Chapter 3

2nd six weeks (2013-2014) AP Stats

MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY

Oct 7 8 9 10 11
Standard Normal Assessing Chapter 2 Review Chapter 2 Test
Calculations Normality

Quiz
HW: MC 11-14
Ch 2 Review #2
HW: Worksheet HW: 2.26,2.27 WS AP Problem Set
MC HW 1-10
14 15 16 17 18
Scatterplots Correlation Quiz 3.1 Linear Regression

HW: 3.38 (use data

HW: 3.2-3.4, 3.16, HW: 3.28,3.30,3.31, HW: AP Problem on page 127),3.40,
3.19,3.20, 3.21 3.34,3.37 Set 3.41, 3.52
21 22 23 24 25
LSRL and r-squared Residuals Quiz 3.3 Modeling Non-
Interpret Slope Modeling non-linear Linear Data (Natural
Data Logs)

HW: Log WS
HW: 3.45, 3.71 HW: 3.46-3.48, 3.61 HW: WS
ws - Slope HW: WS #2
28 29 30 31 November 1
Quiz 4.1 Power Models Chapter 3/4 Review Chapter 3/4 Test

HW: Chapter 3
HW: AP Problem Review
Set HW: WS HW: AP Problem Read Chapter 5
Set Sampling Designs
4 5 6 7 8
Relations in Interpreting Intro to Random 6 Weeks
Categorical Data Correlation and Digit Table Cumulative Test
Regression

Go over HW.
HW: 4.51-4.53, 4.58 HW: 4.33,4.37,4.38, HW: 5.1, 5.2, 5.7, Assign Project
4.42,4.44,4.76 5.8, 5.10

**All assignments are subject to last minute changes!!!

1
 Scatterplot Example

Boat Manatees Boat Manatees

Year Year
Registrations* Killed Registrations* Killed
1977 447 13 1984 559 34
1978 460 21 1985 585 33
1979 481 24 1985 614 33
1980 498 16 1987 645 39
1981 513 24 1988 675 43
1982 512 20 1989 711 50
1983 526 15 1990 719 47
* (in thousands)

(a) We want to examine the relationship between the number of powerboats and the number of
manatees killed. State the explanatory and response variables.

(b) Make a scatterplot of these data below. Include all the necessary elements required of any graph.
Interpret your graph by stating what it shows about the relationship between these variables.

2
Notes for Correlation:
Correlation is a ________________. It measures the ____________ and ______________of a linear
relationship between two variables (x,y) – denoted as “r”.

The correlation “r” is the average of the products of the standardized x’s and standardized y’s.

Standardizing removes the ______ and allows us to calculate “r” while combining unrelated
variables. It does not matter which variable you call x or y –_________________________.

Both variables must be quantitative.

How to Interpret r:

Positive “r”:

Negative “r”:

-Correlation falls between ____________. Values near ______indicate a very weak linear relationship.
-Because r uses a standardized values of the observations, r does not change when we change
units.
-Correlation only measures _______________of a linear relationship. It does not describe curved
relationships.
-Like mean and standard deviation, the correlation is strongly affected by ______________.
“r” is not a complete description of two variable data.

See page 141 and 145 for graphs.

Page 138 data from problem 3.20.
Put Guessed calories in L2 and Correct Calories in L1.
Run a 2VarsStat to get mean and standard deviation of both variables.
Plot the scatterplot on calculator.
Run a LinReg to see diagnostics. What is the strength of the relationship?
Remove 394 and 419 (outliers) and run the diagnostics again.
What did you notice about the “r”? Do outliers affect the correlation? Did the strength of the
relationship get stronger or weaker after the outliers were removed?

3
Note for Linear Regression:

A regression line is a straight line that describes __________________________

________________________________________________________________________.

We often use a regression line to ____________the value of _______.

_______________ unlike ___________________, requires that we have an explanatory and a response variable.

See graph on page 150.

The Least Squares Regression Line (LSRL) is a mathematical model suitable for a linear trend.

