Introduction to the Practice of Statistics

9th Edition Moore Test Bank

Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/introduction-to-the-practice-of-statistics-9th-edition-moore-test-bank/
1. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

Determine whether each of the following statements is true or false.

A) The explanatory variable in this study is the average number of putts per hole.
B) The value of the correlation between 1993 winnings and average number of putts per hole
is 0.285.
C) If the average number of putts per hole increases by 1, we estimate the winnings in 1993
to decrease by $4139.20.
D) We predict the 1993 winnings for a golf pro with an average of 1.75 putts per hole to be
$653,600.00.

Page 1
2. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

The quantity s = 281.8 is an estimate of the standard deviation  of the deviations in the
simple linear regression model. What are the degrees of freedom for this estimate?
A) 66
B) 67
C) 68
D) 69

Page 2
3. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

What is the value of the intercept of the least-squares regression line?

A) 1698.4
B) 3023.8
C) –4139.2
D) 7897.2

Page 3
4. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

Suppose the researchers conducting this study wish to test the hypotheses H0: 1 = 0
versus Ha: 1< 0. What is the value of the t statistic for this test?
A) –2.44
B) –1.91
C) 2.44
D) 2.61

Page 4
5. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

What is an approximate 95% confidence interval for the slope 1?

A) –4139.2 ± 1698.2
B) –4139.2 ± 3396.4
C) 7897.2 ± 3023.8
D) 7897.2 ± 6047.6

Page 5
6. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

Which of the following conclusions seems most justified?

A) There is no evidence of a relation between the average number of putts per round
and the 1993 winnings of PGA tour pros.
B) There is distinct evidence (P-value less than 0.05) that there is a positive
correlation between 1993 winnings and average number of putts per round.
C) There is some evidence that PGA tour pros who averaged fewer putts per round
had higher winnings in 1993.
D) The presence of strongly influential observations in these data makes it impossible
to draw any conclusions about the relationship between 1993 winnings and average
number of putts per round.

Page 6
7. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

Suppose the researchers conducting this study wish to estimate the 1993 winnings when
the average number of putts per hole is 1.75. The following results were obtained from
statistical software.

Predicted winnings Standard error 95.0% C.I. 95.0% P.I.

653.6 61.62 (530.6, 776.6) (77.7, 1229.4)

If the researchers wish to estimate the mean winnings for all tour pros whose average
number of putts per hole is 1.75, what would be a 95% confidence interval for the mean
winnings?
A) (77.7, 1229.4)
B) (530.6, 776.6)
C) 653.6 ± 61.62
D) 653.6 ± 123.24

Page 7
8. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

Suppose the researchers conducting this study wish to estimate the 1993 winnings when
the average number of putts per hole is 1.75. The following results were obtained from
statistical software.

Predicted winnings Standard error 95.0% C.I. 95.0% P.I.

653.6 61.62 (530.6, 776.6) (77.7, 1229.4)

If the researchers wish to estimate the winnings for a particular tour pro whose average
number of putts per hole is 1.75, what would be a 95% prediction interval for the
winnings?
A) (77.7, 1229.4)
B) (530.6, 776.6)
C) 653.6 ± 61.62
D) 653.6 ± 123.24

Page 8
9. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

The following (partial) ANOVA table was obtained from statistical software.

Source DF Sum of squares

Model 1 471,605
Error 67 5,319,690

What is the value of the SST, the total sum of squares?

A) 79,398
B) 471,605
C) 5,319,690
D) 5,791,295

Page 9
10. There is an old saying in golf: “You drive for show and you putt for dough.” The point
is that good putting is more important than long driving for shooting low scores and
hence winning money. To see if this is the case, data on the top 69 money winners on
the PGA tour in 1993 are examined. The average number of putts per hole for each
player is used to predict the total winnings (in thousands of dollars) using the simple
linear regression model

(1993 winnings)i = 0 + 1(average number of putts per hole)i + i,

where the deviations i are assumed to be independent and Normally distributed with a
mean of 0 and a standard deviation of . This model was fit to the data using the method
of least squares. The following results were obtained from statistical software.

R2 = 0.081
s = 281.8

Variable Parameter estimate Standard error

Constant 7897.2 3023.8
Average putts –4139.2 1698.4

The following (partial) ANOVA table was obtained from statistical software.

Source DF Sum of squares

Model 1 471,605
Error 67 5,319,690

What is the value of the F statistic for testing the hypotheses H0: 1 = 0 versus Ha: 1 
0?
A) 0.089
B) 2.44
C) 4.72
D) 5.94

11. The statistical model for simple linear regression has the form yi   0  1 xi   i , i = 1,
2, …, n. Which of the following statements about this statistical model is/are TRUE?
A) The  i are assumed to be Normally distributed with a mean of 0 and a standard
deviation of .
B) The parameters of this model are  0 , 1 , and .
C)  0  1 xi is the mean response when x  xi .
D) The  i , i = 1, 2, …, n are independent.
E) All of the above are true.

