Know that straight lines are widely used to model relationships between two quantitative variables. For scatter plots that suggest a linear association, informally fit a straight line, and informally assess the model fit by judging the closeness of the data points to the line.
The mathematical purpose of this activity is for students to be able to visually assess the best line that fits data among a set of choices. Students are given a scatter plot and two lines that may fit the data. Students must select the line that best fits the data. The given lines address many common errors in student thinking about best-fit lines, including going through the most points, dividing the data in half, and connecting the points on both ends of the scatter plot.
Listen for students using the terms "slope" and "-intercept."
Launch
Provide students access to the images. Give students 2 minutes of quiet time to work on the questions.
Activity
None
Student Task Statement
Which of the lines is the best fit for the data in each scatter plot? Explain your reasoning.
A scatterplot. Horizontal, from 40 to 90, by 10's, labeled wins. Vertical, 0 to 5, by 0 point 5’s, labeled number of runs allowed per game. 27 dots trend linearly downward and to the right. A dashed, trending linearly upward to the right, a solid line trending linearly downward and to the right. Dot 1 at 42 comma 4 point 5, above solid line, dot 27 at 86 comma 1 point 6, on solid line. 13 dots above solid line, 12 dots below solid line. 16 dots below dashed line, 10 dots above dashed line.
A scatterplot. Horizontal, from 30 to 90, by 10's, labeled average score from 10 surveys. Vertical, 0 to 50, by 5’s, labeled average amount spent on dinner. 27 dots trend linearly upward and to the right. A dashed and solid line of best fit. Dot 1 at 40 comma 17, dot 27 at 88 comma 40, with dashed line. Dashed line passes through 10 points, with remaining points above the line. Solid line passes through 4 points, with 10 points above the line and 13 points below the line.
A scatterplot. Horizontal, from 50 to 75, by 5's, labeled price of crude oil per barrel. Vertical, 2 point 5 to 3 point 8, by 0 point 1’s, labeled cost of gas at local station. 27 dots trend linearly upward and to the right. A dashed and solid line of best fit. Dot 1 at 50 comma 2 point 5 at the bottom left end of the dashed line. Dashed line passes through 10 dots. Solid line passes thorugh 3 points, with 10 points below the line and 15 points above the line.
A scatterplot. Horizontal, from 0 to 70, by 10's, labeled number of floors. Vertical, 300 to 900, by 50’s, labeled height of building, feet. 27 dots trend linearly upward and to the right. A dashed and solid line of best fit. Dot 1 at 25 comma 319, Dot 27 at 60 comma 807, with solid line of best fit, passing through 5 additional points. Dashed line passes through 14 points.
Activity Synthesis
The purpose of this discussion is to understand bad fit, good fit, and best fit. In each scatter plot, the solid line represents the line of best fit—except for the last two graphs, for which the dashed line is the best fit.
Ask a student who uses the term "slope" while working the questions, “Can you explain the relationship between the two lines in the plot of runs and wins using the concept of slope?” (The slope of the dashed line is positive, and the slope of the solid line is negative.)
Ask a student who uses the term "-intercept," “Can you explain the significance of the -intercept in the question about average survey scores and amount spent on dinner?” (The solid line will have a -intercept less than the -intercept for the dashed line. Because the two lines have approximately the same slope, they appear parallel in the scatter plot.)
If time permits, discuss questions such as:
“Is the dashed line shown with the scatter plot of the data for runs and wins a bad fit, good fit, or best fit?” (The line is a bad fit because it does not show the correct relationship between the variables. It shows that the value of increases as the value of increases, rather than the value of decreasing as the value of increases.)
“Is the dashed line shown with the scatter plot of the data for oil and gas a bad fit, good fit, or best fit?” (It is the best fit because it is close to going through the middle of the data and follows the same trend as the data.)
“What factors helped you select the linear model that fits the data best?” (The line should go through the middle of the data, follow the trend of the data, and have a similar number of points on each side of the line.)
“What is a line of best fit?” (the best linear model for the data)
“How do you know that you have a line of best fit?” (We can use technology to generate the line of best fit, but then we need to graph it with the scatter plot to verify that it fits the data well. It needs to follow the trend of the data, it should go roughly through the middle of the data, and it should have roughly the same number of data points on both sides of it.)
“Which line fits the data better: the solid line or the dashed line? Do you think it is the line of best fit? Explain your reasoning.” (The dashed line fits the data better because its slope and vertical intercept more closely resemble the trend of the data than the slope and vertical intercept of the solid line. It is not the line of best fit because it does not go through the middle of the data. It should be a little lower on the graph.)
Student Lesson Summary
Some data appear to have a linear relationship, so finding an equation for a line that fits the data can help us understand the relationship between the variables.