LSRL - of y on x is the line that makes the sum of squares of the vertical distance of the data points of the line as
small as possible. (error)

error =

See graph on page 151.

LSRL equation

______ is the predicted response; _________ is the y-intercept; ___________ is the slope.

To calculate the LSRL by hand we use the following:

b= a=

See example 3.9 page 153.

Interpreting slope for the LSRL: From the problem we did previously (Calories Guessed/Actual Calories). b=1.304

y Guessed calories 1.304

Slope (b) =  =
x Actual calories 1

So for every ___________________________________(state description),

there is a ________________________________(state description) on average.

The purpose of the LSRL is to _____________________________ in your prediction.

error = observed – predicted

Observed is denoted by _________ Predicted is denoted by ________

We want ____________ to be as close to y as possible. In order to keep distances

positive, when minimizing the distance _______________is squared to remove the negative values.

4
Facts about LSRL:

1._______________________________________________________________

2._______________________________________________________________

Implications of the slope b:

1. When r =____________, the predicted response _____ changes the same a x.

2. When _____________, the change in ______ is less than the change in x.

3. If you are graphing an LSRL by hand never use slope and the intercept, use two points that are spread
apart.

Least Squares Regression Line and Correlation

Ex. 3.12 page 132
1. What is the explanatory variable and response variable?

Input the explanatory variable in L1 and the response variable in L2. Plot the data in the
scatterplot on your calculator.
2. What do you notice about the data at first glance? Describe.

Run a 2VarsStat on the data.

3. What is the mean and standard deviation of both variables?

Generate the LinReg(a+bx) and paste the equation in Y1.

4. What is the equation of the line? What is the slope of the line?

5. What is the correlation coefficient? What is the relationship? Record the value of the r
squared. We will discuss this later.

6. Generate the LSRL by hand using the information obtained. How close is the line you
generated to the line calculated by the calculator?

7. Interpret slope in terms of the problem using the slope you generated.

8. Look at the graph again with the calculator generated LSRL imposed on the graph. Knowing
the correlation coefficient and seeing the LSRL graphed, did this change your answer in number
2 above?

5
6
Name ____________________________ Period _______________
Interpreting Slope and Y-intercept
in Least-Squares Regression

Given the following situations, identify the slope and the y-intercept. Then interpret each in the context
of the scenario.

1. The paper “Arthroscopic Distal Clavicle Resection for Isolated Atraumatic Osteolysis in Weight
Lifters” (American Journal of Sports Medicine [1998]: 191-195) gave the following regression
equation with x = age and y = days after arthroscopic shoulder surgery before being able to
return to weight-lifting:

y  5.05  1.272x

2. In the article: Reproductive Biology of the Aquatic Salamander Amphiuma tridactylum in

Louisiana” (Journal of Herpetology [1999]: 100 – 105), 14 female salamanders were studied. Using
regression, the researchers predicted y = clutch size (number of salamander eggs) from
x = snout vent length (in centimeters) as follows:

y  147  6.175x

3. Studies have shown that people who suffer a sudden cardiac arrest have a better chance of
survival if a defibrillator shock is administered very soon after a cardiac arrest. Let y = survival
rate (in percent) and x = call-to-shock time (in minutes). A regression analysis of 5 cardiac
situations where a defibrillator was used gave the following LSRL:

y  101.36  9.30x

7
4. Shells of mollusks function as a part of the skeletal system and as protective armor. It has been
argued that many features of these shells were the result of natural selection in the constant
battle against predators. The paper “Postmortem Changes in Strength of Gastropod Shells”
(Paleobiology [1992]: 367- 377) included scatterplots of data on y = breaking strength (in
newtons) and x = shell heights (in centimeters). The least-squares regression line for a sample of
n  38 hermit crab shells was:

y  275.1  244.9x

5. It has been observed that Andean high-altitude natives have larger chest dimensions and lung
volumes that do sea-level residents. Is this also true of Himalayan residents? The paper
“Increased Vital and Total Lung Capacities in Tibetan Compared to Han Residents of Lhasa”
(American Journal of Physical Anthropology [1991]: 341-351) reported on the results of an
investigation into this question. Included in the paper was a plot of lung capacity (y) in cubic
decimeters versus chest circumference (x) in centimeters. The study produced the following
LSRL:

y  4.54  0.1123x

8
Residuals Notes:

Residual__________________________________________________________________________

Residual =

The mean of a _____________________ is zero.

