Business Statistics, 4th Edition by
Norean R. Sharpe
Complete Chapter Solutions Manual
are included (Ch 1 to 25)
** Immediate Download
** Swift Response
** All Chapters included
,Table of Contents are given below
1.Data and Decisions (H&M)
2.Visualizing and Describing Categorical Data (Dalia Research)
3.Describing, Displaying, and Visualizing Quantitative Data (AIG)
4.Correlation and Linear Regression (Zillow.com)
5.Randomness and Probability (Credit Reports, the Fair Isaacs Corporation, and
Equifax)
6.Random Variables and Probability Models (Metropolitan Life Insurance
Company)
7.The Normal and Other Continuous Distributions (The NYSE)
8.Data Sources: Observational Studies and Surveys (Roper Polls)
9.Data Sources: Experiments (Capital One)
10.Sampling Distributions and Confidence Intervals for Proportions (Marketing
Credit Cards: The MBNA Story)
11.Confidence Intervals for Means (Guinness & Co.)
12.Testing Hypotheses (Casting Ingots)
13.More About Tests and Intervals (Traveler’s Insurance)
14.Comparing Two Means (Visa Global Organization)
15.Inference for Counts: Chi-Square Tests (SAC Capital)
16.Inference for Regression (Nambé Mills)
17.Understanding Residuals (Kellogg’s)
18.Multiple Regression (Zillow.com)
19.Building Multiple Regression Models (Bolliger and Mabillard)
20.Time Series Analysis (Whole Foods Market®)
21.Introduction to Big Data and Data Mining (Paralyzed Veterans of America)
22.Quality Control (Sony)
23.Nonparametric Methods (i4cp)
24.Decision Making and Risk (Data Description, Inc.)
25.Analysis of Experiments and Observational Studies
,Solutions Manual organized in reverse order, with the last chapter displayed first, to ensure that all chapters are included
in this document. (Complete Chapters included Ch25-1)
Chapter 25 – Analysis of Experiments and Observational Studies
SECTION EXCERCISES
SECTION 25.1
1.
a) For this experiment, there are k = 4 treatments and N = 40 subjects (4×10). The degrees of
freedom for treatment are k – 1 = 4 – 1 = 3; degrees of freedom for error are N – k = 40 – 4 = 36;
and total degrees of freedom are N – 1 = 40 – 1 = 39.
b) SSE = SSTotal – SST = 1177.97 – 856.07 = 321.9.
SST 856.07
c) MST 285.36.
k 1 3
SSE 321.9
d) MSE 8.94.
N k 36
2.
a) H0: µ1 = µ2 = µ3 = µ4
HA: at least one mean is different
MST 285.36
b) F 31.92.
k 1; N k MSE 8.94
c) The P-value associated with the calculated F-statistic depends on the numerator and denominator
degrees of freedom, 3 and 36 respectively. It can be found with technology (P-value < 0.0001) or
from Table F in the Appendix. Using Table F, at α = 0.01 the F value for 3 and 35 degrees of
freedom (36 is not in the table) is 4.40. The calculated value is much larger, so we can say that the
P-value is less than 0.001.
d) At α = 0.05, we can reject the null hypothesis and conclude that not all four means are the same (at
least one mean is significantly different).
SECTION 25.2
3. Cell phone adoption.
a) The null hypothesis H0: The mean cell phone adoption rates are equal in the three regions.
The alternative hypothesis HA: The mean cell phone adoption rates are not all equal in the three
regions.
b) Countries were selected randomly from the three regions, so the randomization condition is met.
However, boxplots suggest the spreads are quite different and show several outliers which require
attention. Other plots should be used to check the Nearly Normal Condition.
4. Marketing managers’ salaries.
a) The null hypothesis H0: The mean salaries for marketing managers are equal in all three regions.
The alternative hypothesis HA: The mean salaries for marketing managers are not all equal in the
three regions.
b) States were selected randomly from the three regions, so the randomization condition is met.
Boxplots indicate that the spreads are similar and appear to be symmetric but other plots should be
used to check the Nearly Normal Condition. The boxplots do indicate that the sample data may be
somewhat skewed, but it is difficult to tell if the normal population assumption is violated. Other
plots should be used to check the normal population assumption.
25-1
, 25-2 Chapter 25 Analysis of Experiments and Observational Studies
SECTION 25.3
5. Activating baking yeast.
a) The null hypothesis H0: 1 = 2 = 3 = 4 (the means of all four groups are equal). .
