A remarkable story appeared in the Chicago Tribune last Sunday. It's about a remarkable young man named Derrius Quarles and his determination to get somewhere and be somebody. A foster child whose father was stabbed to death when Derrius was four and whose mother struggled with drugs, he had the strength of character to overcome the vagaries of his life and end up winning scholarships to excellent colleges all over the country, including Morehouse, where he now attends.
Derrius's outlook can be summed up here: "You can't go around thinking you are inferior just because you didn't have parents," he says. "For me, it's about knowing where you are from and accepting it, but more important, knowing where you are going."
At 17, he was living on his own, keeping himself together and focusing on the future. He budgeted his money and when he did the grocery shopping he avoided junk food in favor of fruits and vegetables. He never took his eyes off his goal.
Derrius was fortunate to have someone see his potential. As often happens and as studies have shown, sometimes just one person can have an immense effect on a young person. For Derrius, that person was his summer biology teacher, Nivedita Nutakki, who told him he shouldn't waste his talent. Arriving as a freshman with a 2.5 GPA at Kenwood Academy, Derrius was taking three AP classes and earning a 3.6 by his junior year.
In the middle of this amazing story is a passage that made me angry: "Even his oversize ambition couldn't get Quarles past one roadblock. He dreamed of attending Harvard, until one college adviser told him his 28 ACT score was simply not high enough. He abandoned his plans."
Regardless of whether Harvard or Morehouse (or any other institution) would be the best for him, no college adviser should have told him not to bother applying to Harvard or anywhere else. It is not for that person to say. I always tell students that it is their right and privilege to apply wherever they want so long as they understand clearly what the odds are. In this case, it's a shame that someone assumed Derrius wouldn't get into Harvard on the basis of that score. And it's almost criminal that Derrius was convinced to abandon his plans as a result. Any college adviser who thinks he or she can or should make that determination suffers from a bad case of hubris.
The truth is, we cannot know what the future holds for our advisees. We don't know what colleges and universities will decide, even though we can come up with some pretty accurate guesses if we've had enough experience. We don't know how or when a student will suddenly "take off" and make us proud. But all you have to do is read Derrius's story to know that no matter where he went he'd make good, and that as a result the test scores say very little about him (and even so, they are miles above the average scores of someone from his background).
Again, I'm not saying Derrius should have gone to Harvard or anywhere else or that he's deprived as a result--clearly not. But no one should have told him it wasn't possible. Anything is possible, as this young man has already shown. While we may think we know a lot, the future always confounds us and we should always be humble in its presence.
Edit, Oct. 23: I was so taken with the story that I forgot I had met Derrius through Scholarship Chicago, a program that provides help and mentorship throughout the college process and in college as well as financial assistance during college. When I met him he had already received several admission offers from colleges and was racking up scholarships. I would never have guessed at the hurdles he was going through he was so poised, confident, and focused.
Showing posts with label test scores. Show all posts
Showing posts with label test scores. Show all posts
Tuesday, October 6, 2009
Monday, January 5, 2009
Taming the Testing Dragon
The coming new year for the Chinese is the Year of the Ox, so I'm going to go with that spirit and suggest it's time to tell students applying to college that they can take the ACT or SAT once and that's it. The current discussion about "score choice" and what colleges want versus what the College Board wants them to want versus what ACT is doing and how all the test prep companies figure into all this mess cries out for a Gordian knot solution (sorry to mix cultural references). I've said many times that the admission process is far too complicated, overgrowing students' educations and becoming more important than what they're learning in high school. So, to repeat: Limit students to taking the SAT or ACT once. Period. We'll all be better off.
I know many will shriek at this limitation of students' right to flay themselves in the quest for collegiate Valhalla, but let's think about some of the issues (and for the sake of argument I'll try to limit my own abhorrence of these tests):
Argument: Students should have the right to take the test as many times as they wish and report their best scores to enhance their applications. It's a free country!
Response: Many may see a one-time testing approach as too similar to the European all or nothing tests (and many people still think of the tests as the be-all-and-end-all for college admission). But a test score in the U.S. is not determinative; it can be considered, downplayed, or lauded by any college to whom it is reported. And it doesn't limit where students can apply. Hundreds of institutions don't even use the scores or are score optional, with little effect on the quality of their student bodies. And even those who use scores often downplay them when necessary to enroll athletes, legacies, talented minority students, and so on. In other words, scores are fungible, not fixed; one set of scores or six doesn't really make that much of a difference.