A residual plot allows us to assess how well a LSRL serves as a model for the data.

To graph a Residual Plot:

1. X – values are on the ____________axis.

2. The corresponding residuals goes on the _________axis.

3. They are graphed about the _______________.

See page 168 - 169 graph and data. In Class Example on your own paper.
To graph a Residual Plot by Hand:
L1 L2 L3
X Actual Y L2 – Y1(L1)

1. In Y1 you should have the LSRL.

2. Highlight the L3 and use the formula L2 - Y1(L1). This will compute the residuals (error).

3. Turn off Plot 1 and turn on Plot 2. Deselect the LSRL line in Y1.

4. Type Y = 0 in Y2. This is the reference line for the plot.

5. Graph the scatterplot using for the X – axis L1 and for the Y – axis use L3.
Zoom9. This will show the residual plot for the data.

You are looking for a randomly scattered set of points. No pattern and no curves. If either occurs
then you know that a LSRL is not the best model for the data.

See page 170 the graphs for different residual plots.

An outlier is an observation that lies outside the overall pattern of the other observations.
Outliers have large residuals (in the y-direction). Influential points are outliers in the x-
direction and have small residuals. An observation is “influential” if removing it would markedly
change the position of the regression line. Outliers and influential points both can be considered
“influential”. Also a point can considered an outlier and an influential point at the same time!

9
Residual plot example:

10
Name:___________________________ AP Statistics Ch 3 Review

Part I - Multiple Choice (Questions 1-10) - Circle the answer of your choice.

1. Foresters use regression to predict the volume of timber in a tree using easily measured quantities such as diameter. Let y be
the volume of timber in cubic feet and x be the diameter in feet (measured at 3 feet above ground level). One set of data gives

y = -30 +60x.
The predicted volume for a tree of 18 inches is:

(a) 1050 cubic feet

(b) 600 cubic feet
(c) 105 cubic feet
(d) 90 cubic feet
(e) 60 cubic feet

2. Consider the following scatterplot of midterm and final exam scores for a class of 15 students. Which of the following are true
statements? Exam Scores Scatter Plot
110
I. The same number of students scored 100 on the midterm 100
exam as scored 100 on the final exam.

Final_Exam _Score
90
II. Students who scored higher on the midterm exam tended
80
to score higher on the final exam.
70
III. The scatterplot shows a moderate negative correlation
60
between midterm and final exam scores.
50
40
(a) I and II
30
(b) I and III
20
(c) II and III
(d) I, II, and III 20 40 60 80 100
(e) None of the above gives the complete set of complete true responses. Mdterm_Exam_Score

3. Data are obtained for a group of college freshman examining their SAT scores (math plus verbal) from their senior year of high
school and their GPAs during their first year of college. The resulting regression equation is:

y  0.00161x  1.35 with s x  120 , and s y  .3057

What percentage of the variation in GPAs can be explained by looking at SAT scores?

(a) 0.161%
(b) 16.1%
(c) 39.9%
(d) 63.2%
(e) This value cannot be computed from the information given.

4. Suppose the correlation between two variables is r = 0.23. What will the new correlation be if 0.14 is added to all values of the
x-variable, every value of the y-variable is doubled, and the two variables are interchanged?

(a) 0.23
(b) 0.37
(c) 0.74
(d) -0.23
(e) -0.74

11
5. Given the least-squares regression line: [Cost of a Monopoly Property] = 67.3 + 6.78 * [Spaces From GO],
determine the residual for Reading Railroad which costs $200 and is 5 spaces from GO.
(a) –98.8
(b) –9.88
(c) 98.8
(d) –1418.3
(e) A residual has no meaning since one of the variables is categorical.

