In: Statistics and Probability
In an effort to promote a new product, a marketing firm asks participants to rate the effectiveness of ads that varied by length (short, long) and by type of technology (static, dynamic, interactive). Higher ratings indicated greater effectiveness.
Source of Variation | SS | df | MS | F |
---|---|---|---|---|
Length | 10 | |||
Technology | ||||
Length × Technology | 154 | |||
Error | 570 | 114 | ||
Total | 824 |
(a) Complete the F-table and make a decision to retain or reject the null hypothesis for each hypothesis test. (Assume experimentwise alpha equal to 0.05.)
Source of Variation |
SS | df | MS | F |
---|---|---|---|---|
Length | 10 | |||
Technology | ||||
Length
× Technology |
154 | |||
Error | 570 | 114 | ||
Total | 824 |
Source of Variation |
SS | df | MS | F | P-value |
---|---|---|---|---|---|
Length | 10 | 1 | 10 | 2 | 0.1600 |
Technology | 90 | 2 | 45 | 9 | 0.0002 |
Length × Technology | 154 | 2 | 77 | 15.4 | 0.0000 |
Error | 570 | 114 | 5 | ||
Total | 824 | 119 |
Let Factor A = Length,
Factor B = Technology
Calculation for each of the missing value in table is shown below:
SSTotal = SSA + SSB + SSAxB + SSError
824 = 10 + SSB + 154 + 570
SSB = 824 - (10 + 154 + 570)
SSB = 90
dfA = a - 1 (Here, a is no. of levels of factor A)
dfA = 2 -1 = 1
dfB = b - 1 (Here, a is no. of levels of factor B)
dfB = 3 -1 = 2
dfAxB = (a - 1)*(b - 1)
dfAxB = 1*2 = 2
dfTotal = dfA + dfB + dfAxB + dfError
dfTotal = 1+ 2 + 2 + 114
dfTotal = 119
MSA = SSA/dfA = 10/1
MSA = 10
MSB = SSB/dfB = 90/2
MSB = 45
MSAxB = SSAxB/dfAxB = 154/2
MSAxB = 77
MSError = SSError/dfError = 570/114
MSError = 5
Fstatistic (Factor A) = MSA/MSError = 10/5 = 2
Fstatistic (Factor B) = MSA/MSError = 45/5 = 9
Fstatistic (Factor AxB) = MSAxB/MSError = 77/5 = 15.4
P-value corresponding to Fstatistic (Factor A) is 0.1600 (Obtained using online p-value calculator. screenshot attached)
P-value corresponding to Fstatistic (Factor B) is 0.0002 (Obtained using online p-value calculator. screenshot attached)
P-value corresponding to Fstatistic (Factor AxB) is 0.0000 (Obtained using online p-value calculator. screenshot attached)
Since p-value for Factor A (0.1600) is less than 0.05, there is no statistically significant difference in effectiveness of ads based on length.
Since p-value for Factor B (0.0002) is less than 0.05, there is statistically significant difference in effectiveness of ads based on technology.
Since p-value for Interaction between A and B (<0.0000) is less than 0.05, it shows that interaction effect is statistically significant.