In: Statistics and Probability
The calculations for a factorial experiment involving four levels of factor A, three levels of factor B, and three replications resulted in the following data: SST = 272, SSA = 25, SSB = 24, SSAB = 165.
|
a.
Source of Variation | Sum of Squares |
|
Mean Square | F | p-value | |
Factor A | 25 | 3 | 8.33 | 3.44 | 0.0327 | |
Factor B | 24 | 2 | 12 | 4.95 | 0.0159 | |
Interaction | 165 | 6 | 27.5 | 11.36 | 0 | |
Error | 58 | 24 | 2.42 | |||
Total | 272 | 35 |
SS Error - SS Total - (SSA + SSB + SSAB) = 272 - (25 + 24 + 165) = 58
Given a = 4, b = 3, r = 3
DFA = a - 1 = 4 - 1 = 3
DFB = b - 1 = 3 - 1 = 2
DFAB = (a-1) * (b-1) = 3 * 2 = 6
DF Error = ab(r - 1) = 4 * 3 * (3 - 1) = 24
F = MS / MS for Error
DFT = abr-1 = 4 * 3 * 3 - 1 = 35
Mean Square = Sum of Squares / DF
P-value for A for df = 3, 24 is 0.0327
P-value for B for df = 2, 24 is 0.0159
P-value for A for df = 6, 24 is 0
b.
The p-value for Factor A is between .02 and .05
Since p-value is less than 0.05, Factor A is significant
c.
The p-value for Factor B is between .01 and .025
Since p-value is less than 0.05, Factor B is
significant
d.
The p-value for the interaction of factors A and B is
less than .01
Since p-value is less than 0.05, The interaction of factors A and B
is significant.