside Higher Education. The link to the entire essay given within the article led to an expired page.) </p>
<p>The professor who trained that student was Dr. Les Perelman, who teaches writing at MIT. In 2005, Dr. Perlman announced at a writing conference that he had done an analysis of essays evaluated by the College Board as part of the SAT and concluded that he could predict with 92% accuracy what grade an essay would get based exclusively on its length. Dr. Perlman's claim created quite a stir in the media at the time. There was an article in the New York Times, he was interviewed on NPR, he spoke at conferences, and media reported his findings across the nation.</p>
<p>I disagree with Dr. Perelman. I believe there are at least five major holes in his argument as I understand it. However, I believe there is one point he made that I can agree with wholeheartedly. On that point, I believe, Professor Perelman hit the bullseye.</p>
<p>My understanding of Professor Perelman's argument comes from two primary sources. The first is the transcript of an interview he did May 7, 2005 on NPR's program "Weekend Edition" moderated by Linda Wertheimer. The second was an article that appeared in the magazine Inside Higher Education. That article, which appeared March 26, 2007, was entitled "Fooling the College Board".</p>
<p>Between those two sources I extracted five main points that Perelman made. The first was that there was a 92% positive correlation between the links of an SAT essay and its score when graded by the College Board. The second point was that be essay was highly formulaic. Professor Perelman criticized the five paragraph structure in particular.</p>
<p>Third, Perelman objected to the College Board's acceptance of what he described as "total disregard for the facts". </p>
<p>The fourth point was that the SAT formula "obliterates" the writing process. According to Perelman will "you are punished for taking time to think." He pointed out that a great deal of the writing process occurs during revision. Perelman noted that college work often requires students to write essays on demand during examinations, but he maintained that those essays are on topics that students have previously studied and considered in-depth. The SAT essay, however, is entirely extemporaneous.</p>
<p>Perelmans final point was that the SAT essay was unlike the essays that students would be expected to write in college and was therefore not a good predictor of how students might write after they entered college.</p>
<p>My criticism of Perelman's position includes all of these points. I will take them in my own order.</p>
<p>Hole number one... Perelman noted that his prediction of grades based on essay length was 92% accurate and concluded that essay length was the relevant factor. It wasn't. Length was a natural consequence of other more relevant factors. (I note a rule of logic: correlation does not necessarily indicate causation.)</p>
<p>The length of an essay is determined by two factors. First is the number of ideas and second is the number of words used to express those ideas. (I am omitting handwriting. I note here that both the number and the size of words determine how many pages are filled. But it is more common that 12 essays are written in average or smaller than average handwriting. The handwriting is also, more often than not, atrocious. This is probably because the test-taker is writing like a maniac to get everything down in less than 25 minutes. In the end, I've chosen to concentrate on the number of words rather than the number of pages because Professor Perelman based his correlations on the number of words in an essay. The assertion that he could predict the grade of an essay by looking at it from across the room was a bit of an exaggeration, since there is no way at that distance he would be able to determine the number of words. But I digress.) </p>
<p>Perelman predicted essay grades with 92% accuracy. The 8% remaining were either short essays that scored well or long essays that didn't. A long essay with few ideas is usually wordy, repetitive and circular. A short essay with many ideas is usually assertive and unsupported, but it reveals an analysis in depth which is the consequence of an active intellect. These constituted Perelman's errors.</p>
<p>His 92% accuracy rate probably was the result of the more common situation. That is, students who could only think of a few ideas in response to the prompt generally ran out of things to say earlier than students whose analysis was more complete. Further, students whose insight and maturity of thought led to a greater number of ideas are generally more mature in their use of language as well, largely as a result of more practice reading and writing, either on the internet or elsewhere. In short, if their analysis of the prompt was better, it is most likely that their essay was better on all other counts (coherence, sentence structure, vocabulary, etc.) as well. It seems credible, then, that a longer essay is much more likely to be a better essay. (More on this at the end of my argument.)</p>