Argument: Students should have the opportunity to get their best scores to indicate their true abilities.
Response: The College Board, which produces the SAT, long ago gave up the myth they themselves originated that the test can't be coached since it was an indicator of innate abilities. It even offers its own prep courses to subvert (sorry, prepare for) the test. And companies like Princeton Review and others, whatever one might think of them, have demonstrated that it is possible to raise scores not by knowing more about geometry or American history, but by knowing how the test is structured. How this adds to a student's academic qualities has yet to be determined. Ironically, students who take the test again and raise their scores significantly can be accused of cheating, and a very high test score coupled with low or mediocre grades can brand an applicant a slug in class. So it's damned if you do, damned if you don't, and whatever "true ability" is is certainly not being measured by the ACT or SAT.
Argument: Taking the test several times is just a good way to get a better score; it's not unfair or anything.
Response: Practice tests already exist for the SAT and the ACT, namely, the PSAT and the PLAN. Administered at students' schools, they come back with detailed explanations of what students got right and wrong and what concepts they need to work on for when they take the test for real. In fact, students can take the PLAN and the PSAT in their sophomore and junior years, without scores being reported anywhere, so they have plenty of time to see what they need to improve when the time comes.
Argument: If students want to take the test multiple times, what's wrong with that?
Response: Well, nothing, really, if you think that going through hours of test prep, anxiety, and craziness, not to mention hundreds if not thousands of dollars somehow are positive educational developments. Testing already crowds out actual academic subjects as early as third grade, and drilling for college entrance exams is the most tedious, boring, and retrograde activity a school can indulge in. No wonder students hate it. Students are already idiotically overtested and as to whether it's always the students' choice to take and retakes the tests, I'd look more closely at parental influence.
Argument: Test prep and multiple testing give students a taste of what's expected of them in college.
Response: I for one wouldn't attend an institution that focused on testing like that as an evaluative measure. Does it introduce concepts to think about, encourage intellectual development, accurately measure what a student knows? No. Testing is something to be gotten through, not embraced. It is intellectually deadening and as welcome as plague. Most students will find that, except for huge institutions with classes of hundreds, they will rarely see SAT-like tests.
Argument: A student can have an "off" day on the one day that the test is given, leading to a "false negative" score.
Response: A student's score is always considered in the context of high school strength and GPA; an "off" day could easily be seen as that in the admission process when so-so scores accompany an otherwise strong record. Scores (as well as every other application element) are subject to the sense and good judgment of the individuals reading applications, so there is every reason to believe that a sense of who's "off" can be developed with a one-test limit even as it is now with multi-test scores being reported.
Argument: Colleges need to be able to put the best scores of their applicants together so they can put together the best profiles possible, so allowing students to take the tests multiple times is to their advantage.
Response: This is a college issue, not a student or educational issue. In my experience we spent much more time talking about students' activities, courses, and achievements than their test scores in committee. One argument is that no college wants to have poorer scores to report than its competitors do. But if everyone has only the one score to report, a deflation will occur across the board and equilibrium should be maintained.
Argument: Multiple scores enable colleges to get the best bond ratings and rankings.
Response: Aside from the insidiousness of these methods of rating colleges, the same principle applies as in the answer above: If all institutions have the same one-test figures, it seems likely that everything will reach an equilibrium that would merely lead to a recalibration of the ratings and the rankings.
Argument: A single test date would put more pressure on students because there would be no "safety valve" if the results weren't good.
Response: Probably, but it would be up to colleges and universities to put the test in a more enlightened context by showing how they use it and where it actually stands in the admission hierarchy. In fact, adopting the one-time test might cause colleges to rethink how they use it because it would be a rawer picture of the test-taker, more "authentic," so to speak. A single date would be intense, but knowing it would all be over afterwards might be liberating. If the date were at the end of junior year, results received in the summer might provide motivation for doing better in courses senior year to make up for a poor score.
