In: Statistics and Probability
Consider the following partially completed two-way ANOVA table. Suppose there are 4 levels of Factor A and 2 levels of Factor B. The number of replications per cell is 4. Use the 0.01 significance level. (Hint: estimate the values from the F table.)
Complete an ANOVA table. (Round MS and F to 2 decimal places.)
Sources: Factor A, Factor B, Interaction, Error, total.
SS: 70, 50, 210, 400, 730
Find the critical values to test for equal means. (Round your answers to 2 decimal places.)
Determine if there is a significant difference in Factor A means, Factor B means.
Determine if there is a significant difference in interaction means.
To complete the table above, we should first obtain the table of formulas needed.
n= 4 * 2 *4 = 32
ANOVA |
||||
Source |
SS |
df |
MS |
F |
Factor A |
Factor A |
k-1 |
MSA= SSA/(k-1) |
MSA/MSE |
Factor B |
Factor B |
b-1 |
MSB = SSB/(b-1) |
MSB/MSE |
Interaction |
SSI |
(k-1)(b-1) |
MSI=SSI/(k-1)(b-1) |
MSI/MSE |
Error |
SSE |
n-kb |
MSE = SSE/(n-kb) |
|
Total |
SS Total |
n-1 |
Therefore, the completed ANOVA table is as follows:
ANOVA |
||||
Source |
SS |
df |
MS |
F |
Factor A |
70 |
3 |
23.33 |
1.339 |
Factor B |
50 |
1 |
50 |
2.99 |
Interaction |
210 |
3 |
70 |
4.199 |
Error |
400 |
24 |
16.667 |
|
Total |
730 |
31 |
To test whether there is a significant difference in factor A means, we will use critical value from F table with 0.05 significance level, df numerator (k-1) = 3, and df denominator (n-kb) = 24. The critical value is 3.01. The computed value of factor A is smaller than the critical value, therefore, we fail to reject the null hypothesis. We can conclude that the factor A means are equal.
To test whether there is a significant difference in factor B means, we will use critical value from F table with 0.05 significance level, df numerator (b-1) = 1, and df denominator (n-kb) = 24. The critical value is 4.26. The computed value of factor B is smaller than the critical value, therefore, we fail to reject the null hypothesis. We can conclude that the factor B means are equal.
To test whether there is a significant effect from the interaction between factor A and factor B, we will use critical value from F table with 0.05 significance level, df numerator (k-1)(b-1) = 3, and df denominator (n-kb) = 24. The critical value is 3.01. The computed value of interaction is higher than the critical value, therefore, we reject the null hypothesis of no interaction. We can conclude that the combination between factor A and factor B has a significant effect on the response factor.