Subscribe
Share
Share
Embed
Lesson by lesson podcasts for teachers of Illustrative Mathematics®.
(Based on IM 9-12 Math™ by Illustrative Mathematics®, available at www.illustrativemathematics.org.)
Alright. Let's jump right into it today. We're gonna be, unpacking residuals, you know, like, when your model's prediction is different from what actually happens.
Speaker 2:Yeah. Exactly. Like, you think your commute will be 30 minutes, but then, bam, traffic jam, you're stuck for an extra 10. That right there, that difference, that's a residual.
Speaker 1:A residual in action.
Speaker 2:Yeah.
Speaker 1:And speaking of action, this illustrative mathematics lesson we're diving into today really lays out how teachers can help students not just get what a residual I s, but how we use them to figure out if a linear model is even a good fit for the data.
Speaker 2:Totally. It's all about that goodness of fit. Right? But this lesson's really smart about about it doesn't just throw heavy statistics at the students.
Speaker 1:Yeah. It's gotta be more about building intuition at this stage, right, not full on statistical analysis just yet.
Speaker 2:You got it. It's all about helping students develop that gut feeling for what a residual means before they dive into the more, formal stuff later on.
Speaker 1:Love that. Meeting them where they are. So how does this lesson ease students into those tricky waters?
Speaker 2:Well, they start off with what they call a math talk, and it's really cool. Gets them thinking about differences without even mentioning residuals outright.
Speaker 1:Oh, sneaky. I like it. Give us an example. What kind of stuff are they doing in this math talk?
Speaker 2:They keep it nice and simple. They'll ask students to figure out the difference between, like, an estimate and an actual value. So maybe, like, how far off is saying there are 20 cars when you actually count 21?
Speaker 1:So they're already dipping their toes into the whole overestimating, underestimating thing, which is, like, the core of residuals.
Speaker 2:Exactly. And then from there, they bring back something familiar. The orange weight data, you might remember it from a previous lesson.
Speaker 1:Oh, yeah. The oranges. Right. So, like, we got those data points, 3 oranges weighing 1.027 kilograms, that sort of thing.
Speaker 2:That's the one. And this is where the lesson starts to bridge the gap, connecting those real physical oranges to this more abstract idea of residuals.
Speaker 1:Okay. So we've got our oranges. We've got students thinking about differences.
Speaker 2:Mhmm.
Speaker 1:How do the lesson writers bring this all together?
Speaker 2:This is where they really, hit the nail on the head. They don't just talk about residuals. They get the students right in there working with them, like, actually having them calculate the line of best fit for this orange data.
Speaker 1:Let me guess no dragging out the graph paper and rulers for this. Right?
Speaker 2:You know it. The lesson plan actually encourages using technology for this part.
Speaker 1:Smart move. What's the student's focus on what matters, like the meaning of the residuals, instead of getting lost in the calculations?
Speaker 2:Exactly. Work smarter, not harder. Right. And it goes even further. The lesson gets students thinking about how accurate those calculations trees.
Speaker 2:You know? Like, just memorizing steps
Speaker 1:without really understanding why. Right.
Speaker 2:So they point out that, hey.
Speaker 1:The orange weights are given to the 1,000ths place. Right? Yeah.
Speaker 2:So the lesson's like, hey, let's round the slope and intercept of that line of best fit to the 1,000ths place too.
Speaker 1:Those little details can make a big difference. But I'm guessing the moment comes when the lesson gets to actually calculating the residuals themselves.
Speaker 2:100%. And they make it so clear, breaking it down step by step. They use that example, the 3 oranges, and have students compare the real weight to what their linear model predicts. Boom. First residual right there.
Speaker 1:Nothing like a good concrete example right now.
Speaker 2:Makes all the difference. Takes it from theory to something real. But get this, they don't stop there. They have the students plot those residuals on a graph too.
Speaker 1:Visualization. Yeah. So important for this kind of thing.
Speaker 2:Right. Seeing's believing, as they say, helps students connect those dots literally between the scatterplot, the line of best fit, and whether a residual is positive or negative.
Speaker 1:Okay. So we've seen how the lesson flows. But even with the best lesson plan, students can still hit some snags. Right? Like, are there any common misconceptions about residuals that teachers should be ready for?
Speaker 1:You said it. It's like they can make or break a lesson sometimes. Knowing what to expect. You know? So when it comes to residuals, what are some of those misconceptions teachers should watch out for?
Speaker 2:Well, one that pops up a lot is, students struggling to connect that positive or negative sign on the residual with whether the model's overestimating or underestimating.
Speaker 1:Seems so obvious when you say it like that, but
Speaker 2:Yeah.
Speaker 1:Yeah. I could see how that could trip them up. What's a good way to get around that?
Speaker 2:Honestly, that visual representation is key. Like, really hammering home that connection between the scatterplot, the line of best fit if the point's above the line, model underestimated positive residual. Below the line, overestimated negative residual.
Speaker 1:Gotcha. Anchor their understanding in that visual. Love it. Anything else teachers should be ready for?
Speaker 2:Yeah. There's this other one that's a little trickier. See, sometimes students, they get it in their heads that a small residual automatically means the model's a good fit. Like, they don't look at the overall pattern of the residuals, just the size.
Speaker 1:Oh, interesting. So even if all the individual residuals are small, it could still be misleading.
Speaker 2:Exactly. Think about it like this. You could have data that's clearly curved. Right? But you try to force a straight line onto it.
Speaker 2:You might end up with residuals that are all pretty small, but they're consistently above the line for a bit then consistently below.
Speaker 1:Oh, yeah. So the pattern itself is a red flag even if the individual residuals aren't that big.
Speaker 2:Bingo. And that right there, that's what makes residuals so powerful. It's not just about the numbers themselves, but what they're telling us, you know, about how well that line really fits the data.
Speaker 1:It's like looking for those clues, seeing the bigger picture rather than getting stuck in the weeds.
Speaker 2:100%. And the lesson does a great job encouraging that kind of thinking. Like, having students ask, are these residuals mostly close to 0? Is there a pattern here, or is it just random? Those are the $1,000,000 questions.
Speaker 1:And by teaching students to ask those questions, we're giving them a tool they can use way beyond this lesson, even beyond math class. Right?
Speaker 2:Oh, absolutely. This whole idea of looking at data with a critical eye, understanding how models work, it's everywhere. Economics, science, even just understanding trends in the world, this stuff matters.
Speaker 1:Couldn't agree more. We're not just teaching math here. We're teaching them how to think. Alright. So we've covered a ton of ground today.
Speaker 1:Big thanks to Illustrative Mathematics for this awesome lesson plan on residuals. Really makes you think.
Speaker 2:For sure. Until next time.
Speaker 1:Keep those minds curious, everybody.