There are many reasons to support a one-time only test:
1. The hours and dollars spent on test prep seriously distract from more useful activities like homework and true academic development, whether they're sponsored through schools as classtime sessions or after school. Especially in areas with a high percent of first-generation or poor students, it is critical that time and dollars not be sacrificed for something as ephemeral and uncertain as test prep. I know of one school that has spent nearly $60,000.00 on test prep for students who could have better been served by spending that money on academic enhancement, tutoring, equipment, and so on. A recent article in Harper's magazine (September 2008) documenting a year of test prep in a New York City school is illuminating. Click on this entry's title to go to the article.
2. Test prep as a part of schooling is a kind of regression to the days of rote learning, which has long since been abandoned in this country. It kills motivation, deadens intellectual curiosity, and makes education look like a hoop to jump through rather that an ongoing source of personal development. It makes students and teachers cynical; no good teacher I know will sacrifice a classroom discussion about "Death of a Salesman" for an SAT vocabulary drill. And no student would willingly attend. He may not care for Arthur Miller, either, but at least there's the possibility that something interesting might come up. (Furthermore, it most disadvantages those who can least afford it: First-generation and other underserved students who most need to learn the basics of English, math, and so on to do well in college.)
3. Multiple testing opportunities favor those already privileged; a one-time test date can even the playing field to a certain extent. While privileged students can still afford the books and testing that non-privileged students can't, the one-time test means that what you see is what you get on the other end. Will non-privileged students suffer because they can't afford test prep or the ability to try again? Those students already have extra consideration for their backgrounds and lack of educational support, so test scores will continue to be seen that way. (Remember, test scores have been shown definitiely to have cultural biases.) It will affect privileged students more, because they'll have to live with their scores without being able to tinker with them over and over. (Idea: After adopting the one-test only policy, schools can ask "Estimate how much preparation you received or paid for before the test." The more that's reported, the less credible the test. Fantasy I know, but still...)
4. Multiple testing is a financial bonanza that offers little real improvement in educational environment, siphoning off money from individuals and school systems that could be put to better use. While test prep companies and the College Board get rich coming and going, that money doesn't go to enhancing educational opportunity (although the companies do provide their services pro bono in many circumstances). And schools that can't really afford it are led to chase the ephemara of scores as a way to getting their students into college rather than focusing on building their academic programs.
5. The ability to take the test multiple times fosters the idea that testing is more crucial than it really is. Like Sysiphus, rolling the stone up the hill only to have it roll down again so he has to start all over again, multiple testing really accomplishes very little while creating great strain and anxiety. Nothing gets learned, nothing is accomplished, other more fruitful opportunities are passed up, and in the end, an admission decision can be made in spite of scores as much as because of them. Consistently, according to NACAC, a student's GPA and course strength are the most compelling parts of the application; the scores are really more window dressing. They're easy to look at and mess with; they have acquired a magical quality; and they play into our love of lists and bests/worsts. So while admission officers tend to drool over big scores, they also can see the proverbial "diamonds in the rough" that shine in class without the burnishing of high test scores.
Adopting a one-time only testing policy may be seen as radical, but it would help to simplify and equalize the whole testing universe. Colleges and universities should consider cutting back the testing underbrush while at the same time promoting the importance of academic achievement more forcefully.
I know many will shriek at this limitation of students' right to flay themselves in the quest for collegiate Valhalla, but let's think about some of the issues (and for the sake of argument I'll try to limit my own abhorrence of these tests):
Argument: Students should have the right to take the test as many times as they wish and report their best scores to enhance their applications. It's a free country!
Response: Many may see a one-time testing approach as too similar to the European all or nothing tests (and many people still think of the tests as the be-all-and-end-all for college admission). But a test score in the U.S. is not determinative; it can be considered, downplayed, or lauded by any college to whom it is reported. And it doesn't limit where students can apply. Hundreds of institutions don't even use the scores or are score optional, with little effect on the quality of their student bodies. And even those who use scores often downplay them when necessary to enroll athletes, legacies, talented minority students, and so on. In other words, scores are fungible, not fixed; one set of scores or six doesn't really make that much of a difference.
Argument: Students should have the opportunity to get their best scores to indicate their true abilities.