Page 10
12. Which of the following statements about simple linear regression is/are FALSE?
A) The term “residual” refers to the difference between the observed response and the
predicted response using the regression equation.
B) If b0 and b1 are the intercept and slope, respectively, of the least-squares line,
yˆi  b0  b1 x* is an unbiased estimator of the mean response when x  x* .
C)
The estimate of  is given by s =
 ( y  yˆ )
i i
2

.
n 1
D) The e i sum to zero.

Page 11
13. The data referred to in this question were collected on 41 employees of a large
company. The company is trying to predict the current salary of its employees from
their starting salary (both expressed in thousands of dollars). The SPPS regression
output is given below as well as some summary measures:

What is an approximate 95% confidence interval for the slope 1?

A) (–7.57, 4.39)
B) (–4.52, 1.34)
C) (1.80, 2.41)
D) (1.95, 2.26)

Page 12
14. The data referred to in this question were collected on 41 employees of a large
company. The company is trying to predict the current salary of its employees from
their starting salary (both expressed in thousands of dollars). The SPPS regression
output is given below as well as some summary measures:

Suppose we wish to test the hypotheses H0: 1 = 2 versus Ha: 1  2. Together with an
insignificant constant in this model, this would imply that the employees currently earn
about twice as much as their starting salary. At the 5% significance level, would we
reject the null hypothesis?
A) Yes
B) No
C) This cannot be determined from the information given.

Page 13
15. The data referred to in this question were collected on 41 employees of a large
company. The company is trying to predict the current salary of its employees from
their starting salary (both expressed in thousands of dollars). The SPSS regression
output is given below as well as some summary measures:

What is the value of the estimate for ,the standard deviation of the model deviations i?
A) 0.15
B) 2.93
C) 7.21
D) 52.0

Page 14
16. The data referred to in this question were collected on 41 employees of a large
company. The company is trying to predict the current salary of its employees from
their starting salary (both expressed in thousands of dollars). The SPSS regression
output is given below as well as some summary measures:

John Doe works for this company. He started with a salary of $15,300. Predict his
current salary with a 90% confidence interval. Express the interval in the appropriate
units.
A) ($15,683; $45,537)
B) ($18,204; $43,015)
C) ($28,580; $32,640)
D) ($31,516; $32,885)

Page 15
17. The data referred to in this question were collected on 41 employees of a large
company. The company is trying to predict the current salary of its employees from
their starting salary (both expressed in thousands of dollars). The SPSS regression
output is given below as well as some summary measures:

How would a 90% confidence interval for the average current salary for all employees
who started with a salary of $15,300 compare to a 90% confidence interval for the
current salary of an individual with a starting salary of $15,300?
A) It would be narrower.
B) It would be the same.
C) It would be wider.
D) This cannot be determined from the information given.

Page 16
18. Which of the following statements about the simple linear regression model and its least
squares fit is/are FALSE?
A) The ANOVA table gives degrees of freedom, sum of squares, and mean squares for
the model, error, and total sources of variation.
B) The ANOVA F statistic is the ratio MSM/MSE.
C) The ANOVA F statistic has an F(1, n – 2) under the null hypothesis H0: 1 = 0.
D) The ANOVA F statistic is particularly useful because it tests the H0: 1 = 0
against the one-sided alternative Ha: 1 > 0.
E) The ratio SSM/SST is equal to r 2 .

19. Do heavier cars use more gasoline? To answer this question, a researcher randomly
selected 15 cars. He collected their weight (in hundreds of pounds) and the mileage
(MPG) for each car. From a scatterplot made with the data, a linear model seemed
appropriate. The following output was obtained from SPSS:

Determine whether each of the following statements is true or false.

A) The explanatory variable in this study is the weight of the car.
B) The value of the correlation between weight and mileage is 0.662.
C) An estimate for the standard deviation  of the deviations i in the linear regression
model is 6.275.
D) Approximately 43.8% of the variation in mileage is accounted for by the linear
relationship with the weight of the car.

Page 17
20. Do heavier cars use more gasoline? To answer this question, a researcher randomly
selected 15 cars. He collected their weight (in hundreds of pounds) and the mileage
(MPG) for each car. From a scatter plot made with the data, a linear model seemed
appropriate. The following output was obtained from SPSS:

What is the equation of the least-squares regression line?

A) ŷ = –0.52 x + 0.164
B) ŷ = 40.44 – 0.52x
C) ŷ = 40.44 + 6.28x
D) y = 40.44 + 0.52x

Page 18
21. Do heavier cars use more gasoline? To answer this question, a researcher randomly
selected 15 cars. He collected their weight (in hundreds of pounds) and the mileage
(MPG) for each car. From a scatter plot made with the data, a linear model seemed
appropriate. The following output was obtained from SPSS:

Which of the following descriptions of the value of the slope is the correct description?
A) We cannot interpret the slope because we cannot have a negative weight of a car.
B) We estimate the mileage to decrease by 0.521 miles per gallon when the weight of
a car increases by 1 pound.
C) We estimate the mileage to decrease by 0.521 miles per gallon when the weight of
a car increases by 100 pounds.
D) We estimate the mileage to decrease by 52.1 miles per gallon when the weight of a
car increases by 100 pounds.

