Modules 5 & 6 Practice Problems - Key

Module 5 Practice Problems

Question 1

A study was performed to assess how well a psychiatric treatment reduced anxiety levels of
middle-age patients dealing with anxiety. Eight middle-age patients were randomly selected to
participate in the study, and their anxiety levels were measured before and after the treatment.
Assume the anxiety levels of the patients prior to treatment and after treatment are normally
distributed. Compute a 95% confidence interval for the population mean of paired differences.

Patient 1 2 3 4 5 6 7 8
Before 6.5 6.3 9.1 10.3 11.4 8.2 6.1 11.5
After 6.8 2.4 7.6 8.4 8.12 6.2 3.7 2.3

(a) Compute the mean and standard deviation of the sample differences. The sample
differences should be of the form: before – after. Round to three decimal places.

Differences:
-0.3 3.9 1.5 1.9 3.28 2 2.4 9.2

Mean: x=2.985 Standard deviation: s = 2.804

(b) Determine the critical value Tc that corresponds to the 95% level of confidence. Enter the
value rounded to three decimal places. Hint: Use a T-distribution critical value table or
technology to find the critical value.

n = 8, df = 7, c = 0.95, Tc = 2.365

(c) Determine the margin of error for a 95% confidence interval for the population mean of the
paired differences. Round to two decimal places.

2.804
Margin of error: E=( 2.365 ) =2.34
√8
(d) Compute a 95% confidence interval for the population mean of the paired differences.
Round the lower limit and upper limit to two decimal places.

95% Confidence Interval: ( x−E , x + E )= (2.985−2.34 ,2.985+2.34 )=(0.64 ,5.33)

(e) Interpret the confidence interval.

We are 95% confident that the mean of all differences (anxiety level before – anxiety level
after) is between 0.64 and 5.33.

(f) Can we conclude that the treatment lowers patients’ anxiety levels? Explain.

Yes, since the confidence interval only contains positive values, we can conclude that the mean
of anxiety levels before treatment is greater than the mean of anxiety levels after treatment.
This indicates that the treatment lowers anxiety levels of patents in this population.

Question 2

A researcher is interested in determining if a new diet lowers cholesterol levels of middle-age

adults who experience high cholesterol. Ten middle-age adults were randomly selected to
participate in the study. Participants’ cholesterol levels were measured before the experiment,
prior to the implementation of the new diet, and after the experiment, after subjects
implemented a high protein diet for six weeks. Assume the cholesterol levels before the diet
and after the diet of the population of interest are normally distributed. Compute a 90%
confidence interval for the population mean of paired differences.

Subject 1 2 3 4 5 6 7 8 9 10
Before 229 230 235 218 236 247 233 250 228 245
After 235 225 235 212 235 245 230 248 225 244

(a) Compute the mean and standard deviation of the sample differences. The sample
differences should be of the form: before – after. Round values to three decimal places.

Differences:
-6 5 0 6 1 2 3 2 3 1

Mean: x=1.7 Standard deviation: s = 3.268

(b) Determine the critical value Tc that corresponds to the 90% level of confidence. Enter the
value rounded to three decimal places. Hint: Use a T-distribution critical value table or
technology to find the critical value.

n = 10, df = 9, c = 0.90, Tc = 1.833

(c) Determine the margin of error for a 90% confidence interval for the population mean of the
paired differences. Round to two decimal places.

3.268
Margin of error: E=( 1.833 ) =1.89
√ 10
(d) Compute a 90% confidence interval for the population mean of the paired differences.
Round the lower limit and upper limit to two decimal places.

95% Confidence Interval: ( x−E , x + E )= (1.7−1.89 , 1.7+1.89 ) =(−0.19 , 3.59)

(e) Interpret the confidence interval.

We are 90% confident that the mean of all differences (cholesterol level before – cholesterol
level after) is between -0.19 and 3.59.

(f) Can we conclude that the diet leads to lower cholesterol levels? Explain.