The alternative hypothesis HA: not all means are equal.
b) The data provide strong evidence with a P-value < 0.0001 to reject the null hypothesis and conclude that the
means are not all equal. This experiment provides strong evidence that the mean activation times differ
among the recipes.
c) Yes. Because we have rejected the null hypothesis, we can proceed with a multiple comparisons method to
compare all the groups.
6. Activating baking yeast.
a) The null hypothesis is that the mean Distance thrown using each grip is the same. The alternative
is that not all means are equal.
b) We would not reject the null hypothesis of equal means because the P-value is 0.1543. This
experiment provides insufficient evidence that the mean Distance differs among the three grips.
c) No. We have not rejected the null hypothesis, so we cannot proceed with multiple comparisons.
SECTION 25.4
7. No. This is an observational study. There may be other factors (such as age) that are influencing the
response.
8. If the treatments were randomized, then the researchers should be justified in making the claim.
CHAPTER EXERCISES
9. Managers’ hourly wages.
a) This is an observational study. The factors are not deliberately manipulated to specific treatments.
b) Boxplots of the hourly wages by both manager type and region indicate that the equal variance
assumption is reasonable. Also, the boxplots indicate fairly symmetric distributions. However,
other plots should be used to check the Nearly Normal Condition once the model has been fit. The
states were randomly selected from the three regions, so the randomization condition is satisfied.
c) Based on the P-values, we can conclude that there is no significant interaction effect. There is no
evidence that the mean hourly wages are different across the three regions but are different for
sales and advertising managers.
d) Yes, because the interaction effect is not significant.
10. Concrete testing.
a) This is an experiment. The factors are deliberately manipulated to specific treatments.
b) Boxplots of the compressive strengths by both alkali content and sand type indicate that the equal
variance assumption is reasonable. Also, the boxplots indicate fairly symmetric distributions.
However, other plots should be used to check the Nearly Normal Condition once the model has
been fit. The experimental units were selected randomly for each treatment, so the randomization
condition is satisfied.
c) Based on the P-values, we can conclude that there is no significant interaction effect. The mean
compressive strengths are not the same across the three different alkali contents and not the same
for the three sand types.
d) Yes, because the interaction term is not statistically significant. There is no evidence that the effect
of alkali content differs across the three sand types or that the effect of sand type differs across the
three alkali contents.
11. Cars (fuel efficiency).
Norean R. Sharpe
Complete Chapter Solutions Manual
are included (Ch 1 to 25)
** Immediate Download
** Swift Response
** All Chapters included
,Table of Contents are given below
1.Data and Decisions (H&M)
2.Visualizing and Describing Categorical Data (Dalia Research)
3.Describing, Displaying, and Visualizing Quantitative Data (AIG)
4.Correlation and Linear Regression (Zillow.com)
5.Randomness and Probability (Credit Reports, the Fair Isaacs Corporation, and
Equifax)
6.Random Variables and Probability Models (Metropolitan Life Insurance
Company)
7.The Normal and Other Continuous Distributions (The NYSE)
8.Data Sources: Observational Studies and Surveys (Roper Polls)
9.Data Sources: Experiments (Capital One)
10.Sampling Distributions and Confidence Intervals for Proportions (Marketing
Credit Cards: The MBNA Story)
11.Confidence Intervals for Means (Guinness & Co.)
12.Testing Hypotheses (Casting Ingots)
13.More About Tests and Intervals (Traveler’s Insurance)
14.Comparing Two Means (Visa Global Organization)
15.Inference for Counts: Chi-Square Tests (SAC Capital)
16.Inference for Regression (Nambé Mills)
17.Understanding Residuals (Kellogg’s)
18.Multiple Regression (Zillow.com)
19.Building Multiple Regression Models (Bolliger and Mabillard)
20.Time Series Analysis (Whole Foods Market®)
21.Introduction to Big Data and Data Mining (Paralyzed Veterans of America)
22.Quality Control (Sony)
23.Nonparametric Methods (i4cp)
24.Decision Making and Risk (Data Description, Inc.)
25.Analysis of Experiments and Observational Studies
,Solutions Manual organized in reverse order, with the last chapter displayed first, to ensure that all chapters are included
in this document. (Complete Chapters included Ch25-1)
Chapter 25 – Analysis of Experiments and Observational Studies
SECTION EXCERCISES
SECTION 25.1
1.
a) For this experiment, there are k = 4 treatments and N = 40 subjects (4×10). The degrees of
freedom for treatment are k – 1 = 4 – 1 = 3; degrees of freedom for error are N – k = 40 – 4 = 36;
and total degrees of freedom are N – 1 = 40 – 1 = 39.
b) SSE = SSTotal – SST = 1177.97 – 856.07 = 321.9.