Page 19
22. Do heavier cars use more gasoline? To answer this question, a researcher randomly
selected 15 cars. He collected their weight (in hundreds of pounds) and the mileage
(MPG) for each car. From a scatter plot made with the data, a linear model seemed
appropriate. The following output was obtained from SPSS:

To test if there is a significant straight-line relationship between the weight and the
mileage of a car, we can perform a t test. What is the value of the t statistic for this test?
A) t = –3.182
B) t = 6.445
C) t = 6.780
D) This cannot be determined from the information given.

Page 20
23. Do heavier cars use more gasoline? To answer this question, a researcher randomly
selected 15 cars. He collected their weight (in hundreds of pounds) and the mileage
(MPG) for each car. From a scatter plot made with the data, a linear model seemed
appropriate. The following output was obtained from SPSS:

What is a 95% confidence interval for the slope 1?

A) (–0.875, –0.167)
B) (–0.685, –0.357)
C) (–1.04, 0.001)
D) (26.89, 53.99)

24. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

What is approximately the value of the intercept of the least-squares regression line?
A) 0.0075
B) 0.0925
C) 61.16
D) –176.64

Page 21
25. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

What is an approximate 90% confidence interval for the slope 1?

A) –0.09 ± 0.0075
B) –0.09 ± 0.0125
C) 0.09 ± 0.0075
D) 0.09 ± 0.0125

26. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

Suppose the researchers conducting the study wish to test the hypotheses H0: 1 = 0
versus Ha: 1> 0. What do we know about the P-value of this test?
A) The P-value is greater than 0.10.
B) The P-value is between 0. 05 and 0.10.
C) The P-value is between 0.01 and 0.05.
D) The P-value is less than 0.01.

Page 22
27. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

Is there strong evidence of a straight-line relationship between sales and profits? Explain
briefly.
A) Yes, because the slope of the least-squares line is positive
B) Yes, because the P-value for testing if the slope is zero is quite small
C) No, because the value of the square of the correlation is relatively small
D) It is impossible to say because we are not given the actual value of the correlation.

28. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

Suppose the researchers conducting this study wish to estimate the profits (in hundreds
of thousands of dollars) for companies that had sales (in hundreds of thousands of
dollars) of 500. The following results were obtained from statistical software.

Sales Predicted profit Standard error 95.0% C.I. 95.0% P.I.

500 –130.4 59.3 (–248.5, –12.3) (–1066.4, 805.6)

If the researchers wish to estimate the mean profits for all companies that had sales of
500, what would be a 95% confidence interval for the mean profits?
A) (–1066.4, 805.6)
B) (–248.5, –12.3)
C) –130.4 ± 59.3
D) 500 ± 59.3

Page 23
29. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

Suppose the researchers conducting this study wish to estimate the profits (in hundreds
of thousands of dollars) for companies that had sales (in hundreds of thousands of
dollars) of 500. The following results were obtained from statistical software.

Sales Predicted profit Standard error 95.0% C.I. 95.0% P.I.

500 –130.4 59.3 (–248.5, –12.3) (–1066.4, 805.6)

If the researchers wish to estimate the profits for a particular company that had sales of
500, what would be a 95% prediction interval for the profits?
A) (–1066.4, 805.6)
B) (–248.5, –12.3)
C) –130.4 ± 59.3
D) 500 ± 59.3

Page 24
30. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

The following (partial) ANOVA table was obtained from statistical software.

Source DF Sum of squares

Model 1 32,809,212
Error 16,734,234

What are the degrees of freedom for SSE, the error sum of squares?
A) 2
B) 77
C) 78
D) 79

Page 25
31. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

The following (partial) ANOVA table was obtained from statistical software.

Source DF Sum of squares

Model 1 32,809,212
Error 16,734,234

What is the value of the SST, the total sum of squares?

A) 16,074,978
B) 16,734,234
C) 32,809,212
D) 49,543,448

Page 26
32. A random sample of 79 companies from the Forbes 500 list (which actually consists of
nearly 800 companies) was selected, and the relationship between sales (in hundreds of
thousands of dollars) and profits (in hundreds of thousands of dollars) was investigated
by regression. The following results were obtained from statistical software.

R2 = 0.662
s = 466.2

Variable Parameter estimate Standard error

Constant –176.644 61.16
Sales 0.092498 0.0075

The following (partial) ANOVA table was obtained from statistical software.

Source DF Sum of squares

Model 1 32,809,212
Error 16,734,234

What is the value of the F statistic for testing the hypotheses H0: 1 = 0 versus
Ha: 1  0?
A) 1.96
B) 77
C) 150.97
D) 217,328

Page 27
33. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be Win%  0.932  19.022Goals / G , r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418.

Which of the following statements is/are FALSE?

A) An amount of 39.8% of the variation in the winning percent variable is explained
by the least-squares regression on the number of goals scored per game.
B) An increase of 1 goal per game results in an increase of about 19% in winning
percent.
C) If a team scores 3 goals per game, we would predict the team would have a Win%
of 55%.
D) The Ottawa Senators scored 3.49 goals per game and their Win% was 64.0.The
residual for Ottawa was then –3.32.
E) The mean value of the Win% variable is 0.932 when the Goals/G is 0.