6. A study of the fuel economy for various automobiles plotted the fuel consumption
(in liters of gasoline used per 100 kilometers traveled) vs. speed (in kilometers per hour).
A least-squares regression line was fitted to the data and the residual plot is displayed to
The right. What does the pattern of the residuals tell you about the linear model?

(a) The evidence is inconclusive.

(b) The residual plot confirms the linearity of the data.
(c) The residual plot suggests a different line would be more appropriate.
(d) The residual plot clearly contradicts the linearity of the data.
(e) None of the above.

7. The coefficient of determination of the data described in the scatterplot is:

(a) 0.35
(b) 0.65
(c) -0.80
(d) 0.88
(e) 0

8. With regard to regression, which of the following statements about outliers are true?

I. Outliers have large residuals.

II. A point may not affect the regression equation even though its x-value is displaced in the x-direction and
its y-value is an outlier in the y direction.
III. Removal of an outlier will affect the regression line in a meaningful way.

(a) I and II
(b) I and III
(c) II and III
(d) I, II, and III
(e) None of the above gives the complete set of true responses.

9. As reported in the Journal of the American Medical Association (June 13, 1990), for a study of ten nonagenarians, the following
tabulation shows a measure of strength versus a measure of functional mobility

Strength (kg) 7.5 6 11.5 10.5 9.5 18 4 12 9 3

Walk time (s) 18 46 8 25 25 7 22 12 10 48

What does the slope of the least-squares regression line signify?

(a) The sign is positive, signifying a direct cause-and-effect relationship between strength and mobility.
(b) The sign is positive, signifying that the greater the strength, the greater the functional mobility.
(c) The sign is negative, signifying that the relationship between strength and functional mobility is weak.
(d) The sign is negative, signifying that the greater the strength, the less the functional mobility.
(e) The slope is close to zero, signifying that the relationship between strength and functional mobility is weak.

12
10. Consider the three points (2,11), (3,17), (and (4,29). Given any straight line, we can calculate the sum of the squares of the
three vertical distances from these points to the line. What is the smallest possible value this sum can be?

(a) 6
(b) 9
(c) 29
(d) 57
(e) cannot be determined

Part II – Free Response (Questions 11-13) – Show your work and explain your results clearly.
Data Scatter Plot

11. (a) Find the equation of the least-squares regression line of

5
the data represented in the scatterplot at the right.
4

Y
2

(b) Draw in the LSRL on the scatterplot using two points on either 0
extreme of the scatterplot. Record these points below.
0 1 2 3 4 5
X

2 2
(c) Calculate r and r . What does r describe?

12. An analysis of the relationship between the number of telephones in a household (x) and the annual family income(y) revealed
the following statistics:

x  3.8 y  65, 000 n  26

s x  1.2 s y  15, 500 r  0.65

(a) Determine the least-squares regression line.

(b) Determine coefficient of determination. Explain what this index means.

13
13. Lydia and Bob were searching the Internet to find information on air travel in the United States. They found
data on the number of commercial aircraft flying in the United States during the years 1990-1998. The dates
were recorded as years since 1990. Thus, the year 1990 was recorded as year 0. They fit a least squares
regression line to the data. The graph of the residuals and part of the computer output for their regression are
given below.
Aircraft Scatter Plot
^ 50
y  2939.93  233.517 x
r = 0.88 30

Res iduals
10
(a) Is a line an appropriate model to use for these data? What information tells
you this? -10

-30

-50
1 2 3 4 5 6 7 8 9
Years_Since_1990

(b) What is the value of the slope of the least squares regression line?
Interpret the slope in the context of this situation.

(c) What is the value of the intercept of the least squares regression line?
Interpret the intercept in the context of this situation.

(d) What is the predicted number of commercial aircraft flying in 1992?

(e) What was the actual number of commercial aircraft flying in 1992?

14
15
16