In: Statistics and Probability
Consider the data in the table collected from three independent populations. |
Sample 1 |
Sample 2 |
Sample 3 |
||
---|---|---|---|---|---|
4 |
3 |
4 |
|||
a) Calculate the total sum of squares (SST) and partition the SST into its two components, the sum of squares between (SSB) and the sum of squares within (SSW). |
2 |
2 |
2 |
||
9 |
1 |
1 |
|||
5 |
b) Use these values to construct a one-way ANOVA table.
c) Using alphaαequals=0.05,
what conclusions can be made concerning the population means?
For the given data using Anova Single Factor in Excel we get output as
Anova: Single Factor | ||||||
SUMMARY | ||||||
Groups | Count | Sum | Average | Variance | ||
Sample 1 | 3 | 15 | 5 | 13 | ||
Sample 2 | 3 | 6 | 2 | 1 | ||
Sample 3 | 4 | 12 | 3 | 3.333333 | ||
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Between Groups | 14.1 | 2 | 7.05 | 1.298684 | 0.331368 | 4.737414 |
Within Groups | 38 | 7 | 5.428571 | |||
Total | 52.1 | 9 |
( a ) From the above output
SST = 52.1
SST = SSB + SSW
= 14.1 + 38
= 52.1
( b ) one-way ANOVA table.
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Between Groups | 14.1 | 2 | 7.05 | 1.298684 | 0.331368 | 4.737414 |
Within Groups | 38 | 7 | 5.428571 | |||
Total | 52.1 | 9 |
( c ) P value < l.o.s
0.3314 > 0.05
So fail to reject H0
At α = 0.05 l.o..s there is enough evidence that all population means are equal