12.4: Testing the Significance of the Carry-Over Effect
- Page ID
- 33198
To test for the overall significance of carry-over effects, we can drop the carry-over covariates (\(x_{1}\) and \(x_{2}\) in our example) and re-run the ANOVA. Because the reduced model is a subset of the full model that includes the covariates, we can construct a likelihood ratio test. \[\Delta G^{2} = \left(-2 \log L_{Reduced}\right) - \left(-2 \log L_{Full}\right) \quad \text{with } df_{Reduced} - df_{Full} \text{ degrees of freedom}\]
The \(-2 \log L\) values are provided in the SAS Fit Statistics output for each model. For our example, the SAS output for the Full model with carry-over covariates is:
Fit Statistics |
|||||
---|---|---|---|---|---|
-2 Res Log Likelihood |
122.5 |
||||
AIC (smaller is better) |
130.5 |
||||
AICC (smaller is better) |
132.6 |
||||
BIC (smaller is better) |
132.5 |
And for the reduced model without the carry-over covariates is:
Fit Statistics |
|||||
---|---|---|---|---|---|
-2 Res Log Likelihood |
136.5 |
||||
AIC (smaller is better) |
144.5 |
||||
AICC (smaller is better) |
146.4 |
||||
BIC (smaller is better) |
146.4 |
So, \[\Delta G^{2} = 136.5 - 122.5 = 14 \nonumber\] and with \[\chi_{.05, 2}^{2} = 5.991 \nonumber\] we conclude that there are significant carry-over effects.