It depends on the skill, some are more complicated than others. I would say never more than 10 minutes. This is where the check for understanding is key: sometimes I'll see it's all one common error, clarify that, and then we're off, so something like 6 minutes total. That's often the case toward the end of the sequence. If I were to generalize I would say allow 10 minutes for day 2, and then more like 5-7 minutes after that. For some quick skills you can also get away with less than 5 minutes of practice to save more time.
I give my students a written outline of the steps and they can use it for as long as they want. Eventually they all stop using it and in the mean time I can keep moving on. I've never understood the rush to remember. Rehearsal over time is much more effective than short term cramming. I also ask the students to tell me how they are going to remember it. They have to come up with their own strategy and assess how well it works. This develops their metacognition. I give them strategy ideas to try if they are not sure.
I am a homeschooling parent, so there’s some slightly different things that are possible with an individual student, but I’ve found that the approach you outline in “Strands” combined with this more practice-oriented approach can lead to some fantastic progress.
It seems to really boost confidence and motivation. It sounds weird, but I liken it to tasting food. Kids don’t like strange tasting food. When you use this strand-based scheduling and kids are constantly getting a “taste” of certain types of problems, they develop a taste for them and actually start to like doing them… and if you do something where you gradually drip in more difficult variations of the same problem, kids (or at least, mine) start to be curious and excited to notice differences.
Yea there’s something about the motivation side of this strategy that’s hard to get across without experiencing it yourself. This is how I approach Spanish learning — I’m working on my Spanish skills and when I practice I spend a few minutes on vocab, a few minutes on listening, a few minutes on speaking, a few minutes on grammar. I’m much more motivated to do that than I am to spend a large chunk of time on one of those topics.
Hi, would you equate something like this to when teaching phonics skills making sure kids are reading connected text on subsequent days that is more consistent with all types of words, but then maybe going back to something more controlled for automaticity building and fluency work? Like switching between an article and a decodable that might focus on a past skill?
I would connect this more to Engelmann's approach to a high success rate and introducing concepts over multiple lessons. The goal is a high success rate, and the method is repeated practice in relatively small chunks over multiple days -- "repeat until firm" is a common phrase in DI. So I think the switch between connected text and decodable would be based on success rate while also making sure to revisit past skills. The delay plays a role here too -- for a given phonics skill, can the student do it on a "cold read" the next day? That's the best measure of whether to increase text complexity.
I do think early literacy instruction is just more complex here. In math we can break skills down into nice little chunks that are largely separate from one another, but with phonics all of the skills need to be blended and interleaved together. So the transition isn't as neat and tidy. I know you're a walk to read school which I think is relevant. Walk to read helps to put each student in a place where they can achieve a high success rate while also moving forward with students who are ready for more complexity. Curious if that makes sense, this is a little outside my wheelhouse.
Yes, definitely makes sense. I see that a lot when I do Reading Mastery lessons...the high success rate and breaking skills into very small digestible pieces. I have seen success in both ways using EBLI and Reading Mastery, and I'm just trying to fit more pieces together in my head. I would agree that the reading piece seems a bit more complex and woven together, but in Walk to Read we do try to match to a specific skill deficit, but I feel like we try to use box curriculum for interventions and I'm always wondering if it would be more useful to just work on the specific skill that shows a deficit or because literacy is made up of so many intertwined strands is integrating the more effective approach. Your work is so tangible and reader-friendly, and obviously, you have a deep knowledge base, so I just thought I would ask your thoughts. Thanks for the reply!
I have changed my thinking recently on this, in part because of your post on First-time correct. For the majority of skills now, on the second practise session I won't ask them to do it on Mini-whiteboards initially - if you expect that the success rate will be 30% (or, for that matter, less than say 70%), the con of students practising doing it wrongly is not worth taking, both for the time taken but also them developing harder-to-dislodge misrules. Instead, I prefer either a re-explanation or a hinted hands-up question (controversial I know), then choose a student who you expect to get it right and give a v good answer (not an indicator student). Then quick example then whiteboard question then practise session. Then ideally homework for the following class which features questions of that type. Then 3rd a classic MWB CFU.