<p>Hole number two..... Professor Perelman criticizes the SAT on the grounds that it is not a good predictor of how a student will write in a college classroom. He objects to the five paragraph format, which he says will require an entire freshman year to be "deprogrammed". He also criticizes the timed format on the grounds that the student is "punished for taking time to think." His contrast of the SAT essay and the essays students will be expected to write in college examinations fits here is well.</p>
<p>One problem here is that there are many different kinds of essays and that a competent writer needs to have command of a wide variety of them. The infamous five paragraph essay is only one kind of essay, but it is not an invalid kind of essay. I think of it as an essay format with training wheels attached. For a beginning writer the strong structure is perfect for developing a students understanding of the variety of kinds and purposes of introductions, the many ways to develop ideas in a body paragraph, and the variety of kinds and purposes of conclusions. Once the student is equipped with this basic understanding, then the training wheels can be removed and the student is free to take their basic understanding of the elements of an essay in any direction that they wish.</p>
<p>Generally, this is the point students have reached by the time they're entering college. That many of those students will still be relying on their five paragraph format at that time, and that many of those students will therefore have to be "deprogrammed" during their freshman year is not really so surprising. If a student has moved beyond that point and mastered many kinds of essays, then the SAT can be viewed as asking such a student to "give us a number seven today".</p>
<p>Another problem, hole number three.... To criticize the SAT because it is not like college writing is to overlook the fact that, if students entered college already writing on the level of college writing, then there would be no purpose for college composition classes. The SAT essay is not designed to show how well a student will write when they graduate from college, or when they are a junior in college, or when they are a sophomore in college, or even when they are at the end of their freshman year in college. It is designed to show how well a student is writing when they begin college, </p>
<p>The essay written to be included with the student's application is another kind of essay written under other circumstances, particularly with regard to the opportunity for revision, but it has the same purpose. (And I note here that both kinds of essays can be, and most often are, 'gamed' to one degree or another.)</p>
<p>The fourth hole in Professor Perelmans argument concerns the timed format of the essay which obliterates the writing process and punishes the student for taking the time to think. I agree that revision and preparation are essential parts of the writing process, but I believe Perelman's comparison of the extemporaneous nature of an SAT essay to the essays written in a college professor's blue book exam overlooks a more analogous situation. I'm referring here to the extemporaneous nature of the responses a student might give in a college entrance interview. In both cases the questions that students may be asked are expected to be answered within the context of the student's experience of the world in general and not within the context of the student's in-depth study of a particular academic topic. The student does not have all day to compose and revise answers to the questions asked in an interview and has no specific advanced knowledge of what those questions will be.</p>
<p>The final hole relates specifically to Perelman's objection that the SAT essay shows "total disregard for the facts". I've addressed this point in other posts. I hope I'll be forgiven if I indulge in a moment of cutting and pasting.</p>
<p>In my opinion, there are five pertinent points related to a fictional essay.</p>
<ol>
<li><p>Your essay is being graded based on criteria related to your ability to write. Your grades in history, political science, biology, psychology etc. will be reflected in the subject area tests. Your essay reader is not likely to be proficient in those areas and will not have the time to fact-check your writing, nor would they want to. Consider that one person's fact is another person's fallacy. Do the Federalist Papers represent the intentions of the framers of the Constitution? Should we take them as our guide today when we deal with Constitutional issues? The SAT essay graders don't desire to deal with issues like that for all 1.5 million different essays they'll read each year, and couldn't if they did. Experienced essay readers will know there is a strong possibility that your essay evidence is fake, but they are unlikely to make an issue of it. </p></li>
<li><p>What essay readers WILL be looking for is your understanding of the requirements of a good essay. First, they will be looking for your understanding and insight into the issues raised by the essay prompt. Can you analyze those issues? Are your points and sub-points significant and relevant, or are they superficial, obvious and elementary? You can fake the evidence, but you can't fake the analysis. You have to think to do analysis. This involves two things. The first is logic. Logic is logic. It is objective. (If you haven't learned it already, you should know that the rules of formal logic are about HOW you think about propositions, not about whether or not those propositions are actually true.) The second requirement of good analysis is you, more specifically, your experience and understanding of the world and the different kinds of issues and people in it. Does your awareness extend beyond yourself to the larger world around you? The answer to that will show in your analysis.</p></li>
