If a few students students bomb a test, do you provide any way to get some points back, i.e. corrections, retakes, etc.? Or does the fact that there is another quiz coming soon help alleviate some of the stress of a low grade for students?
The answer here is sorta complicated by details too boring to name, but basically I used to do a whole big retake system as part of Standards Based Grading, and I eventually came to hate juggling all the retakes. Currently I'm not doing retakes. I don't have a great answer to this one, there are plusses and minuses either way.
I teach in the UK and cannot agree more - we too have a heavily test-based culture. Yet there is so much research that shows that frequent low-stakes recall of information of learnt information spaced out over time leads to better retention in memory. Plus quizzing is fun whereas (for most) testing is not.
Oh, I totally disagree. I don't like unit tests--the idea of cramming everything in and then giving a test and then never using it again. All of my tests are cumulative and I often integrate concepts. But quizzes are simply not enough to challenge your best kids. Surprised you'd think otherwise.
I think you're right to be always cumulative. You're also right to be worried about the degree of challenge. The way I make this work with my quizzes is trying to be very clear that "extra credit" is required to excel in the class. I make the distinction like this: everybody should be getting the first 3 or 4 questions right. But if you finish the first three, you need to try the extra credit before handing it in. Those questions are ones you SHOULD be getting right if you're excelling in the class.
I considered doing that back in the day, but here's the problems I see with that:
1) the parents *and* the kids want to do all the problems. And if the student doesn't know how to do the problem, the answer of "don't worry, that's extra credit" won't help because they want the fucking extra credit. If you say look, it's not extra credit it's just a problem to do if you can then eventually the conversation comes around to the kid realizing that the highest grade he can get without doing that problem is a B, or a C. And if you're saying oh, no, doing the normal quiz gets you an A then you're giving out a lot of As. I mean, if EVERYONE should get the first 3 or 4 questions right, then how hard is the quiz?
2) Can/Can't do isn't as binary as we'd like it. So the kids who you don't intend to be challenged by the hard problem might miss some points on the easier stuff, but might know how to do the set up on the extra credit and get some points that way.
But neither of these is my major objection. The major objection is that the quiz is a straightforward application of 3-5 days of learning with few surprises. A test needs to challenge everyone. So I design tests in such a way that I weight the easiest questions the most, and the hardest questions the least. But a test also must be more than just testing functional repetitive knowledge. It's got to challenge the kids to apply multiple ideas on the same situation.
I just recently tried something new because I only have one prep (which is a reason to shoot myself, frankly): I gave my lowest ability kids a different test entirely. It's the same format, uses the same answer sheet, but is just easier questions. My weak kids will be able to get legitimate Cs on the test, while the strong kids will have a different test entirely.
I totally agree with you on this. Unit tests in public high school have become completely pathologized. You have to reserve time to "review," "practice," "remediate" in some fashion, deal with individual emails about it, and in public school, you also have to plan for and implement all of the extra individualized supports, plans, and guidance. So every "unit test" eats up 3-5 days' worth of instructional time. This can be used in more effective, healthier ways.
I think it's so much better to just lower the temperature and do a lot of face-to-face formative assessment and spend time actually DOING mathematics.
At my school our common assessments are summative/final exams at the end of the semester, which does mean kids freak out about something worth a relatively high percentage of their total points for the term. But if they're doing everything they need to do all the way along, it provides a cushion against the freakout. And in 8 or so years, I haven't seen any difference in final exam results, so...
I like your every-other-week quiz idea. I may have to think about how to implement that at scale.
If a few students students bomb a test, do you provide any way to get some points back, i.e. corrections, retakes, etc.? Or does the fact that there is another quiz coming soon help alleviate some of the stress of a low grade for students?
The answer here is sorta complicated by details too boring to name, but basically I used to do a whole big retake system as part of Standards Based Grading, and I eventually came to hate juggling all the retakes. Currently I'm not doing retakes. I don't have a great answer to this one, there are plusses and minuses either way.
I teach in the UK and cannot agree more - we too have a heavily test-based culture. Yet there is so much research that shows that frequent low-stakes recall of information of learnt information spaced out over time leads to better retention in memory. Plus quizzing is fun whereas (for most) testing is not.
Oh, I totally disagree. I don't like unit tests--the idea of cramming everything in and then giving a test and then never using it again. All of my tests are cumulative and I often integrate concepts. But quizzes are simply not enough to challenge your best kids. Surprised you'd think otherwise.
I think you're right to be always cumulative. You're also right to be worried about the degree of challenge. The way I make this work with my quizzes is trying to be very clear that "extra credit" is required to excel in the class. I make the distinction like this: everybody should be getting the first 3 or 4 questions right. But if you finish the first three, you need to try the extra credit before handing it in. Those questions are ones you SHOULD be getting right if you're excelling in the class.
I considered doing that back in the day, but here's the problems I see with that:
1) the parents *and* the kids want to do all the problems. And if the student doesn't know how to do the problem, the answer of "don't worry, that's extra credit" won't help because they want the fucking extra credit. If you say look, it's not extra credit it's just a problem to do if you can then eventually the conversation comes around to the kid realizing that the highest grade he can get without doing that problem is a B, or a C. And if you're saying oh, no, doing the normal quiz gets you an A then you're giving out a lot of As. I mean, if EVERYONE should get the first 3 or 4 questions right, then how hard is the quiz?
2) Can/Can't do isn't as binary as we'd like it. So the kids who you don't intend to be challenged by the hard problem might miss some points on the easier stuff, but might know how to do the set up on the extra credit and get some points that way.
But neither of these is my major objection. The major objection is that the quiz is a straightforward application of 3-5 days of learning with few surprises. A test needs to challenge everyone. So I design tests in such a way that I weight the easiest questions the most, and the hardest questions the least. But a test also must be more than just testing functional repetitive knowledge. It's got to challenge the kids to apply multiple ideas on the same situation.
I just recently tried something new because I only have one prep (which is a reason to shoot myself, frankly): I gave my lowest ability kids a different test entirely. It's the same format, uses the same answer sheet, but is just easier questions. My weak kids will be able to get legitimate Cs on the test, while the strong kids will have a different test entirely.
I totally agree with you on this. Unit tests in public high school have become completely pathologized. You have to reserve time to "review," "practice," "remediate" in some fashion, deal with individual emails about it, and in public school, you also have to plan for and implement all of the extra individualized supports, plans, and guidance. So every "unit test" eats up 3-5 days' worth of instructional time. This can be used in more effective, healthier ways.
I think it's so much better to just lower the temperature and do a lot of face-to-face formative assessment and spend time actually DOING mathematics.
At my school our common assessments are summative/final exams at the end of the semester, which does mean kids freak out about something worth a relatively high percentage of their total points for the term. But if they're doing everything they need to do all the way along, it provides a cushion against the freakout. And in 8 or so years, I haven't seen any difference in final exam results, so...
I like your every-other-week quiz idea. I may have to think about how to implement that at scale.