No, since the confidence interval contains 0, we cannot conclude that there is a difference
between the mean cholesterol level before the diet compared to the mean cholesterol level
after the diet. We cannot conclude that the mean cholesterol level after the diet is lower, so
we can’t conclude that the diet lowers adults’ cholesterol levels.

Question 3

Does a new writing app help students reduce the number of errors they have when writing a 2-
page paper? Six students were randomly selected to participate in an experiment. Participants
wrote a paper prior to using the app, and after using the app, and the number of errors in their
writing was identified. Assume the number of errors that students make before and after using
the app are normally distributed. At the 5% level of significance, determine if the app reduces
the number of errors that students make in their writing.

Student 1 2 3 4 5 6
Before 5 7 4 12 11 6
After 3 6 4 10 11 5

(a) Determine the null and alternative hypotheses.

Hypotheses: H 0 : μ=0 , H a : μ>0 (Right-tailed test)

(b) Compute the mean and standard deviation of the sample differences. The sample
differences should be of the form: before – after. Round to three decimal places.

Differences:
2 1 0 2 0 1

Mean: x=1 Standard deviation: s = 0.894

(c) Compute the standardized test statistic? Round to two decimal places.

x−μ 1−0
T= = ≈ 2.74
Test statistic: s 0.894
√n √6
(d) What is the P-value? Explain how you determined the P-value.

The degrees of freedom are df = 5. The P-value is the area to the right of T = 2.74, which is
0.02. Since the P-value is less than 5%, we reject the null hypothesis and support the
alternative hypothesis.

(e) State a decision about the hypotheses and answer the research question.

The sample provides evidence that the mean number of errors student have before using the
app is greater than the mean number of errors students have after using the app. The sample
provides evidence that the app reduces the number of errors of all users.

Question 4

A researcher was interested in examining whether the acidity of water in households in a city
had changed from 2010 to 2020. Nine households were randomly selected to participate in the
study. In 2010 and 2020, researchers collected the pH level of water in the same nine
households. Assume the pH levels of water in the households in the city in 2010 and 2020 are
normally distributed. At the 5% level of significance, determine if there is a difference in the pH
levels of water in the households in the city.

Household 1 2 3 4 5 6 7 8 9
2010 5.5 6 7 6.5 4.5 6.5 7 5.5 4.5
2020 6 6 6.5 5.5 4 6 6 5.5 5

(a) Determine the null and alternative hypotheses.

Hypotheses: H 0 : μ=0 , H a : μ ≠ 0 (Two-tailed test)

(b) Compute the mean and standard deviation of the sample differences. The sample
differences should be of the form: 2020 – 2010. Round to three decimal places.

Differences:
0.5 0 -0.5 -1 -0.5 -0.5 -1 0 0.5

Mean: x=−0.278 Standard deviation: s = 0.565

(c) Compute the standardized test statistic? Round to two decimal places.

x−μ −0.278−0
T= = ≈−1.47
Test statistic: s 0.565
√n √9
(d) What is the P-value? Explain how you determined the P-value.

The degrees of freedom are df = 8. The P-value is twice the area to the left of T = -1.47, which is
2(0.089) = 0.178. Since the P-value is greater than 5%, we fail to reject the null hypothesis and
fail to support the alternative hypothesis.

(e) State a decision about the hypotheses and answer the research question.

The sample does not provide evidence that the mean pH of water in the households in the city
in 2010 is different than the mean pH level of water in the households in the city in 2020. There
is not sufficient evidence to conclude that the acidity in the water in the households is different
from 2010 to 2020.

Module 6 Practice Problems

Question 1

The scatterplot below shows the free throw percentage (FTPct) and field goal percentage
(FGPct) of 80 professional basketball players. A player’s free throw percentage is the
percentage of free throws (shots attempted from the free throw line) that a player makes. A
player’s field goal percentage is the percentage of shots that a player makes.
a. What type of relationship exists between the two variables?