Response: The College Board, which produces the SAT, long ago gave up the myth they themselves originated that the test can't be coached since it was an indicator of innate abilities. It even offers its own prep courses to subvert (sorry, prepare for) the test. And companies like Princeton Review and others, whatever one might think of them, have demonstrated that it is possible to raise scores not by knowing more about geometry or American history, but by knowing how the test is structured. How this adds to a student's academic qualities has yet to be determined. Ironically, students who take the test again and raise their scores significantly can be accused of cheating, and a very high test score coupled with low or mediocre grades can brand an applicant a slug in class. So it's damned if you do, damned if you don't, and whatever "true ability" is is certainly not being measured by the ACT or SAT.
Argument: Taking the test several times is just a good way to get a better score; it's not unfair or anything.
Response: Practice tests already exist for the SAT and the ACT, namely, the PSAT and the PLAN. Administered at students' schools, they come back with detailed explanations of what students got right and wrong and what concepts they need to work on for when they take the test for real. In fact, students can take the PLAN and the PSAT in their sophomore and junior years, without scores being reported anywhere, so they have plenty of time to see what they need to improve when the time comes.
Argument: If students want to take the test multiple times, what's wrong with that?
Response: Well, nothing, really, if you think that going through hours of test prep, anxiety, and craziness, not to mention hundreds if not thousands of dollars somehow are positive educational developments. Testing already crowds out actual academic subjects as early as third grade, and drilling for college entrance exams is the most tedious, boring, and retrograde activity a school can indulge in. No wonder students hate it. Students are already idiotically overtested and as to whether it's always the students' choice to take and retakes the tests, I'd look more closely at parental influence.
Argument: Test prep and multiple testing give students a taste of what's expected of them in college.
Response: I for one wouldn't attend an institution that focused on testing like that as an evaluative measure. Does it introduce concepts to think about, encourage intellectual development, accurately measure what a student knows? No. Testing is something to be gotten through, not embraced. It is intellectually deadening and as welcome as plague. Most students will find that, except for huge institutions with classes of hundreds, they will rarely see SAT-like tests.
Argument: A student can have an "off" day on the one day that the test is given, leading to a "false negative" score.
Response: A student's score is always considered in the context of high school strength and GPA; an "off" day could easily be seen as that in the admission process when so-so scores accompany an otherwise strong record. Scores (as well as every other application element) are subject to the sense and good judgment of the individuals reading applications, so there is every reason to believe that a sense of who's "off" can be developed with a one-test limit even as it is now with multi-test scores being reported.
Argument: Colleges need to be able to put the best scores of their applicants together so they can put together the best profiles possible, so allowing students to take the tests multiple times is to their advantage.
Response: This is a college issue, not a student or educational issue. In my experience we spent much more time talking about students' activities, courses, and achievements than their test scores in committee. One argument is that no college wants to have poorer scores to report than its competitors do. But if everyone has only the one score to report, a deflation will occur across the board and equilibrium should be maintained.
Argument: Multiple scores enable colleges to get the best bond ratings and rankings.
Response: Aside from the insidiousness of these methods of rating colleges, the same principle applies as in the answer above: If all institutions have the same one-test figures, it seems likely that everything will reach an equilibrium that would merely lead to a recalibration of the ratings and the rankings.
Argument: A single test date would put more pressure on students because there would be no "safety valve" if the results weren't good.
Response: Probably, but it would be up to colleges and universities to put the test in a more enlightened context by showing how they use it and where it actually stands in the admission hierarchy. In fact, adopting the one-time test might cause colleges to rethink how they use it because it would be a rawer picture of the test-taker, more "authentic," so to speak. A single date would be intense, but knowing it would all be over afterwards might be liberating. If the date were at the end of junior year, results received in the summer might provide motivation for doing better in courses senior year to make up for a poor score.
There are many reasons to support a one-time only test:
1. The hours and dollars spent on test prep seriously distract from more useful activities like homework and true academic development, whether they're sponsored through schools as classtime sessions or after school. Especially in areas with a high percent of first-generation or poor students, it is critical that time and dollars not be sacrificed for something as ephemeral and uncertain as test prep. I know of one school that has spent nearly $60,000.00 on test prep for students who could have better been served by spending that money on academic enhancement, tutoring, equipment, and so on. A recent article in Harper's magazine (September 2008) documenting a year of test prep in a New York City school is illuminating. Click on this entry's title to go to the article.