In Theory of Instruction, Engelmann suggests using an initial teaching sequence in two consecutive sessions, elaboration(/practise here) on the second and 3rd, then possibly integration on 3rd and 4th (i.e. interleaved practise here). (From memory - though it may be even more prolonged than that)
Yea that makes sense, I think two days for initial instruction is probably a good rule of thumb. I will say that as I have gotten better at the atomisation side of things, when I do a good job isolating a small atom and sequencing it well, I can get above 70% the next day. But that's a minority, for most larger-grain-size skills I think multiple rounds of instruction should be expected. Of course Engelmann knew this decades ago and we're here figuring it all out again.
Glad you liked the first time correct post, it's one of my favorites and somehow it's the least-liked post I've written in the last few months.
I like your idea about strands. I’ve thought about reorganizing eighth grade math to teach the very basics of each unit in the first two or three months, then spend the rest of the year diving deeper. It feels like it would take forever to figure out how to pull it off.
But I think it would be helpful for students to have to think each day about the relationships in the problem instead of leaning on “I know this is the Pythagorean Theorem unit so I’m going to have to use a squared plus b squared equals c squared”
But I also think that might be pretty challenging for students who don’t learn quickly. They might feel even more lost than in a traditional unit where they can rely a little bit more on what happened the day before.
I do something with a little more continuity than that. I end every class with a short (10 question) mixed practice assignment, so students do need to distinguish between different types of questions. But in general, I'm doing multiple strands that feel continuous.
I'm doing kindof random stuff right now because we're mopping up different skills students need more work with, but my class today:
This is day ~five evaluating expressions with negatives (think 2 - x^2 where x = 9 and similar stuff, practicing multiple skills at once). It was just a check for understanding, quick clarification on order of operations, then practice.
This is day two on fractions within fractions (can't format nicely but think (1/2) / (2/5) in a big fraction). Checked for understanding, still have a while to go, did some modeling and more checking for understanding, then practice.
Then this was our big introduction to volume of a prism. Walked through a series of questions building up to V = Bh, after some very simple volume questions yesterday.
Each of those strands builds off what we did the day before. So students do have to keep track of three strands at once, but it's not randomly interleaved all day every day.
I find students have responded really well. Working for an entire class period on one topic can feel repetitive, and frustrating if you feel like you don't get it. Mixing in multiple topics keeps the energy high. It's not perfect but I've been a huge fan.
For these day 2, 3, 4 practice sessions, how much total time typically? It said 5 minutes of practice, but more time total?
It depends on the skill, some are more complicated than others. I would say never more than 10 minutes. This is where the check for understanding is key: sometimes I'll see it's all one common error, clarify that, and then we're off, so something like 6 minutes total. That's often the case toward the end of the sequence. If I were to generalize I would say allow 10 minutes for day 2, and then more like 5-7 minutes after that. For some quick skills you can also get away with less than 5 minutes of practice to save more time.
I give my students a written outline of the steps and they can use it for as long as they want. Eventually they all stop using it and in the mean time I can keep moving on. I've never understood the rush to remember. Rehearsal over time is much more effective than short term cramming. I also ask the students to tell me how they are going to remember it. They have to come up with their own strategy and assess how well it works. This develops their metacognition. I give them strategy ideas to try if they are not sure.
I am a homeschooling parent, so there’s some slightly different things that are possible with an individual student, but I’ve found that the approach you outline in “Strands” combined with this more practice-oriented approach can lead to some fantastic progress.
It seems to really boost confidence and motivation. It sounds weird, but I liken it to tasting food. Kids don’t like strange tasting food. When you use this strand-based scheduling and kids are constantly getting a “taste” of certain types of problems, they develop a taste for them and actually start to like doing them… and if you do something where you gradually drip in more difficult variations of the same problem, kids (or at least, mine) start to be curious and excited to notice differences.
Yea there’s something about the motivation side of this strategy that’s hard to get across without experiencing it yourself. This is how I approach Spanish learning — I’m working on my Spanish skills and when I practice I spend a few minutes on vocab, a few minutes on listening, a few minutes on speaking, a few minutes on grammar. I’m much more motivated to do that than I am to spend a large chunk of time on one of those topics.
Hi, would you equate something like this to when teaching phonics skills making sure kids are reading connected text on subsequent days that is more consistent with all types of words, but then maybe going back to something more controlled for automaticity building and fluency work? Like switching between an article and a decodable that might focus on a past skill?
