4 ms·

AP Calculus isn't a good example of the "teach to the test" problem. Most elementary/middle and even high school tests aren't nearly as good as AP tests, and m

AP Calculus isn't a good example of the "teach to the test" problem. Most elementary/middle and even high school tests aren't nearly as good as AP tests, and most subjects aren't best taught through rote - the best way to learn US history is NOT to be quizzed repeatedly on which year the Missouri Compromise was passed or how many men died at Sharpsburg. The problem that "teach to the test" introduces is that students focus on factoids and identifying the correct answer in multiple choice, but in the real world (or in college), I'm less likely to have to pick WW2 generals out of a lineup and more like to need to come up with a few paragraphs on what caused it or how it was won or something. Now AP tests do (at least partially) solve for this with their free response and short answer sections, but Standards of Learning (or other state tests) may not.

dkarl 14y ago
most subjects aren't best taught through rote - the best way to learn US history is NOT to be quizzed repeatedly on which year the Missouri Compromise was passed or how many men died at Sharpsburg As far as I know, there aren't any subjects that are best taught through rote. (The confusion between rote learning and practicing skills is a pet peeve of mine.) Learning and testing don't need to resemble each other, not at all, not any more than a physical exam at the doctor's office needs to include a balanced meal, eight hours of sleep, and some light exercise. I think I established that pretty well in my comment: the way I was taught calculus was very different from how I was tested. I was taught calculus through lecture, demonstration, and practice, using applications to motive a new concept, covering the concept intuitively before learning the mechanics, and finally learning and practicing the relevant computational skills. My ex was the one who was taught according to the philosophy that learning should consist of mimicking the test over and over again. Most elementary/middle and even high school tests aren't nearly as good as AP tests.... Now AP tests do (at least partially) solve for this with their free response and short answer sections, but Standards of Learning (or other state tests) may not. In other words, standardized tests are capable of reflecting and encouraging progressive teaching ideals OR regressive fact-regurgitation models of learning. I have no doubt at all that poor tests can be written, have been used in the past, and are currently in use in some places. Tests can't help but reflect the beliefs and competence of the people who write them, but that applies to everything else in education. Nothing is immune to being screwed up by individual or institutional incompetence. If you assume competent, progressive teachers and incompetent, benighted test-writers, it's easy to conclude that tests will drag down educational standards. That isn't fair, though. There's no reason to assume that the test designers are any less enlightened and competent than the teachers manning the classrooms. in the real world (or in college), I'm less likely to have to pick WW2 generals out of a lineup and more like to need to come up with a few paragraphs on what caused it or how it was won or something I have to disagree with you there. That NEVER happens in the real world. In real life, you most commonly draw on historical knowledge to generate context for things you read or hear. You often have to rely on little factoids such as names and dates to generate connections between what you're reading and what you already know, because you might be hearing something from somebody who doesn't know the context themselves. (Or you might be reading something by a political pundit who is being selective with the facts.) Knowing dates is highly underrated. In a classroom, you learn that dates are the most sterile and inert form of historical knowledge. That may be true for academics or for students, but for laypeople applying a casual knowledge of history in the real world, dates come in really handy. They're the Kevin Bacon of historical connections.
- intended 14y ago
  > In other words, standardized tests are capable of reflecting and encouraging progressive teaching ideals OR regressive fact-regurgitation models of learning. You are putting the cart before the horse. Standardized tests in America circa 2004 allowed for that flexibility. You had an education system that allowed for progressive thought precisely because you weren't forced to standardize. as I have argued before. Your standardized test earlier, and your teaching system, was NOT comparable in ANY way to what Korea/Japan/India have. You are now recreating those systems with Standardized COMPULSORY tests. Your comparisons are superficial and invalid. > . If you assume competent, progressive teachers and incompetent, benighted test-writers, it's easy to conclude that tests will drag down educational standards. That isn't fair, though. There's no reason to assume that the test designers are any less enlightened and competent than the teachers manning the classrooms. You are painting a far too rosy picture of test taking limits. And part of that reason is because you aren't straying into the details. Tests can be only varied on a few axes - if you are testing about the revolt of 1857 or the various ways to calculate capacitance there is only a small subset of problems that you can come up with. Don't forget - These are STANDARDIZED tests - they have to be graded either through an answer key, or through grader judgement - if you allow too much on grader judgement then results aren't standardized. SO it ALWAYS skews to answer keys. This results in all tests becoming predictable. You do not have that many degrees of freedom when standardizing. > I have to disagree with you there. That NEVER happens in the real world. Studying/testing is never about real life. IT can't be. No one knows what sort of life a student is going to lead, so you can only hope to give everyone a similar base of information which should help them find their own paths eventually. As a result, data has to be produced in long form story format, and tested to see that key pieces of information were retained. Real life doesn't come into it.

