Homework Assignment 7

1. Suppose you are planning an experiment to test the latest conflicting diet programs. The
factor for diet will involve three different levels: low fat, low carbohydrate, and low calorie.
The other factor will involve planned exercise. One level will be coached aerobics and the
other will be normal activity. We hope to find out which regime is best at reducing weight.

The degrees of freedom for interaction are:

a) 2.
b) 5.
c) 6.
d) 7

2. A major benefit of two-way ANOVA over several one-way ANOVAs is:

a) two-way is more efficient and gives more information.
b) two-way takes less time.
c) two one-ways are better than one two-way.

3. If an F test for interaction is significant:

a) one always checks to see if the two single factors are significant.
b) there may be interest in the individual factors.
c) a careful examination of the means is necessary to fully understand the data.
4. An experiment was conducted to examine the yield of six different types of alfalfa. They were
planted in four different locations, so we will consider the field as a second factor because
different soil and moisture types can also have an effect on yield. The interactions plot below
was obtained.

This plot indicates:

a) an effect for alfalfa type only.

b) an effect for field only.
c) that interactions are present.

5. An experiment was performed to investigate lifetime of a component exposed to high heat.

Three different oven types were used at either 550°F or 600°F. The means plot below was
obtained.

This plot suggests that:

a) An interaction effect is present.

b) That there are main effects of oven and temperature.
c) That there are no main effects or interactions.
 6. A study of cardiovascular risk factors compared heavy smokers, light smokers, and
nonsmokers. Both men and women were included in the study and gender differences were
also to be investigated. One of the variables measured was heart rate after six minutes of
exercise.

The two factors being studied in this experiment are:

a) men and women.

b) heart rate and smoking level.
c) gender and smoking level.

7. Which of the following statements about one-way analysis of variance is/are TRUE?

I. The total sums of squares represent variation present in the data, and they are calculated by
summing squared deviations.

II. There are three distinct sources of variation represented in the table: groups, error, and
total.

III. The test statistic is the ratio of sum of squares for groups (SSG) to the sum of squares due to
error (SSE).

IV. The sum of squares for total (SST) is composed of two parts, one due to groups (SSG) and one
due to error (SSE).

a) I and II only
b) I, II, and III only
c) I, II, and IV only

8. A farmer is trying to figure how to yield the largest amount of a particular type of crop. He is
trying to choose between three fertilizers (A, B, and C) and four different types of protein
additives (A, B, C, and D).

The factors being studied are:

a) all combinations of fertilizer and protein additives.

b) fertilizer and protein additive.
c) crop yield.
 9. The degrees of freedom for interaction are:
a) 3.
b) 6.
c) 12.
d) 14

10. The response variable is:

a) protein additive.
b) fertilizer.
c) crop yield.

11. A scientist examined the effect of temperature and catalyst on the yield of a reaction. He ran
the experiment at both 150° and 200°C with two types of catalyst. A partial ANOVA table is
shown below.

Analysis of Variance

Source DF SS MS

Temp 1 30.54 30.54

Catalyst 1 0.10 0.10

Interaction 1 0.03 0.03

Error 12 39.09 3.26

Total 15 69.77

The F statistic for interaction is:

a) 0.009.
b) 0.0008.
c) 0.006.
12. Suppose you are planning an experiment to test the latest conflicting diet programs. The
factor for diet will involve three different levels: low fat, low carbohydrate, and low calorie.
The other factor will involve planned exercise. One level will be coached aerobics and the
other will be normal activity. We will have 25 subjects assigned to each treatment
combination. We hope to find out which regime is best at reducing weight.

The degrees of freedom for error are:

a) 149.
b) 144.
c) 5.

13. Researchers tested the effects of hypnosis, nicotine gum, nicotine patch, and nicotine
inhaler. They believed the effects may differ for men and women, so men and women were
randomized to the treatments separately. Twenty-five subjects are randomized into each
treatment combination. They obtained the following results:

Source DF SS MS F

Gender 1 386.2

Method 3 479.3

Interaction 3 1286.2

Error 192 970.2

Total 199

The value of the MS-Method is:

a) 5.05.
b) 159.77.
c) 386.2.
d) 428.73.
14. An experiment tests the effects of new devices to help stop smoking. They tested hypnosis,
nicotine gum, nicotine patch, and nicotine inhaler. They believed the effects may differ for
men and women, so men and women were randomized to the treatments separately.
Twenty-five subjects are randomized into each treatment combination. They obtained the
following results:

Source DF SS MS F

Gender 1 386.2 386.2

Method 3 479.3 159.77

Interaction 3 1286.2 428.73

Error 192 970.2 5.05

Total 199

The value of the F statistic for Method is:

a) 31.62.
b) 32.93.
c) 23.71.
 15. A study of cardiovascular risk factors comparing smoking level and gender was funded. There
were 25 subjects included in each of the combinations of gender and smoking status. One of
the variables measured was the heart rate after six minutes of exercise. A two-way ANOVA is
performed with the resulting ANOVA table given below.

Source DF SS M F P

Smoking 2 30850.4 15425.2 41.16 0.000

Gender 1 331.5 331.5 0.88 0.349

Smoking*Gender 2 341.0 170.5 0.46 0.635

Error 144 53961.1 374.7

Total 149 85484.0

We can conclude from the ANOVA table that:

a) the effect of smoking on heart rate appears to be the same for men and women.
b) there is strong evidence that heart rate increases with the amount smoked.
c) both (a) and (b) are correct.

16. The mean blood pressures of male employees of a large company are to be compared. An
SRS of 5 male employees from each of the following levels of employment are selected:
Executive level, Assistant level, Intern level. It is of interest to test the hypothesis of equality
of mean blood pressure level for the three levels of employment. The test statistic is:
a) an F statistic with 2 and 12 df.
b) an F statistic with 3 and 15 df.
c) an F statistic with 2 and 15 df.

17. The mean blood pressures of male employees of a large company are to be compared. An
SRS of 5 male employees from each of the following levels of employment are selected:
Executive level, Assistant level, Intern level. It is of interest to test the hypothesis of equality
of mean blood pressure level for the three levels of employment. If we reject the null
hypothesis we conclude that:
a) none of the mean blood pressures equal the others.
b) at least one mean blood pressure differs from the others.
c) all the blood pressures differ.
 Answers
1. A
2. A
3. C
4. C
5. B
6. C
7. C
8. B
9. B
10. C
11. A
12. B
13. B
14. A
15. A
16. A
17. B