2. Test prep as a part of schooling is a kind of regression to the days of rote learning, which has long since been abandoned in this country. It kills motivation, deadens intellectual curiosity, and makes education look like a hoop to jump through rather that an ongoing source of personal development. It makes students and teachers cynical; no good teacher I know will sacrifice a classroom discussion about "Death of a Salesman" for an SAT vocabulary drill. And no student would willingly attend. He may not care for Arthur Miller, either, but at least there's the possibility that something interesting might come up. (Furthermore, it most disadvantages those who can least afford it: First-generation and other underserved students who most need to learn the basics of English, math, and so on to do well in college.)
3. Multiple testing opportunities favor those already privileged; a one-time test date can even the playing field to a certain extent. While privileged students can still afford the books and testing that non-privileged students can't, the one-time test means that what you see is what you get on the other end. Will non-privileged students suffer because they can't afford test prep or the ability to try again? Those students already have extra consideration for their backgrounds and lack of educational support, so test scores will continue to be seen that way. (Remember, test scores have been shown definitiely to have cultural biases.) It will affect privileged students more, because they'll have to live with their scores without being able to tinker with them over and over. (Idea: After adopting the one-test only policy, schools can ask "Estimate how much preparation you received or paid for before the test." The more that's reported, the less credible the test. Fantasy I know, but still...)
4. Multiple testing is a financial bonanza that offers little real improvement in educational environment, siphoning off money from individuals and school systems that could be put to better use. While test prep companies and the College Board get rich coming and going, that money doesn't go to enhancing educational opportunity (although the companies do provide their services pro bono in many circumstances). And schools that can't really afford it are led to chase the ephemara of scores as a way to getting their students into college rather than focusing on building their academic programs.
5. The ability to take the test multiple times fosters the idea that testing is more crucial than it really is. Like Sysiphus, rolling the stone up the hill only to have it roll down again so he has to start all over again, multiple testing really accomplishes very little while creating great strain and anxiety. Nothing gets learned, nothing is accomplished, other more fruitful opportunities are passed up, and in the end, an admission decision can be made in spite of scores as much as because of them. Consistently, according to NACAC, a student's GPA and course strength are the most compelling parts of the application; the scores are really more window dressing. They're easy to look at and mess with; they have acquired a magical quality; and they play into our love of lists and bests/worsts. So while admission officers tend to drool over big scores, they also can see the proverbial "diamonds in the rough" that shine in class without the burnishing of high test scores.
Adopting a one-time only testing policy may be seen as radical, but it would help to simplify and equalize the whole testing universe. Colleges and universities should consider cutting back the testing underbrush while at the same time promoting the importance of academic achievement more forcefully.
Monday, November 24, 2008
Harvard's SAT Exceptionalism
We all know that Harvard can do whatever it wants and usually does. And what it does usually sets the course for the rest of the country's colleges and universities (at least in the sense of giving them something to think about...) But this recent comment in the Boston Globe really has me steaming:
Harvard's dean of admissions, Bill Fitzsimmons, said standardized tests that are based on high school course work have proven superior to the SAT at determining college readiness and said he hoped such tests will begin to play a larger role in admissions decisions.
"Wouldn't it be better for students to study chemistry and math and language, than trying to game a somewhat esoteric set of test-taking skills?" he asked.
Yet Harvard "could never be SAT-optional," he said, because of the need for a national measure to identify top students, including those from urban or rural high schools that don't send many students to elite colleges.
Mr. Fitzsimmons recently chaired a committee that explored the use of standardized testing in college admission. It recommended that the SAT be de-emphasized in admission decisions for all the reasons that many of us have been giving for many years. Clearly, however, this recommendation is meant to apply only to lesser institutions, and not Harvard itself. Harvard couldn't possibly do what the plebes do because it needs to have a "national measure" to identify top students, unlike everyone else, who presumably only need, what, "local" measures? Or other more scurrilous ways of evaluating applications?