Page 28
34. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be Win%  0.932  19.022Goals / G , r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418.

A plot of the residuals from the least-squares fit against the Goals/G variable is shown
below.

What statements about residuals and/or about this residual plot is/are FALSE?
A) There does not appear to be any particular pattern to the residuals on the plot.
B) The residual plot shows that the residuals do approximately follow a Normal

Page 29
 distribution, as the statistical model requires.
C) Residuals from a least-squares fit in simple linear regression always sum to zero.
D) None of the residuals look as though they would be considered to be outliers.
E) The residuals appear to vary randomly about their mean of zero.

35. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be , r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418.

If a test of hypothesis were conducted of H0: 1 = 0 against Ha: 1  0, what would be the
value of the test statistic?
A) t = 0.07
B) z = 0.07
C) z = 4.31
D) F = 4.31
E) t = 4.31

Page 30
36. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be , r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418. We are told that s 2 =
60.29. How many degrees of freedom are associated with this statistic?
A) 29
B) 1
C) 30
D) 28
E) None of the above

Page 31
37. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be , r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418. What would the
approximate 96% confidence interval be for the true slope 1 ?
A) 19.022 ± 2.054(12.800)
B) 19.022 ± 2.154(4.418)
C) 0.932 ± 2.154(12.800)
D) 19.022 ± 2.054(4.418)
E) 0.932 ± 2.154 (12.800)

Page 32
38. As in most professional sports, statistics are collected in the National Hockey League. In
the 2006–2007 season, teams played 82 games. A team was awarded 2 points for a win
and 1 point if the game was tied at the end of regulation time but then lost in overtime.
For each of the 30 teams, data on the number of goals scored per game (Goals/G) and
the percentage of the 164 possible points they won (Win%) during the season were
collected. The following graph shows the plotted points for the variables Win% and
Goals/G and the simple linear regression line fitted using least squares.

From the computer output for the least-squares fit, the estimated equation was found to
be r 2 = 0.398, and s 2 = 60.29. Also, it was
determined from the output that SE b0 = 12.800 and SE b1 = 4.418. For the 2006-2007
season, teams scored an average of x = 2.88 goals per game. For the population of
teams that score 2.5 goals per game, the standard error of the estimated mean Win% is
SE ˆ = 2.197. What is the estimated mean Win% for the population of teams that score
2.5 goals per game?
A) 42.7%
B) 53.6%
C) 48.5%
D) 55.7%
E) Not within ± 2% of any of the above

Page 33
39. The statistical model for linear regression can be expressed as which of the following
equations?
A) DATA = FIT + RESIDUAL
B) FIT = DATA + RESIDUAL
C) RESIDUAL = DATA + FIT
D) DATA = FIT

40. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the

explanatory variable. The parameters a 0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
The explanatory variable x is ________.
A) quantitative
B) qualitative
C) categorical
D) None of the above

41. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the

explanatory variable. The parameters a 0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
What is the slope of the regression line?
A) 10
B) 70
C) 80
D) 60

42. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the
explanatory variable. The parameters  0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
What is the y-intercept for the regression line?
A) 10
B) 70
C) 80
D) 60

Page 34
43. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the
explanatory variable. The parameters  0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
What is the subpopulation mean when x = 20?
A) 270
B) 100
C) 70
D) None of the above

44. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the
explanatory variable. The parameters  0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
What is the range of values for 68% of the observed responses when x = 20 using the
68-95-99.7 rule?
A) 265 to 275
B) 270 to 275
C) 265 to 270
D) None of the above

45. The statistical model for simple linear regression is written as , where
 y represents the mean of a Normally distributed response variable and x represents the
explanatory variable. The parameters  0 and 1 are estimated, giving the linear
regression model defined by , with standard deviation  = 5.
What is the distribution of the test statistic used to test the null hypothesis
against the alternative hypothesis ? (Note: Assume n is the sample size.)
A) N(0,1)
B) N(0,2)
C) t(n – 1)
D) t(n – 2)

Page 35
46. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

What is the test statistic to test the null hypothesis that the slope is zero versus the
alternative hypothesis that the slope is not zero?
A) 100
B) 4
C) 2
D) None of the above

47. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

What is the P-value for a test of the null hypothesis that the slope is zero versus the
alternative hypothesis that the slope is not zero?
A) .04
B) .05
C) .02
D) None of the above

48. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

Is there a statistically significant linear relationship between the response and

explanatory variable when the significance level, , is .01?
A) Yes
B) No

Page 36
49. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

What is the predicted value of the response when x = 5?

A) 120
B) 100
C) 20
D) None of the above

50. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

Suppose an observed response value is 150 when x = 5.What is the value of the
residual?
A) 150
B) 30
C) –30
D) None of the above

51. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

What is the critical value, t*, that is used to compute an 80% confidence interval for 1?
(Note: Use software to compute the exact value.)
A) .84
B) 1.29
C) 1.98
D) None of the above

Page 37
52. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

What is an 80% confidence interval for 1?

