In: Statistics and Probability
The Kenton Food Company wished to test four different package designs
for a new breakfast cereal. Twenty stores, with approximately equal sales volumes, were
selected as the experimental units. Each store was randomly assigned one of the package
designs, with each package design assigned to five stores. A fire occurred in one store during
the study period, so this store had to be dropped from the study. Hence, one of the designs
were tested in only four stores. The stores were chosen to be comparable in location and
sales volume. Other relevant conditions that could affect sales, such as price, amount and
location of shelf space, and special promotional efforts, were kept the same for all of the
stores in the experiment. Sales, in number of cases, were obtained for the study period.
Each of Four Package Designs | |||||||||
Store (replication) | |||||||||
Package Design | 1 | 2 | 3 | 4 | 5 | ||||
1 | 11 | 17 | 16 | 14 | 15 | ||||
2 | 12 | 10 | 15 | 19 | 11 | ||||
3 | 23 | 20 | 18 | 17 | |||||
4 | 27 | 33 | 22 | 26 | 28 |
Conduct a test to determine whether or not the mean response rates for the packages differ. Solve the hypothesis testing about the equivalence of the package designs. Use level of significance a=.10. State the alternatives, decision rule, and conclusion. What is the p-value of the test.
Since, the effecting variables are all controlled for different stores, this problem can be modeled using one way ANOVA, where -
Null hypothesis - There is no significant difference in the mean response rate for all the pairs of package design
Alternate hypothesis - There is significant difference in the mean response rate for at least one pair of package design
Package Design | 1 | 2 | 3 | 4 | 5 |
1 | 11 | 17 | 16 | 14 | 15 |
2 | 12 | 10 | 15 | 19 | 11 |
3 | 23 | 20 | 18 | 17 | |
4 | 27 | 33 | 22 | 26 | 28 |
Summary of Data | |||||
Treatments | |||||
1 | 2 | 3 | 4 | Total | |
N | 5 | 5 | 4 | 5 | 20 |
∑X | 73 | 67 | 78 | 136 | 370 |
Mean | 14.6 | 13.4 | 19.5 | 27.2 | 18.5 |
∑X2 | 1087 | 951 | 1352 | 3762 | 7598 |
Std.Dev. | 2.3 | 3.6 | 2.3 | 4 | 6.3 |
Result Details | ||||
Source | SS | df | MS | |
Between-treatments | 585 | 3 | 195 | F = 18.57143 |
Within-treatments | 158 | 15 | 10.5 | |
Total | 753 | 18 |
Where SS = sum of squared errors and df = degrees of freedom
F = Ratio variance between between-treatments variance and within-treatments variance
Decision rule -
The p-value for F value = 18.57 and numerator df =3 and denominator df = 15 from the normal distribution table is .000018
The p value is less than 0.1 (significance level)
Hence, the result is significant
There is significant difference in the mean response rate for at least one pair of package design