Other data may follow nonlinear trends or not have an apparent trend at all.
When modeling data with a linear function seems useful, it is important to find a linear function that is close to the data. The line should have a -intercept and slope that follow the shape of the data represented by the scatter plot as much as possible.
Technology can be used to quickly find a line of best fit for the data and provide the equation of the line that we can use to analyze the situation.
Have feedback on the curriculum?
Help us improve by sharing suggestions or reporting issues.
The mathematical purpose of this activity is for students to:
Distinguish between linear and nonlinear relationships in bivariate, numerical data.
Informally assess the fit of a linear model.
Compare the slope and the vertical intercepts of different linear models.
Describe the relationship between two variables.
Students sort different scatter plots showing a linear model during this activity. A sorting task gives students opportunities to analyze representations, statements, and structures closely and make connections (MP2, MP7).
Launch
Arrange students in groups of 2, and distribute the pre-cut cards. Allow students to familiarize themselves with the representations on the cards:
Give students 1 minute to sort the cards into categories of their choosing.
Pause the class after students have sorted the cards.
Select groups to share their categories and how they sorted their cards.
Discuss as many different types of categories as time allows.
Attend to the language that students use to describe their categories and scatter plots, giving them opportunities to describe their scatter plots more precisely. Highlight the use of terms like fit, slope, and intercept. After a brief discussion, invite students to complete the remaining questions.
MLR8 Discussion Supports. Students should take turns finding a match and explaining their reasoning to their partner. Display the following sentence frames for all to see: “I noticed _____ , so I matched . . . .” Encourage students to challenge each other when they disagree. Advances: Speaking, Representing
Action and Expression: Internalize Executive Functions. To support organization, provide students with a template for organizing their observations. Provide a template or invite students to fold a blank piece of paper in thirds and label with three headers of “,” “,” and “linear model fits?” to collect their answers. Explain that in the first column, they will always write increasing, since they will be reading each graph from left to right, then ask them to fill in the behavior of the -values in the next column, and in the last column, their conclusions about whether a linear model fits well. Supports accessibility for: Language, Organization
Activity
None
Student Task Statement
Your teacher will give you a set of cards that show scatter plots.
Arrange all the cards in three different ways. Ensure that you and your partner agree on the arrangement before moving on to the next one. Sort all the cards in order from:
Best to worst for representing with a linear model.
Least to greatest slope of a linear model that fits the data well.
Least to greatest vertical intercept of a linear model that fits the data well.
For each card, write a sentence that describes how changes as increases and whether the linear model is a good fit for the data or not.
Student Response
Loading...
Building on Student Thinking
Activity Synthesis
The purpose of this discussion is for students to discuss the goodness of fit for linear models.
Here are some questions for discussion.
“How are scatter plots of A and F the same? How are they different?” (They have the same slope, and the linear model for each scatter plot is equally well fit. They are different because each of them has a different vertical intercept.)
“How do you know if a linear model is a good fit?” (We need to look at the scatter plot and the line of best fit and make a decision about whether or not the data follows a linear trend.)
“Why is the goodness fit for the linear model in scatter plot B better than the fit for the linear model in scatter plot A?” (The data in B falls on or very close to the linear model. The data in A is scattered around the line of best fit and has roughly the same number of values below the line of best fit as it does above the line of best fit.)
It is recommended to use the digital version of this activity. The mathematical purpose of this activity is for students to use technology to compute a line of best fit for data given in a table and to understand the meaning of the slope and -intercept.
In the digital version of the activity, students use an applet to create a scatter plot from data. The applet allows students to enter data into a table and accurately create a scatter plot, then drag given points to create a linear model to fit the data, and finally use the technology to find a least-squares regression line. The digital version may be helpful for quickly creating the scatter plot and increasing the accuracy of the plot and of fitting a line to the data.
If the digital version of the activity is not available, students should be guided through using available technology to find the least-squares regression line as the line of best fit.
Launch
Provide data tables for the graphs for the cards from the previous activity that fit well with a linear model. Assign one table to each group. For students using the paper task, show them how to use technology to create a scatter plot of the data in a table. After groups have had a chance to estimate the best-fit lines, pause the class. Show students how to use technology to find the least-squares regression line for data and display the line with the scatter plot.