A) (–0.71, 3.29)
B) (.71, 3.29)
C) (–2.71, 5.29)
D) None of the above

53. Suppose we are given the following information:

Sample size, n, = 100

Standard error of slope of the regression line, =2

y = 100 + 4x

Would a 95% confidence interval for 1 be larger, smaller, or the same as an 80%
confidence interval for 1?
A) Larger
B) Smaller
C) Same
D) It cannot be determined from the information given.

54. An interval used to predict a future value is called a ______.

A) prediction interval
B) confidence interval
C) standard error
D) population interval

55. Prediction intervals are usually wider, narrower, or the same as confidence intervals?
A) Wider
B) Narrower
C) The same

Page 38
56. Suppose you are testing the null hypothesis that the slope of the regression line is zero
versus the alternative hypothesis that the slope is different than zero.Would a very small
P-value (i.e., less than .0001) indicate a strong relationship between the explanatory
variable and the response variable?
A) Yes, because the P-value will give the strength of the association between the
explanatory variable and the response variable
B) Yes, because if the P-value is small, then R2 is large
C) No, because there could be a lot of scatter about the regression line, indicating a
weak fit of the data to the model
D) No, because a large P-value would indicate a strong relationship between the
explanatory variable and the response variable

Page 39
57. A recent study was done to assess factors that put Hispanic populations more at risk for
obesity and related chronic diseases, such as diabetes and heart disease, than
non-Hispanic populations. Data were collected on several factors, such as the crude
morality rate of Hispanics, obesity prevalence, percent of adults diagnosed with
diabetes, and median income at the state level. Pearson's Correlations were used to
examine the strength of the relationship between obesity and the other variables, as a
way of observing which characteristics were associated with high prevalence of obesity.
In addition, a simple linear regression was used to model the relationship between
diabetes and obesity. The results from SPSS are shown below.

Page 40
Page 41
What is the sample correlation between obesity prevalence and percent adults diagnosed
with diabetes?
A) –.240
B) .141
C) 1
D) None of the above

Page 42
58. A recent study was done to assess factors that put Hispanic populations more at risk for
obesity and related chronic diseases, such as diabetes and heart disease, than
non-Hispanic populations. Data were collected on several factors, such as the crude
morality rate of Hispanics, obesity prevalence, percent of adults diagnosed with
diabetes, and median income at the state level. Pearson's Correlations were used to
examine the strength of the relationship between obesity and the other variables, as a
way of observing which characteristics were associated with high prevalence of obesity.
In addition, a simple linear regression was used to model the relationship between
diabetes and obesity. The results from SPSS are shown below.

Page 43
Page 44
What is the P-value to test that the population correlation is zero verses the alternative
that the population correlation is greater than zero?
A) 0
B) 1
C) It cannot be determined from the given information.
D) None of the above

Page 45
Another random document with
no related content on Scribd:
 A Sanitary Drinking Tube

To make this tube, procure a piece of ³⁄₄-in. gauge glass, about 10

in. long, and bend it by heating the glass with a blowtorch 2¹⁄₂ in.
from one end, as shown. If this is done in a place where no air will
strike the glass while it is heated and the tube is then allowed to cool
slowly, it will not crack. It is used by putting the short end under the
faucet and drawing on it as on a pipe.
 A Comb Cleaner

The Twisted Wires Form an Excellent Device for Cleaning a Comb

A very handy device for cleaning combs can be made from a piece
of spring sheet brass, 4¹⁄₂ in. long and about 3 in. wide. Notches are
cut in the ends of the brass with a hacksaw, making the projections
as wide as the saw cut. The brass is then bent into shape with a
special clamp made for the purpose, which consists of three pieces
of wood, two being cut to form a curved slot, then fastened to the
third piece. The brass is sprung into the slot, and then fine wire is
stretched between the ends in the notches. The wire should be very
fine and two strands twisted together and run through the notches.
 Concealing the House Key
The time-honored custom of concealing the house key under the
door mat, or in the letter box, when the family has not enough keys
to go around, is so well known that an unauthorized person seeking
to enter the house would look in these places first of all.

A simple and effective hiding place for the key can be quickly and
easily made with the aid of an auger and two pieces of tin. Pick out
an obscure section of the porch railing, and in the edge of this bore a
³⁄₄-in. hole, about ¹⁄₄ in. deeper than the length of the key. Make a
piece of tin into a cylinder, the same length as the key, so that the
latter will slide easily into the hole. At one end of this cylinder solder
a 1-in. disk of tin, which will make it appear as in the illustration.
If the key is placed in the cylinder and the latter pushed into the
hole until it is flush with the surface, it will scarcely be noticed by
anyone not in the secret, and by painting it the same color as the
railing it will become still more inconspicuous.—Contributed by Frank
L. Matter, Portland, Ore.
 A Mysterious Watch