I would connect this more to Engelmann's approach to a high success rate and introducing concepts over multiple lessons. The goal is a high success rate, and the method is repeated practice in relatively small chunks over multiple days -- "repeat until firm" is a common phrase in DI. So I think the switch between connected text and decodable would be based on success rate while also making sure to revisit past skills. The delay plays a role here too -- for a given phonics skill, can the student do it on a "cold read" the next day? That's the best measure of whether to increase text complexity.
I do think early literacy instruction is just more complex here. In math we can break skills down into nice little chunks that are largely separate from one another, but with phonics all of the skills need to be blended and interleaved together. So the transition isn't as neat and tidy. I know you're a walk to read school which I think is relevant. Walk to read helps to put each student in a place where they can achieve a high success rate while also moving forward with students who are ready for more complexity. Curious if that makes sense, this is a little outside my wheelhouse.
Yes, definitely makes sense. I see that a lot when I do Reading Mastery lessons...the high success rate and breaking skills into very small digestible pieces. I have seen success in both ways using EBLI and Reading Mastery, and I'm just trying to fit more pieces together in my head. I would agree that the reading piece seems a bit more complex and woven together, but in Walk to Read we do try to match to a specific skill deficit, but I feel like we try to use box curriculum for interventions and I'm always wondering if it would be more useful to just work on the specific skill that shows a deficit or because literacy is made up of so many intertwined strands is integrating the more effective approach. Your work is so tangible and reader-friendly, and obviously, you have a deep knowledge base, so I just thought I would ask your thoughts. Thanks for the reply!
I have changed my thinking recently on this, in part because of your post on First-time correct. For the majority of skills now, on the second practise session I won't ask them to do it on Mini-whiteboards initially - if you expect that the success rate will be 30% (or, for that matter, less than say 70%), the con of students practising doing it wrongly is not worth taking, both for the time taken but also them developing harder-to-dislodge misrules. Instead, I prefer either a re-explanation or a hinted hands-up question (controversial I know), then choose a student who you expect to get it right and give a v good answer (not an indicator student). Then quick example then whiteboard question then practise session. Then ideally homework for the following class which features questions of that type. Then 3rd a classic MWB CFU.
In Theory of Instruction, Engelmann suggests using an initial teaching sequence in two consecutive sessions, elaboration(/practise here) on the second and 3rd, then possibly integration on 3rd and 4th (i.e. interleaved practise here). (From memory - though it may be even more prolonged than that)
Yea that makes sense, I think two days for initial instruction is probably a good rule of thumb. I will say that as I have gotten better at the atomisation side of things, when I do a good job isolating a small atom and sequencing it well, I can get above 70% the next day. But that's a minority, for most larger-grain-size skills I think multiple rounds of instruction should be expected. Of course Engelmann knew this decades ago and we're here figuring it all out again.
Glad you liked the first time correct post, it's one of my favorites and somehow it's the least-liked post I've written in the last few months.
I like your idea about strands. I’ve thought about reorganizing eighth grade math to teach the very basics of each unit in the first two or three months, then spend the rest of the year diving deeper. It feels like it would take forever to figure out how to pull it off.
But I think it would be helpful for students to have to think each day about the relationships in the problem instead of leaning on “I know this is the Pythagorean Theorem unit so I’m going to have to use a squared plus b squared equals c squared”
But I also think that might be pretty challenging for students who don’t learn quickly. They might feel even more lost than in a traditional unit where they can rely a little bit more on what happened the day before.
I do something with a little more continuity than that. I end every class with a short (10 question) mixed practice assignment, so students do need to distinguish between different types of questions. But in general, I'm doing multiple strands that feel continuous.
I'm doing kindof random stuff right now because we're mopping up different skills students need more work with, but my class today:
This is day ~five evaluating expressions with negatives (think 2 - x^2 where x = 9 and similar stuff, practicing multiple skills at once). It was just a check for understanding, quick clarification on order of operations, then practice.
This is day two on fractions within fractions (can't format nicely but think (1/2) / (2/5) in a big fraction). Checked for understanding, still have a while to go, did some modeling and more checking for understanding, then practice.
Then this was our big introduction to volume of a prism. Walked through a series of questions building up to V = Bh, after some very simple volume questions yesterday.
Each of those strands builds off what we did the day before. So students do have to keep track of three strands at once, but it's not randomly interleaved all day every day.
I find students have responded really well. Working for an entire class period on one topic can feel repetitive, and frustrating if you feel like you don't get it. Mixing in multiple topics keeps the energy high. It's not perfect but I've been a huge fan.