Display the tables for students to use for the last question:
A. (card A in the previous activity)
1
2
2.2
4
3.3
5
3.3
4.5
3.6
6
3.8
6.5
3.9
5.7
4
7
4.4
6.5
4.5
7
4.7
7
4.8
6
4.9
8.7
5
7
5.1
7.7
5.2
6.7
5.5
8
5.5
8.5
6
9.5
6.6
8.6
7
9
7.7
10.313
B. (card B in the previous activity)
1
11.86
2.2
11.332
3.3
10.848
3.4
10.741
3.6
10.716
3.8
10.628
3.9
10.584
4
10.54
4.4
10.364
4.5
10.32
4.7
10.232
4.8
10.188
4.9
10.144
5
10.1
5.1
10.056
5.2
10.5
5.5
9.88
5.7
9.753
6
9.66
6.6
9.396
7
9.22
7.7
8.912
C. (card C in the previous activity)
1
6.11
2.2
7.142
3.3
8.088
3.5
8.19
3.6
8.346
3.8
2.92
3.9
8.604
4
8.69
4.4
9.034
4.5
9.12
4.7
9.292
4.8
13.6
4.9
9.464
5
9.55
5.1
9.636
5.2
9.722
5.5
9.98
5.8
10.32
6
10.41
6.6
10.926
7
11.27
7.7
11.872
D. (card E in the previous activity)
1
13.9
2.2
11.5
3.3
9.3
3.5
9.2
3.6
8.7
3.8
8.3
3.9
8.1
4
7.9
4.4
7.1
4.5
6.9
4.7
6.5
4.8
6.3
4.9
6.1
5
5.9
5.1
5.7
5.2
5.5
5.5
4.9
5.8
4.3
6
3.9
6.6
1.3
7
1.9
7.7
0.5
E. (card F in the previous activity)
1
6.5
2.2
8.5
3.3
9.5
3.3
9
3.6
10.5
3.8
11
3.9
10.2
4
11.5
4.4
11
4.5
11.5
4.7
11.5
4.8
10.5
4.9
13.2
5
11.5
5.1
12.2
5.2
11.2
5.5
12.5
5.5
13
6
14
6.6
13.1
7
13.5
7.7
14.813
Engagement: Develop Effort and Persistence. Provide prompts, reminders, or checklists that focus on increasing the length of on-task orientation in the face of distractions. For example, provide two copies of the steps: Graph the table, find the best-fit line, find the slope, and find the -intercept. Include phrases such as “as increases, . . .” to activate knowledge from prior activities. Supports accessibility for: Attention, Social-Emotional Functioning
Activity Synthesis
The purpose of this discussion is for students to make connections between the scatter plot and the equation of the line of best fit. Display each scatter plot, the line of best fit, and the equation of the line of best fit.
A.
B.
C.
D.
E.
Here are some questions for discussion:
“How does using technology help model the data represented by the scatter plot?” (It allows different people to come up with the same equation for the line of best fit. If the line is just drawn by hand, there can be different linear equations that seem to fit the data well, but there is only one “best” fit line.)
“What does the -intercept represent in each scatter plot? When is it reasonable to use this interpretation?” (It represents the value of estimated by the linear model when . When the intercept is near the range of the data, it can be reasonable to use this interpretation because otherwise, the linear trend may not continue. There are also some situations in which a value of 0 for does not make sense.)
“Why is the slope the same in scatter plot A and scatter plot F?” (It is the same because the data in scatter plot F is the same data as in scatter plot A, except that the values for have all been increased by 4.5 units.)
Tell students they should be careful when predicting values outside the range of the data, in particular, for the -intercept. Even when the data is fit well by a linear model, the behavior of the variables farther away may not be linear. It is important to remember that all predictions using the best-fit line are estimates and the reasonableness of the predictions should be considered.
MLR7 Compare and Connect. Lead a discussion comparing, contrasting, and connecting the different representations. Ask, “How are the slope and vertical intercept represented?” Advances: Representing, Conversing
Standards Alignment
Building On
Addressing
HSS-ID.C.7
Interpret the slope (rate of change) and the intercept (constant term) of a linear model in the context of the data.
The weight of ice cream sold in a day at a small store in pounds () and the average temperature outside during the day in degrees Celsius () are recorded in the table.
For this data, create a scatter plot and sketch a line that fits the data well.
Use technology to compute the best-fit line. Round any numbers to 2 decimal places.
What are the values for the slope and -intercept for the best-fit line? What do these values mean in this situation?
Use the best-fit line to predict the -value when is 10. Is this a good estimate for the data? Explain your reasoning.
Your teacher will give you a data table for one of the other scatter plots from the previous activity. Use technology and this table of data to create a scatter plot that also shows the line of best fit, then interpret the slope and -intercept.
Student Response
Loading...
Building on Student Thinking
Students may struggle with interpreting slope and -intercept. Remind students of how each relates to a situation. To help students interpret slope, ask them, “What does the -variable represent? What does the -variable represent? How is slope connected to the - and -variables? What happens to as increases (or decreases)?” To help students interpret the -intercept, ask them: “What does the point on the scatter plot mean? What are the coordinates of the -intercept? What do each of the coordinates mean in the situation described? What is the -value when is 0? Which variable has a value of 0? Which variable is represented with ?”