A very interesting experiment may be made with the ordinary

dollar watch in illustrating the law of the pendulum. A pendulum 39.1
in. long will make 60 one-way swings per minute, the number of
swings varying inversely as the square root of the length. By actual
count it was found that the balance wheel of the watch in question
made 240 one-way swings per minute, which is just 4 times as fast
as the 39-in. pendulum. Therefore, according to the foregoing law, a
pendulum ¹⁄₁₆ as long, or about 2¹⁄₂ in., would swing in unison with
the wheel of the watch. The question then arises as to what would
happen if the watch itself were suspended so as to swing as a
pendulum of the latter length. The experiment was made as
illustrated, with the result that the watch keeps on swinging
continuously. The swing amounts to about ³⁄₈ in., and appears so
vigorous that it is almost incomprehensible that the small spring in a
watch should be able to maintain so much weight in continuous
motion for 24 hours.
 Bait Casting for
Game Fish
by Stillman Taylor

There are many enjoyable phases of the fisher’s art, but bait casting
from the free reel probably has a wider appeal than any other
branch of angling. The pleasure of handling a short bait-casting rod
is in itself a good sport, even when casting done in the back yard is
made for distance and accuracy only. Get two or three enthusiastic
casters together, and you may have an interesting little tournament,
held on a vacant lot, on the lake, or on the greensward of a city park.
There is plenty of action in casting from the reel, and it is its variety
that has made angling of this type so universally popular. True,
considerable practice is required before the caster is able to shoot
an accurate plug far off, but the knack may be gained after
reasonable application. The handling of the short rod differs from all
other angling methods—it is an active sport rather than a
contemplative recreation—and when the fun of handling a good outfit
is combined with the sport of coaxing out a black bass, shooting a
wooden minnow among the lily pads for pickerel, or casting the bait
for those pirates of our fresh-water lakes, the wall-eyed pike and the
muskellunge, the angler gets a taste of fishing—plus.

The Outfit and Its Selection

A good firearm is necessary to the good marksman and a well-

made, nicely balanced rod, to which is clamped a smooth-running
reel, is the essential equipment of an able angler. Bait casting is, of
course, more a matter of skill than of outfit, but that is no reason why
the angler should be handicapped by poor tackle, although large
expenditure is not necessary in obtaining a satisfactory outfit. The
casting rod may be from 5 to 6 ft. long. Where the caster regards
length of cast as the important factor, the 5-ft. length will probably
give the best satisfaction. However, the superiority of one over the
other in this regard is rather to be found in the elasticity and quality
of the rod than its length. The skill of the caster is, moreover, the
ruling factor. The 5¹⁄₂-ft. rod is better than the 5-ft. length for fishing
purposes, and the 6-ft. rod is more resilient than the 5¹⁄₂-ft. tip. If the
greatest range is not of more importance than the landing of the fish,
the longer rod will prove the most pleasant to handle.
The split-bamboo rod has many friends, and a well-made rod of
this material is a thoroughly satisfactory casting tool. A fine-quality
casting rod of split bamboo costs from $10 to $25. A fairly good rod
may be bought for $2 or a trifle less, and a smooth-running reel for
another $2, while a line for it will bring the total up to about $5. An
outfit of this kind purchased with discrimination, while not embodying
everything that a critical angler would desire, will cast a long and
accurate line and stand up under the strain of handling our heaviest
game fish. The one-piece rod is classed by many anglers as the
ideal casting rod. The two-piece rod is handier to pack and is more
generally used.
The chief fault in bait-casting rods is the stiff action of the tip
section. For tournament casting, a fairly stiff rod is desirable. The
same requirement holds good for muskellunge and other heavy
fishing. A flexible rod affords more pleasure in playing black bass
and other active “gamey” fish. However, the best way to test out a
rod is by casting with it.

A Serviceable Reel Costing $4, a High-Grade Reel Costing $13.50, and a Fine
Tournament Casting Reel Costing $26, are Shown from Left to Right. The
Center One Has a Level-Winding Device and a Thumb-Click Drag on the
Rims of the Plates. The Tournament Casting Reel Has a Cork Arbor and
Jeweled Bearings