dkarl 14y ago

most subjects aren't best taught through rote - the best way to learn US history is NOT to be quizzed repeatedly on which year the Missouri Compromise was passed or how many men died at Sharpsburg As far as I know, there aren't any subjects that are best taught through rote. (The confusion between rote learning and practicing skills is a pet peeve of mine.) Learning and testing don't need to resemble each other, not at all, not any more than a physical exam at the doctor's office needs to include a balanced meal, eight hours of sleep, and some light exercise. I think I established that pretty well in my comment: the way I was taught calculus was very different from how I was tested. I was taught calculus through lecture, demonstration, and practice, using applications to motive a new concept, covering the concept intuitively before learning the mechanics, and finally learning and practicing the relevant computational skills. My ex was the one who was taught according to the philosophy that learning should consist of mimicking the test over and over again. Most elementary/middle and even high school tests aren't nearly as good as AP tests.... Now AP tests do (at least partially) solve for this with their free response and short answer sections, but Standards of Learning (or other state tests) may not. In other words, standardized tests are capable of reflecting and encouraging progressive teaching ideals OR regressive fact-regurgitation models of learning. I have no doubt at all that poor tests can be written, have been used in the past, and are currently in use in some places. Tests can't help but reflect the beliefs and competence of the people who write them, but that applies to everything else in education. Nothing is immune to being screwed up by individual or institutional incompetence. If you assume competent, progressive teachers and incompetent, benighted test-writers, it's easy to conclude that tests will drag down educational standards. That isn't fair, though. There's no reason to assume that the test designers are any less enlightened and competent than the teachers manning the classrooms. in the real world (or in college), I'm less likely to have to pick WW2 generals out of a lineup and more like to need to come up with a few paragraphs on what caused it or how it was won or something I have to disagree with you there. That NEVER happens in the real world. In real life, you most commonly draw on historical knowledge to generate context for things you read or hear. You often have to rely on little factoids such as names and dates to generate connections between what you're reading and what you already know, because you might be hearing something from somebody who doesn't know the context themselves. (Or you might be reading something by a political pundit who is being selective with the facts.) Knowing dates is highly underrated. In a classroom, you learn that dates are the most sterile and inert form of historical knowledge. That may be true for academics or for students, but for laypeople applying a casual knowledge of history in the real world, dates come in really handy. They're the Kevin Bacon of historical connections.

intended 14y ago
> In other words, standardized tests are capable of reflecting and encouraging progressive teaching ideals OR regressive fact-regurgitation models of learning. You are putting the cart before the horse. Standardized tests in America circa 2004 allowed for that flexibility. You had an education system that allowed for progressive thought precisely because you weren't forced to standardize. as I have argued before. Your standardized test earlier, and your teaching system, was NOT comparable in ANY way to what Korea/Japan/India have. You are now recreating those systems with Standardized COMPULSORY tests. Your comparisons are superficial and invalid. > . If you assume competent, progressive teachers and incompetent, benighted test-writers, it's easy to conclude that tests will drag down educational standards. That isn't fair, though. There's no reason to assume that the test designers are any less enlightened and competent than the teachers manning the classrooms. You are painting a far too rosy picture of test taking limits. And part of that reason is because you aren't straying into the details. Tests can be only varied on a few axes - if you are testing about the revolt of 1857 or the various ways to calculate capacitance there is only a small subset of problems that you can come up with. Don't forget - These are STANDARDIZED tests - they have to be graded either through an answer key, or through grader judgement - if you allow too much on grader judgement then results aren't standardized. SO it ALWAYS skews to answer keys. This results in all tests becoming predictable. You do not have that many degrees of freedom when standardizing. > I have to disagree with you there. That NEVER happens in the real world. Studying/testing is never about real life. IT can't be. No one knows what sort of life a student is going to lead, so you can only hope to give everyone a similar base of information which should help them find their own paths eventually. As a result, data has to be produced in long form story format, and tested to see that key pieces of information were retained. Real life doesn't come into it.