A study was done to measure customer satisfaction for 4 different smart phones. The data is given below (the data is a satisfaction score, out of 100; the higher the score, the more satisfied the customer was with their phone)
1. iPhone | 2. Nokia | 3. Blackberry | 4. Android | |
Sample Size | 5 | 4 | 3 | 4 |
Sample Mean | 83.0 | 66.5 | 73.0 | 77.0 |
Overall Mean = 75.5
MSE = 20.54
a) Run a one-factor ANOVA and test H0: μ1 = μ2 = μ3 = μ4 at the 5% level of significance. Set up an ANOVA table (show your work calculating SSC, sum of squares between groups), state H0 and Ha in words, calculate the value of your test statistic, and make your decision/draw your conclusion. Make sure your conclusion is given in words that are in the context of the question.
b) Use the Tukey-Kramer procedure to test H0: μ1 – μ3 = 0 vs Ha: μ3 – μ4 ≠ 0 at the 5% level of significance. What is your decision? You must also clearly state your conclusion in words
(a) The hypothesis being tested is:
H0: µ1 = µ2 = µ3 = µ4
Ha: At least one means is not equal
xi | n | n*(xi - xgrand)² | |||
iPhone | 83 | 5 | 281.25 | ||
Nokia | 66.5 | 4 | 324 | ||
Blackberry | 73 | 3 | 18.75 | ||
Android | 77 | 4 | 9 | ||
xgrand | 75.5 | SSB | |||
633 | |||||
Source | SS | df | MS | F | p-value |
Between | 633 | 3 | 211 | 10.27264 | 0.001238 |
Error | 246.48 | 12 | 20.54 | ||
Total | 879.48 | 15 |
The p-value is 0.001238.
Since the p-value (0.001238) is less than the significance level (0.05), we can reject the null hypothesis.
Therefore, we can conclude that at least one means is not equal.
(b)
Group | Mean Difference | HSD | Significant Difference |
μ1 – μ3 | 10 | 5.099229 | Yes |
Therefore, we can conclude that μ3 – μ4 ≠ 0.
Get Answers For Free
Most questions answered within 1 hours.