SST 856.07
c) MST 285.36.
k 1 3
SSE 321.9
d) MSE 8.94.
N k 36
2.
a) H0: µ1 = µ2 = µ3 = µ4
HA: at least one mean is different
MST 285.36
b) F 31.92.
k 1; N k MSE 8.94
c) The P-value associated with the calculated F-statistic depends on the numerator and denominator
degrees of freedom, 3 and 36 respectively. It can be found with technology (P-value < 0.0001) or
from Table F in the Appendix. Using Table F, at α = 0.01 the F value for 3 and 35 degrees of
freedom (36 is not in the table) is 4.40. The calculated value is much larger, so we can say that the
P-value is less than 0.001.
d) At α = 0.05, we can reject the null hypothesis and conclude that not all four means are the same (at
least one mean is significantly different).
SECTION 25.2
3. Cell phone adoption.
a) The null hypothesis H0: The mean cell phone adoption rates are equal in the three regions.
The alternative hypothesis HA: The mean cell phone adoption rates are not all equal in the three
regions.
b) Countries were selected randomly from the three regions, so the randomization condition is met.
However, boxplots suggest the spreads are quite different and show several outliers which require
attention. Other plots should be used to check the Nearly Normal Condition.
4. Marketing managers’ salaries.
a) The null hypothesis H0: The mean salaries for marketing managers are equal in all three regions.
The alternative hypothesis HA: The mean salaries for marketing managers are not all equal in the
three regions.
b) States were selected randomly from the three regions, so the randomization condition is met.
Boxplots indicate that the spreads are similar and appear to be symmetric but other plots should be
used to check the Nearly Normal Condition. The boxplots do indicate that the sample data may be
somewhat skewed, but it is difficult to tell if the normal population assumption is violated. Other
plots should be used to check the normal population assumption.
25-1
, 25-2 Chapter 25 Analysis of Experiments and Observational Studies
SECTION 25.3
5. Activating baking yeast.
a) The null hypothesis H0: 1 = 2 = 3 = 4 (the means of all four groups are equal). .
The alternative hypothesis HA: not all means are equal.
b) The data provide strong evidence with a P-value < 0.0001 to reject the null hypothesis and conclude that the
means are not all equal. This experiment provides strong evidence that the mean activation times differ
among the recipes.
c) Yes. Because we have rejected the null hypothesis, we can proceed with a multiple comparisons method to
compare all the groups.
6. Activating baking yeast.
a) The null hypothesis is that the mean Distance thrown using each grip is the same. The alternative
is that not all means are equal.
b) We would not reject the null hypothesis of equal means because the P-value is 0.1543. This
experiment provides insufficient evidence that the mean Distance differs among the three grips.
c) No. We have not rejected the null hypothesis, so we cannot proceed with multiple comparisons.
SECTION 25.4
7. No. This is an observational study. There may be other factors (such as age) that are influencing the
response.
8. If the treatments were randomized, then the researchers should be justified in making the claim.
CHAPTER EXERCISES
9. Managers’ hourly wages.
a) This is an observational study. The factors are not deliberately manipulated to specific treatments.
b) Boxplots of the hourly wages by both manager type and region indicate that the equal variance
assumption is reasonable. Also, the boxplots indicate fairly symmetric distributions. However,
other plots should be used to check the Nearly Normal Condition once the model has been fit. The
states were randomly selected from the three regions, so the randomization condition is satisfied.
c) Based on the P-values, we can conclude that there is no significant interaction effect. There is no
evidence that the mean hourly wages are different across the three regions but are different for
sales and advertising managers.
d) Yes, because the interaction effect is not significant.
10. Concrete testing.
a) This is an experiment. The factors are deliberately manipulated to specific treatments.
b) Boxplots of the compressive strengths by both alkali content and sand type indicate that the equal
variance assumption is reasonable. Also, the boxplots indicate fairly symmetric distributions.
However, other plots should be used to check the Nearly Normal Condition once the model has
been fit. The experimental units were selected randomly for each treatment, so the randomization
condition is satisfied.
c) Based on the P-values, we can conclude that there is no significant interaction effect. The mean
compressive strengths are not the same across the three different alkali contents and not the same
for the three sand types.
d) Yes, because the interaction term is not statistically significant. There is no evidence that the effect
of alkali content differs across the three sand types or that the effect of sand type differs across the
three alkali contents.
11. Cars (fuel efficiency).
