Lesson 7: Statistical Predictions Using Two Variables
Lesson 7: Statistical Predictions Using Two Variables
Objective:
Students will learn how to predict height using arm span data  and vice versa  visually on a scatterplot.
Materials:

Arm span vs. Height Scatterplot (LMR_4.6_Arm Span vs Height)
Note: This handout will be referenced in subsequent lessons.

Assorted color markers (dry erase or overhead)—See step 3 of lesson.

Overhead or LCD projector
Essential Concepts:
Essential Concepts:
When predicting values of a variable y  and if y is associated with x  then we can get improved predictions by using our knowledge about x. We essentially “subset” the data for a given value of x and use the mean y for those subset values. If the resulting means follow a trend, we can model this trend to generalize to asyet unseen values of x.
Lesson:

Remind students that in the previous lessons they were working with height data to predict the typical height of all the students at a large high school. They also used one method statisticians use to help them make a good prediction.

In addition to the height data, it turns out that each student’s arm span data was also collected and recorded.

Display the Arm Span vs. Height Scatterplot (LMR_4.6) on a white board or overhead projector (you will write on the board or the transparency later in the lesson—see step 9).
[LMR_4.6](../IDS_Curriculum_v_5.0/2_IDS_LMRs_v_5.0/IDS_LMR_Unit 4_v_5.0/LMR_4.6_Arm Span vs Height.pdf)

Distribute the Arm Span vs. Height handout (LMR_4.6). Students will also refer to this handout in a subsequent lesson.

In teams, ask students to analyze the plot and discuss the following questions:
• What kind of plot is this? Scatterplot.
• How many variables are displayed in this plot? Two variables.
• Which variable is shown on the xaxis? On the yaxis? Arm span is shown on the xaxis and height is shown on the yaxis.
• What is this plot showing? It is showing the relationship between a person’s height and the person’s corresponding arm span measurement.
• How can I find out the height of the person whose arm span measures 68 inches? Find 68 on the xaxis. Then find the data point located at 68. Place finger on the data point and track its location on the yaxis. The height is also 68 inches.

Using Talk Moves, conduct a class discussion of the questions in step 5.

Remind students that we’ve learned that the mean is the best way of predicting heights. The mean heights of these people is 64 inches.

Ask the students: Do you think we can do better? Is 64 a good prediction for someone whose arm span is 72”? What about 60”? How can you come up with a rule for determining the best predicted height if you know the person’s arm span?
Note to teacher: Lead students to realize that they can do this by “subsetting” the data for the fixed x value. For example, if arm span is 60, they should consider only the heights of people whose arm span is 60 and find the mean.

In teams, ask students to find the mean height for people whose arm span is 60, 64, 68, and 72. Then plot these points on the graph (we’ll use this later – the points should be roughly along a straight line). These arm spans have a range of height values associated with them. Students may take a mean of the heights, but answers may vary.

Ask students if they see any patterns or rules they can use from this to help with predictions. There were probably multiple answers from students since there were multiple height values associated with each arm span length. The goal now is to come up with a rule that suggests a plausible height value for anyone with a particular arm span.

A sentence starter to guide students: If a person has a bigger arm span, then we should predict… [a bigger height]. If time permits, you might push them to be more precise. Let’s take someone who has a 60 inch arm span. You predicted a height of . How much should we increase our prediction for people with a 62 inch arm span? Can you do this without subsetting the data and recalculating?

Conceptually, students are wrestling with the notion of the slope of the regression line but there’s no need to point this out just yet. Important: The equation of the line of best fit will be revealed in lesson 10.
Class Scribes:
One team of students will give a brief talk to discuss what they think the 3 most important topics of the day were.
Homework & Next 2 Days