This smug exceptionalism not only throws the committee's study and recommendations into doubt (were you just wasting everyone's time?) it also reeks of a "Let them eat cake" mentality that makes us common folk want to grab our pitchforks and settle someone's hash. Why can de-emphasizing the SAT work for everyone else but not possibly for Harvard? Surely with its 372 years of experience it knows how to identify a talented student by now without a test that has only been in existence for 80 years or so. And surely, if it's good enough for Harvard, why should anyone else give it up, despite the fact that many colleges and universities have, without any diminution in their ability to attract and identify able applicants.
Fitzsimmons connects using the SAT with the necessity of finding "urban or rural" students who might otherwise, presumably, be overlooked without it. But this is just protective coloring, meant to reassure us that Harvard needs the scores to find talented first generation and minority students it would otherwise miss. But most of those students won't do well on the SAT, so Harvard would either have to reject them or ignore the scores. And Harvard has the resources to find anyone it wants, so why rely on the scores when it's just finished downplaying them?
So the message and value of the study become muddied and pointless. Whatever we may think about America's top university "brand," we must acknowledge that Harvard's imprimatur on anything carries great weight. Without Harvard's taking the lead by adopting a more enlightened view of admission testing (even if it stops short of de-emphasizing it), what was the point of doing the study in the first place? Of course, it's not bound to follow through on any conclusions, but wouldn't its participation suggest it was willing to lead where those conclusions might point? To say categorically that it couldn't possibly risk its reputation by de-emphasizing scores, even though that was the conclusion of the study seems arrogant at best, cynical and unilateral at worst.
If Harvard wants to avoid being the Marie Antoinette of colleges and universities, perhaps it should get out with the people a little instead of simply visiting its faux village to commune with the peasants. It might experiment with how it uses the SAT by making decisions on a sampling of students without using scores and following them through over the years. With the immense resources at its disposal, Harvard could actually perform a service rather than retreat into its opulence. Leadership on this issue would be to take the study's conclusions seriously, as if they applied to ALL institutions and not just everyone else.
One interesting irony of this situation is that the SAT was once touted specifically as a way to find otherwise hidden talents throughout the country when many colleges had narrowly specific entrance exams of their own. It was conceived of as a great leveller. But with the increasing connection of test performance with income, this seems no longer defensible; now it's as much a barrier to admission as a way into college. The idea of the SAT's being a "national standard" that is somehow equal across the country has been definitvely refuted over and over again. And being able to find talented students in out of the way places has never been easier. So what's Harvard's excuse? Apres moi, le deluge...
Harvard's dean of admissions, Bill Fitzsimmons, said standardized tests that are based on high school course work have proven superior to the SAT at determining college readiness and said he hoped such tests will begin to play a larger role in admissions decisions.
"Wouldn't it be better for students to study chemistry and math and language, than trying to game a somewhat esoteric set of test-taking skills?" he asked.
Yet Harvard "could never be SAT-optional," he said, because of the need for a national measure to identify top students, including those from urban or rural high schools that don't send many students to elite colleges.
Mr. Fitzsimmons recently chaired a committee that explored the use of standardized testing in college admission. It recommended that the SAT be de-emphasized in admission decisions for all the reasons that many of us have been giving for many years. Clearly, however, this recommendation is meant to apply only to lesser institutions, and not Harvard itself. Harvard couldn't possibly do what the plebes do because it needs to have a "national measure" to identify top students, unlike everyone else, who presumably only need, what, "local" measures? Or other more scurrilous ways of evaluating applications?
This smug exceptionalism not only throws the committee's study and recommendations into doubt (were you just wasting everyone's time?) it also reeks of a "Let them eat cake" mentality that makes us common folk want to grab our pitchforks and settle someone's hash. Why can de-emphasizing the SAT work for everyone else but not possibly for Harvard? Surely with its 372 years of experience it knows how to identify a talented student by now without a test that has only been in existence for 80 years or so. And surely, if it's good enough for Harvard, why should anyone else give it up, despite the fact that many colleges and universities have, without any diminution in their ability to attract and identify able applicants.
Fitzsimmons connects using the SAT with the necessity of finding "urban or rural" students who might otherwise, presumably, be overlooked without it. But this is just protective coloring, meant to reassure us that Harvard needs the scores to find talented first generation and minority students it would otherwise miss. But most of those students won't do well on the SAT, so Harvard would either have to reject them or ignore the scores. And Harvard has the resources to find anyone it wants, so why rely on the scores when it's just finished downplaying them?
