4.1 4.2 4.3 4.

4.1 Vitruvian Man

Relating Body Measurements

More than 2,000 years ago, a Roman architect and writer named Vitruvius
found patterns by relating two body measurements. He claimed a person’s
arm span is equal to his or her height.

• Do you think the relationship between arm span and height

applies to the students in your class?
• How would you display and analyze data collected to test the
claim made by Vitruvius?

Problem 4.1
The table shows the height and arm span of students in a CMP class.

Height (in.) 56 57 57 58 59 60 60 60 62 64 64 66 67 67 67 68
Arm span (in.) 54 57 54 61 56 58 59 60 62 63 62 62 65 67 69 67

Do you think the data support the claim that arm span and
height are about equal?

A Analyze the data to test your ideas.

1. Plot the (height, arm span) data on a coordinate graph. The resulting
graph is called a scatter plot.
2. Do you think the scatter plot supports the claim that arm span and
height are about equal for most people?
continued on the next page >
2. Yes the data seems to support the claim that
they are equal.

82 Thinking With Mathematical Models

 4.1 4.2 4.3 4.4

Problem 4.1 continued

3. If each student in the class had arm span s equal to height h, what
equation would relate the two variables? h = s
(57,57)
a. Graph the equation on your scatter plot.
(60,60)
(62,62) b. Which data points (if any) does your line pass through? Explain
(67,67) The arm how arm span and height are related in those points.
span and the
height are the c. Choose several data points that are not on your line. Explain how
same. arm span and height are related in each case. How do you describe
the relationship shown on the graph? Arm span is greater than
height.
B The tallest person in recorded history was Robert Pershing Wadlow. Height is greater
At age 22, he was 8 feet 11.1 inches (272 cm) tall. His arm span was than the arm span.
9 feet 5.75 inches (289 cm).
1. Where would you plot the point (height, arm span) for Robert
Wadlow? Would the point be on, above, or below the line you drew
in Question A, part (3)? Above the line because his arm span is
greater than his height.
2. Does the data point for Robert Wadlow support the claim that arm
span and height are roughly equal? No because the difference is 17 cm
which is a significant difference.
C The accuracy of fit for a linear model is measured by calculating errors from
the model. These errors, called residuals, are the differences between the
actual data and what the model predicts.
1. Find the arm span residuals (actual arm span - predicted arm span)
using the model s = h for the CMP class data.

Height (inches) 56 57 57 58 59 60 60 60 62 64 64 66 67 67 67 68
Arm span (inches)
Actual 54 57 54 61 56 58 59 60 62 63 62 62 65 67 69 67
Predicted by Model 56 57 57 58 59 60 60 60 62 64 64 66 67 67 67 68
Residual −2 0 −3 ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■

2. To see if there is any pattern in the residuals it helps to plot the (height,
residual) data. Use such a plot and data in the table to describe the
pattern (if any) in the residuals. Then explain why you think the
equation s = h is or is not an accurate model for predicting arm span
from height. The equation is not perfect. There are many
residuals that are negative suggesting that the
model is a bit high. continued on the next page >

Investigation 4 Variability and Associations in Numerical Data 83

 4.1 4.2 4.3 4.4

Problem 4.1 continued

D The dinosaur Tyrannosaurus rex grew to 20 feet in height with an arm span
of about 10 feet.
1. Do you think the T. rex data point fits the pattern that arm span and
height are roughly equal? Explain. No, because the arm span is half
of the dinosaur's height.
2. If you plot the data point, would it be on, above, or below the line you
drew in Question A, part (3)? It would be below the line
because the height is greater
than the arm span.

Homework starts on page 96.

4.2 Older and Faster

Negative Correlation

Magnolia Elementary is a school with students who are 5 to 14 years old.

One field day, all students were timed in a 100-meter race. The table shows
data for some of the students.

Student Age
5 5 6 8 8 8 9 9 10 10 10 11 11 12 13 13 14
(years)
Race Time
25 22 23 18 16 17 15 16 17 20 14 15 13 14 17 12 13
(seconds)

84 Thinking With Mathematical Models

 4.1 4.2 4.3 4.4

The graph below shows the data from the table and a line that models
the data.

Race Time and Age

30 y

Race Time (seconds)

25
20
15
10
5
x
0
0 2 4 6 8 10 12 14 16
Student Age (years)

• How would you describe the relationship between age and

race time?
• Would you say the relationship is strong or weak?
• Are the data points close to the line or spread out?

Problem 4.2
Use the Race Time and Age graph.

A The line drawn on the graph models the relationship between age and
race time.
The approximate slope is
1. What is the approximate slope of the line? -5/4.
2. How does the slope help you understand the relationship between age
and race time? Your race time gets down
as you get older.
3. Do you think it makes sense to predict a race time for a 7-year-old
student using the line? If so, what do you predict for a 7-year-old? How
confident are you in your prediction? Yes, a 7 year old is within the rang of graph. I
predict around 19 second for the race time.
4. Do you think it makes sense to predict a race time for a 21-year-old
person using the line? If so, what do you predict for a 21-year-old?
How confident are you in your prediction? No, because the graph predicted that
a 21 year old would have a race time
of around 5 second which does not
match since the world record is only
around 9.5 seconds.

Investigation 4 Variability and Associations in Numerical Data 85

 4.1 4.2 4.3 4.4

Problem 4.2 continued

B Some data points are very close to the line while others are far from it. The
points far from the line don’t seem to fit the model.
1. Find two points that don’t seem to fit the model. What are their
coordinates (age, race time)? (5,25) and (10,20)
2. Why do you think the points don’t match the overall pattern? Explain.
Think about the relationship between race time and age. People do not run at the
same speed.
3. In Problem 4.1, you used a line to model (height, arm span).
a. If a 6-foot-9-inch NBA basketball player has a 7-foot-5-inch arm
span, would that data point fit the model? No, his arm span is 8 inches
longer than his height.
b. Would you plot the data point, on, above, or below the s = h line?
Explain. I would plot it above the line because his
arm span is greater than height.
C The table and graph show age and grade point average (GPA) for
14 students at Magnolia High School.

Student Age
14 14 14 15 15 15 15 16 16 17 17 17 18 18
(years)
GPA 2.5 3.2 3.8 1.8 2.6 3.7 1.2 1.6 3.9 2.0 3.4 3.5 1.9 3.0

GPA and Age

4 y
3.5
3
2.5
GPA

2
1.5
1
0.5
x
0
13 14 15 16 17 18 19
Age (years)

1. Are age and GPA strongly related for these students? Explain.
No, they're not strongly related because GPA vary greatly.
2. How is your answer to part (1) supported by the table?
For the same ages of students we have a wide range of GPAs.
3. How is your answer to part (1) supported by the scatter plot?
The dots are spread apart. There's no real pattern.

Homework starts on page 96.

86 Thinking With Mathematical Models