 |
 |
 |
|
|
|
Simple Linear Regression (using the method of least squares)
For 
Using:
We determine least squares estimates with: 
1 Way Analysis of Variance: Equal Replication r
| Source | DF | SS | MS | F |
| Trt | t-1 |  |
SS(T)/(t-1) | MS(T)/MS(E) |
| Error | t(r-1) |  |
SS(E)/(t(r-1)) | |
| Total | rt-1 |  |
2 Way Analysis of Variance
| Source | DF | SS | MS | F |
| Blocks | r-1 |  |
SS(B)/(r-1) | MS(B)/MS(E) |
| Trt | t-1 |  |
SS(T)/(t-1) | MS(T)/MS(E) |
| Error | (t-1)(r-1) |  |
SS(E)/(t(r-1)) | |
| Total | rt-1 |  |
Contrasts
Using totals, define a contrast as: 
The corresponding contrast sum of squares is: 