So the message and value of the study become muddied and pointless. Whatever we may think about America's top university "brand," we must acknowledge that Harvard's imprimatur on anything carries great weight. Without Harvard's taking the lead by adopting a more enlightened view of admission testing (even if it stops short of de-emphasizing it), what was the point of doing the study in the first place? Of course, it's not bound to follow through on any conclusions, but wouldn't its participation suggest it was willing to lead where those conclusions might point? To say categorically that it couldn't possibly risk its reputation by de-emphasizing scores, even though that was the conclusion of the study seems arrogant at best, cynical and unilateral at worst.
If Harvard wants to avoid being the Marie Antoinette of colleges and universities, perhaps it should get out with the people a little instead of simply visiting its faux village to commune with the peasants. It might experiment with how it uses the SAT by making decisions on a sampling of students without using scores and following them through over the years. With the immense resources at its disposal, Harvard could actually perform a service rather than retreat into its opulence. Leadership on this issue would be to take the study's conclusions seriously, as if they applied to ALL institutions and not just everyone else.
One interesting irony of this situation is that the SAT was once touted specifically as a way to find otherwise hidden talents throughout the country when many colleges had narrowly specific entrance exams of their own. It was conceived of as a great leveller. But with the increasing connection of test performance with income, this seems no longer defensible; now it's as much a barrier to admission as a way into college. The idea of the SAT's being a "national standard" that is somehow equal across the country has been definitvely refuted over and over again. And being able to find talented students in out of the way places has never been easier. So what's Harvard's excuse? Apres moi, le deluge...
Monday, October 20, 2008
Rolling & Reading the Bones
Humans have always wanted to know the future. We want to know whether we'll be president or get an A or marry the person of our dreams. Despite the prominence of science, horoscopes still get printed in the newspaper and tarot card readers are still in business. Not so many people plunge their hands into steaming sheep entrails any more but we still read our fortunes from the fortune cookie and scan the papers for clues to a big score or the winner of the next race. We want to minimize our risks and maximize our benefits, so if we know what's going to happen, we can all make the right choices.
College admission testing is one way of trying to know the future. It was initially designed for one purpose and one purpose only: To predict academic success in a student's freshman year in college. It's a way to feel that we're not just relying on hunches or guesses to decide who should get into a college and who shouldn't. Wearing the trappings of science, testing looks like a no-lose proposition for college admission officers. The numbers, the percentages, the graphs and charts, all point to rational and dispassionate conclusions. They tell us how the person will perform over the next year, and they make us feel good. We have seen into the future and believe we know how it will turn out.
Numerous studies, however, including an extensive 20-year longitudinal study at Bates College, have shown that college admission officers can predict a student's performance just as well with or without the use of testing. But it looks too much like tarot card reading when you do it without the numbers, so many institutions are afraid to do without. Putting students through the ritual of testing provides a superstructure for our superstition. It feels concrete, something you can really get your arms around.
It's still a prediction, though, and therefore not a certainty by any means. Unlike predictions in science, where physical laws enable scientists to tell when a planet will be where, testing predictions can't do anything similar because they attempt to rationalize the non-rational: human behavior. It might be more accurate to say that they attempt to set in stone something that flows in unknowable directions. More than any of the other elements in a student's application, testing feels more like divination than fact. At least when you look at a student's overall trajectory you can get a good picture of the possible result. With testing, all bound up in a few hours of highly stylized behavior and perhaps many more hours of self-abnegation for its sake, there's a sense of inevitability that is entirely unearned. Seeing into the future often depends on heightening present reality (think of psychic trances or Ouija board concentration), and testing is a good example of this phenomenon. It looks like we know the future when we get a number or a ghostly emanation or a particular arrangement of cards, but we're really getting only what we see.
The future remains unknowable, especially for unpredictable humans and the unpredictable world. As much as we try to get some control or even act on whatever predicting devices we consult, the future very often eludes us. And we should never be too proud to remember that even when something turns out as we intended, it's only because we were lucky, not because the world bowed to our intent.