A weak, negative relationship. There is a lot of scatter about the line of best fit.

b. What is the explanatory variable in this scatterplot?

FTPct – Free throw percentage.

c. What is the response variable in this scatterplot?

FGPct – Field goal percentage.

d. Which number below is a reasonable estimate of the correlation coefficient? Why?

(-0.95, -0.45, 0.50, 0.85)

-0.45. The relationship is negative but weak, so r will be negative but not very close to -1.

Question 2

The table below shows the relationship between the pages-per-minute (PPM) and price for a
random sample of 8 inkjet printers.

PPM Price
3.9 300
2.9 199
2.7 79
 2.9 129
2.4 70
4.1 348
3.4 299

a. What is the correlation coefficient? Round to three decimal places.

r = 0.943

b. Is it appropriate to use page-per-minutes to predict price? Why?

Yes, the scatterplot shows a strong, positive linear relationship between the two variables.

c. Determine the equation of least-squares regression line?

Equation: ^y =170 x−338.12

d. Interpret the slope of the least-squares regression line.

If the PPM (pages per minute) increases by 1, the predicted cost of the printer will increase
by $170.

e. Interpret the y-intercept of the least-squares regression line.

An inkjet printer that prints 0 PPM (pages per minute) will cost -$338.12. This is not
reasonable.

f. Predict the price of an inkjet printer that prints 2.6 pages per minute?

^y =170 ( 2.6 )−338.12=103.88 Price = $103.88

g. Find and interpret the coefficient of determination?

r =0.889. This means that 88.9% of the total variability in price is explained by the page-
2

per-minute of a printer.

Question 3

The table below shows the relationship between the number of gallons of gas used and the
miles driven by a student on several long trips.

Number of Miles
Gallons of Gas
Driven
 7.4 245
8.6 280
5.4 175
9.1 295
8.3 265
4.6 235

a. What is the correlation coefficient? Round to three decimal places.

r = 0.814

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

^y =18.86 x +112.72

c. Interpret the slope of the least-squares regression line.

If the trip used 1 additional gallon of gas, the number of miles driven increased by 18.86
miles.

d. Interpret the y-intercept of the least-squares regression line.

The student would drive 112.72 miles on 0 gallons of gas. This is not reasonable.

e. Predict the number of miles driven if the student used 6 gallons of gas.

^y =18.86 (6 ) +112.72=225.88 225.88 miles

Question 4

The table below shows the 3rd exam scores and final exam scores for a random sample of
students from a large class of students in a mathematics class.

3rd Exam Score Final Exam Score

65 140
67 133
71 185
71 163
66 126
 75 198
67 173
70 180
73 162
69 151
69 159

a. Determine the correlation coefficient of the data. Round to three decimal places.

r = 0.761

b. At the 1% level of significance, perform a hypothesis test to determine whether there is a

positive correlation between the 3rd exam score and final exam score of all students in the
class.

 Do the sample data meet the criteria for a hypothesis test for the population
correlation coefficient: (1) random selection, (2) linearity, and (3) equal variance?
Explain.
 Determine the hypotheses.
 Compute the T-test statistic for this test. Round the value to two decimal places.
 Specify the degrees of freedom and the test statistic, and find the P-value.
 State a decision about the hypotheses and answer the research question.

The sample satisfies the three criteria listed. The sample is random so it is representative of the
population. The scatterplot shows a linear relationship between the variables. The equal
variance criteria is met since the vertical thickness of the scatterplot is roughly the same for all x
values.
Hypotheses: H 0 : ρ=0 , H a : ρ> 0

r 0.761
T= = ≈ 3.52
Test statistic:
√ √
2
1−r 1−(0.761)2
n−2 11−2

The degrees of freedom are df = 9. The P-value is the area to the right of T = 3.52, which is
0.0003. Since the P-value is less than 1%, we reject the null hypothesis and support the
alternative hypothesis. The sample provides evidence that there is a positive correlation
between the 3rd exam score and the final exam score among all students in the class.