The steel casting rod is well liked by some casters, but only a
high-quality steel rod is worthy of consideration. It is generally not
classed as the equal of a well-made solid-wood or split-bamboo rod,
for casting. The better-grade steel rod is springy and flexible, and for
this reason is a good rod for casting for black bass. Still, most
anglers have a steel rod in their outfits, but it is not as generally used
at tournaments as the time-tried wooden rod.
The guide is an important detail of a casting rod. Inexpensive rods
are usually provided with large metal guides, while the better rods
are fitted with agate guides, or with agate tip and hand guides. A
guide, ¹⁄₂ in. in diameter, at the butt of the rod is large enough and
will handle the line smoothly.
The single grip is sometimes preferred by casters, but the
additional grip placed above the reel is desirable if much fishing is
done. Solid-cork hand grasps are most satisfactory, being less
slippery when wet and less tiring to the hand than handles made of
other materials. If the lower grip is properly shaped with a forward
edge, the forefinger will find a firm grip upon it, and the finger hook
will not be needed. However, some anglers prefer to use the hook,
and if the rod is not so fitted, a detachable hook may be fastened to
it. A reel band that locks securely is desirable, and all the better
grades of casting rods are so fitted.
 The quadruple reel is the logical choice of the bait caster, the
spool turning four times to one turn of the handle. As the work
demanded of a bait-casting reel is different and more exacting than
that of the click reel used in fly casting, its design is radically
different. The best type of click reel is of large spool diameter and
narrow between plates, while the typical casting reel is its direct
opposite—being long-barreled, with plates of comparatively small
diameter. For tournament casting the finest outfit will naturally be
selected, but for practical work it is unnecessary to pay $25 for a
finely wrought reel. Three typical reels are shown in the illustration.
The one at the left holds 80 yd. of line, can be easily taken apart for
cleaning, and costs $4. In the center is a level-winding reel with
thumb click and adjustable drag placed on the rims of the plates. It
holds 60 yd. of line, and costs $13.50. At the right is a very high-
grade reel for tournament casting. It has a cork arbor, jeweled
bearings holds 60 yd. of line, and costs $26.
Any of the better grades of braided silk are suitable for the bait-
casting line. The average price for a dependable line is about $1 for
50 yd. A small size line is best, and sizes G and E are preferred. A
brown and green, or other dark-colored, line is preferable for fishing.
The black and white, all white and other light-colored, lines are
satisfactory for tournament use. After a day’s fishing the line should
be unreeled and looped around the back of a chair in loose coils to
dry, or loosely wound upon a drying reel.
For practice and tournament casting one of the standard ¹⁄₄ and
¹⁄₂-oz. loaded wooden or metal plugs may be used. The popularity of
bait casting from the reel has brought forth a large variety of artificial
baits. These wooden and enameled lures are used by anglers
because they bring the catches.
The wooden minnows, and other forms of artificial bait, may be
roughly classed in three divisions: the surface lures for use on top of
the water; the diving and under-water baits; and the weedless
varieties, which are usually of the surface type. No bait caster’s outfit
is complete unless it includes one bait of each class, and to meet
fully the different and varying conditions of weather, season, and
water, it is a good plan to select one of each type in light and
medium-dark colors. A variety of typical baits are shown in the page
plate at A, B, C, etc. Many of the baits are provided with one or more
eyes for attaching the line, thus enabling the caster to spin his bait at
different depths. All of the wooden-body baits will float, and only dive
under the surface when reeled in; the quicker they are retrieved the
deeper they spin. While many of these baits have been designed
especially to entice the black bass, practically all of them are good
for pickerel and general fresh-water casting, particularly the minnow
baits.
The other items which complete the bait caster’s outfit are the
landing net and tackle box. The net should be well made, with a steel
hoop 13 in. or larger in diameter, and should have a handle about 3
ft. long. The folding type of net is the most convenient, and is
provided with a separate handle. The tackle box may be of sheet
metal or leather, with partitions for holding reels, baits, and the
angling “what not.” A convenient and serviceable landing net costs
$1 or more, and a tackle box will cost about as much.

Suggestions on Making the Cast

Although much practice is required before the caster can take his
place in a tournament, “getting the hang” of handling the short rod
and free-running reel comes quickly. As soon as the simple
principles are understood, length and accuracy will come only
through practice. Bait casting is easier than fly casting and the angler
need concern himself only with the length and accuracy of his cast,
since delicacy does not enter into this phase of fishing. The coarser
sweet-water fish, like black bass and pickerel, are not frightened
when the bait hits the water with a splash—indeed, the splash of the
bait attracts them.
Two methods of casts are used by bait casters, the overhead and
the underhand or side cast. The overhead cast is used largely by the
practical angler, as well as by those who take part in tournament
contests, because the bait is projected a longer distance and may be
more accurately placed. If the novice will thumb the line, as shown in
Fig. 1, with thumb bearing directly on the cross rod of the reel with
the ball of the thumb resting lightly on the line, one or two trials will
show the correct manner of spooling. The line is retrieved as shown
in Fig. 2, care being taken to form correct habits from the start. To
make the cast, reel in the bait until it is some 6 in. from the tip,
release the click or drag, and press the ball of the thumb firmly upon
the spooled line. Carry the rod over the shoulder until it is horizontal
or nearly so, as in Fig. 3, and then bring it smartly forward, checking
it quickly when it reaches an angle of some 45°, as shown in Fig. 4.
This projects the bait forward in a straight line. When the rod is
brought forward the tension of the thumb on the spool is released so
that the spool will revolve freely, but enough pressure must be
communicated by the thumb to prevent the reel from spinning faster
than the line is unreeled from the spool, otherwise backlash will
occur. The use of the whole arm and the weight of the body,
combined with the elasticity of the pliant rod, enables the caster to
cover a surprisingly long distance after a little practice. A distance of
150 ft. is not difficult to attain, but for fishing, a well-placed cast of 60
or 75 ft. is long enough to reach promising stretches of water.
 The Method of Thumbing the
A Line is Shown in Figure 1, and
the Correct form in Retrieving
the Line, in Figure 2. The
Overhead Cast is Illustrated in
Fig. 3 Figures 3 and 4, and the
B Underhand or Side Cast in
Figures 5 and 6
C
D
Fig. 4 Fig. 1
E
Fig. 2 Fig. 5
F
Typical Baits:
A, Under-Water Minnow; G
B, Wabbler Spoon;
C, Bass;
D, Fluted Wabbler;
E, Weedless; H Fig. 6
F, Diving;
G, Combination Color, Taste,
and Smell;
H, Spoon and Minnow Bass; I
I, Surface