Knowing the future has its problems, too. Let's assume for a moment that testing really can predict student success. How will that affect the behaviors not only of the student being tested, but of all those who come into contact with him or her? How often have you noticed how someone's attitude toward a student changes when his or her test scores are revealed? Glimpses into the future can alter our behavior and cause us to distort that very future. Think of Macbeth, told by the three witches that he'd eventually be King of Scotland. He's faced with a truly miserable puzzle (helped not at all by his wife): Should he just go along as he has been and assume that he'll get the throne sooner or later or should he take action to make sure it happens? Or are these the same things? Once you throw the eye of newt into the pot, you're done for because you can't not act, so what is the reality of your situation? Who controls it? Your future will arrive no matter what you do, so you just have to make the best of it.
When we predict the future with test scores, are we acting as though the future were in our power to command it or are we simply allowing it to proceed as it should? To me, scores are more akin to tarot cards than science. Tarot cards have just as complex an interpretive system surrounding them as testing does. And whether we're talking tarot or psychic readings, what the subject brings to the table is most important. The future is just extrapolation from the past coupled with wishes and expectations; isn't a student's past a better thing to base our predictions on?
College admission testing is one way of trying to know the future. It was initially designed for one purpose and one purpose only: To predict academic success in a student's freshman year in college. It's a way to feel that we're not just relying on hunches or guesses to decide who should get into a college and who shouldn't. Wearing the trappings of science, testing looks like a no-lose proposition for college admission officers. The numbers, the percentages, the graphs and charts, all point to rational and dispassionate conclusions. They tell us how the person will perform over the next year, and they make us feel good. We have seen into the future and believe we know how it will turn out.
Numerous studies, however, including an extensive 20-year longitudinal study at Bates College, have shown that college admission officers can predict a student's performance just as well with or without the use of testing. But it looks too much like tarot card reading when you do it without the numbers, so many institutions are afraid to do without. Putting students through the ritual of testing provides a superstructure for our superstition. It feels concrete, something you can really get your arms around.
It's still a prediction, though, and therefore not a certainty by any means. Unlike predictions in science, where physical laws enable scientists to tell when a planet will be where, testing predictions can't do anything similar because they attempt to rationalize the non-rational: human behavior. It might be more accurate to say that they attempt to set in stone something that flows in unknowable directions. More than any of the other elements in a student's application, testing feels more like divination than fact. At least when you look at a student's overall trajectory you can get a good picture of the possible result. With testing, all bound up in a few hours of highly stylized behavior and perhaps many more hours of self-abnegation for its sake, there's a sense of inevitability that is entirely unearned. Seeing into the future often depends on heightening present reality (think of psychic trances or Ouija board concentration), and testing is a good example of this phenomenon. It looks like we know the future when we get a number or a ghostly emanation or a particular arrangement of cards, but we're really getting only what we see.
The future remains unknowable, especially for unpredictable humans and the unpredictable world. As much as we try to get some control or even act on whatever predicting devices we consult, the future very often eludes us. And we should never be too proud to remember that even when something turns out as we intended, it's only because we were lucky, not because the world bowed to our intent.
Knowing the future has its problems, too. Let's assume for a moment that testing really can predict student success. How will that affect the behaviors not only of the student being tested, but of all those who come into contact with him or her? How often have you noticed how someone's attitude toward a student changes when his or her test scores are revealed? Glimpses into the future can alter our behavior and cause us to distort that very future. Think of Macbeth, told by the three witches that he'd eventually be King of Scotland. He's faced with a truly miserable puzzle (helped not at all by his wife): Should he just go along as he has been and assume that he'll get the throne sooner or later or should he take action to make sure it happens? Or are these the same things? Once you throw the eye of newt into the pot, you're done for because you can't not act, so what is the reality of your situation? Who controls it? Your future will arrive no matter what you do, so you just have to make the best of it.
When we predict the future with test scores, are we acting as though the future were in our power to command it or are we simply allowing it to proceed as it should? To me, scores are more akin to tarot cards than science. Tarot cards have just as complex an interpretive system surrounding them as testing does. And whether we're talking tarot or psychic readings, what the subject brings to the table is most important. The future is just extrapolation from the past coupled with wishes and expectations; isn't a student's past a better thing to base our predictions on?
Subscribe to:
Posts (Atom)