<li><p>Your analysis leads to your major points and sub-points. Your evidence and explanation of those points come next. Here again, the reader is looking for the quality of your thinking. Are the explanations of your ideas clear and complete? Is your evidence adequate? Again, adequate doesn't mean factual. It means relevant, significant, tied logically to the conclusions you intended the evidence to support. IF the evidence were true, would it be good evidence, logically presented?</p></li>
<li><p>After the analysis of the issues and the development of your points and evidence, comes the organization, structure and coherence of your writing. Actually, the three elements cited are all aspects of the same thing. They make your writing flow smoothly and logically through your points as the essay proceeds. Introductions, conclusions, transitions, summaries, proportion (the amount of time you spend on each point according to its importance) and unity (everything is related and fits into a single clear idea) are all important here.</p></li>
<li><p>Finally, the reader checks your writing style. Word choice, sentence structure, grammar and usage, and punctuation are all relevant here. The higher the score, the more relevant they become, but perfection is not required even in a 12 essay. You are writing a first draft, after all.</p></li>
</ol>
<p>I hope you can see that factual evidence is not really a requirement for executing a high quality essay. The CB and your reader know that, and can evaluate the essay whether the evidence is true or not. (P.S. GULLIVERS TRAVELS is not true. Does that mean it isnt good writing?)</p>
<p>There is a final comment that I could make relates to the professor's demonstration of the College Board's incompetence by way of sending his 'mole' into an actual SAT exam, the intention being that said mole would produce an SAT essay designed to be low in quality and high in score. </p>
<p>It's unfortunate I can't put this paragraph into the context of the entire essay. (One paragraph of that essay was quoted in full in the opening of the article published in Inside Higher Education. The link to the entire essay given within the article led to an expired page.) </p>
<p>I'm assuming that this paragraph represents one of the examples the mole student used to in support some larger theme, but there really is no way to validate that assumption without reference to the entire original essay. That having been said, I would still invite the reader to compare this paragraph with the paragraphs given in the sample essays in the College Board blue book. Remember that the score should not be a reflection of what is expected in college, but rather of what is produced by high school juniors and seniors. What score would you give it? (I will post the answer after youve had a chance to consider the matter and post your scores.)</p>
<p>In conclusion, I would highlight the point on which Dr. Perelman has my wholehearted agreement. Too often Ive read posts by people who want to know the sure-fire formula
for a 12 essay. They want to know prompt archetypes, how many and what kind of examples, how many paragraphs and what to say in the introduction and conclusion. This is born of the thought that the SAT can be cracked, hacked, and B.S.ed. Just learn the secret formula and give those moronic SAT essay readers what they want.</p>
<p>And the result is what both Perelman and I detest. The result is the formulaic, all to typical Hitler-Gatsby-Say-The-Most-Obvious-Thing-IN-5-Paragraphs-BS-When-You-Haven't-Any-Idea-What-You're-Talking-About-Smartass-Test-Prep-Tutor crap.</p>
<p>Heres the big secret. Thats not what the readers want. They want some significant and original ideas, adequately supported and coherently expressed. They want you to show some genuine personal involvement in what you are saying. Thats what they want, but its only very rarely that thats what they get. </p>
<p>What they get is average essay after average essay. They get 4s. Not much to say, not much originality, and lots of formula-written, 5 paragraph essays. Theres nothing really bad about that, just nothing really good, either. Its about average.</p>
<p>You are getting 4s and you want a 12? Heres the formula. Recognize that there is a difference between ten years of experience and one year of experience ten times. Get your head out of your own rear and look at the world and the people around you. Learn about them. Read about them. Think about them. Talk about them. Do things you havent done before. Let new things into your life. After a year or two of that kind of involvement, you might actually start having some original and significant ideas. When someone asks you a question about the world, you might even have something intelligent to say. And that, first and foremost, is what good writing is about: having something worthwhile to say.</p>