The underhand or side cast is somewhat easier to master, and is

made much in the same way as the overhead cast, only the rod is
swung horizontally to the rear, the tip on a level with the angler’s
chest, as shown in Fig. 5. Bring the rod forward smartly in the
direction it is desired to shoot the bait, check the rod when about
horizontal, partly releasing the pressure of the thumb, just enough to
allow the spool to revolve freely, and check the line as the bait hits
the water. Figure 5 shows the beginning of the side cast and Fig. 6
the cast completed. This cast is useful when fishing from a shore
overhung with trees, and with practice the bait may be projected a
long distance.
 The chief factor of skill in bait casting from the reel lies in
controlling the line, and the thumb must be trained to do this through
practice. Some casters turn the reel sideways with the handle
downward while making the cast, but the majority keep the reel level
and cast with it in the same position as when reeling in. Thumbing
the line is easiest done when the spool is well filled, and if the
casting line is not long enough to do this, sufficient common line
should be first wound on to fill it.
The successful bait caster is one who possesses an intimate
knowledge of the fish he casts for, and while a lengthy exposition is
not to form a part of this article, reference to the fish commonly
sought by the caster will not be amiss. The fish generally of interest
to the bait caster are small and large-mouth black bass, pike,
muskellunge, and pickerel. The small-mouth black bass is classed
as the finest game fish, ounce for ounce, by many casters. The
muskellunge, often confused with the Great Lakes or English pike, is
more limited in range than the pike, being found in the Great Lakes
region and the waters of the St. Lawrence River.
Since the habits and general characteristics of the large-mouth
bass are much the same as its near relative, the small-mouth, the
following hints apply to both. When wading, it is well to cast
downstream. Cast in the shallows, near the weeds and snags, and
close to stumps. As little noise should be made as possible.
For early-spring casting, bass are likely to be found in the shallows
and sheltered places during the day, because the water is cool at
this season. The pickerel will also be found in the warmer and
shallower waters. The most successful baits at this time are the
diving minnows, and other under-water lures. As the warmer days of
late spring and early summer appear, the bass swim out into deeper
and cooler water, while the pickerel remain close to the lily pads. The
wooden minnows are excellent baits to use at this season. When the
weeds are very thick, one of the several kinds of weedless baits may
be used.
The early-morning and late-afternoon hours are the best for
midsummer casting, and the shallow places are the good spots to
work over. If there are any brooks or streams emptying into the lake
or pond, do not neglect to work this stretch well at the mouth. When
the sun is low, the white or light-colored surface baits may be
substituted for the medium-dark colored diving varieties, and for
casting after dark, a luminous or moonlight bait will often give good
results. During the fall months fishing conditions are similar to
casting in the spring, but the deep water is then the warmest, and
the under-water or diving baits are usually the best.
To catch bass, the angler must cast his bait where the fish are
feeding, and he will find it an advantage to match natural conditions,
so far as possible, in selecting his baits. For casting on bright days in
clear waters, the red, white and red, white, or green minnows are
good lures, as are also the silverplated spoon baits. Cast well ahead
of the boat, and reel in very slowly to keep the plug spinning well
below the surface. On cloudy days and in muddy waters, after a
heavy rain, for example, use red and yellow baits and goldplated or
burnished-copper spoons. The best all-around bait—equally good for
bass and pickerel—is one of the green-backed minnows, but no one
bait can be expected to meet all conditions, hence the caster should
have a fair assortment of baits.
A stiff casting rod with plenty of backbone is essential for
muskellunge. The angler is out after big game and must be prepared
for the sport. A hard, braided silk line, size E, and 3-0 or 4-0 hooks,
snelled and swiveled, are the best. Any of the wooden baits suitable
for pickerel—the minnows especially—are good for muskellunge. A
man to handle the oars is a handy companion, and after casting the
bait just outside the weed bed, make for deep water as soon as a
fish is hooked. The fish will be seen and felt quickly enough, and
when it breaks water and leaps in the air, give it a little slack line
quickly, otherwise the strain will snap the leader. A twisted, bronzed
trace is the best to use between line and bait, and this need not be
longer than a foot—6 in. is enough. When the fish is brought
alongside, shoot him through the head with a revolver which is
probably the most satisfactory manner of making an easy landing.
A Cat-Proof Bird Table
 Our bird table is a source of great enjoyment, particularly since the
birds feel secure from cats or other enemies because of the
construction of this ornament in our garden. The sketch shows the
arrangement of the table braced at the top of a 6-ft. post. Shrubbery
surrounds the table and a light evergreen climber clings to the post,
yet does not give the cats a good foothold.
Experience has taught us that birds in general prefer bread
crumbs to other varieties of food, and they are also fond of cracked
wheat. The linnets like oranges particularly. We cut an orange in two
and place the halves on the table. It is amusing to see the birds
balance on one side of the orange while they peck at the fruit. Soon
the orange peel is almost entirely emptied. We provide a small basin
of fresh water on the table, and the birds use it as a drinking cup as
well as a bathtub.—F. H. Sweet, Waynesboro, Va.