Covariance Parameters (generalized linear mixed models)
- Click the Covariance Parameters view thumbnail.
This view displays the covariance parameter estimates and related statistics for residual and random effects. The residual (R) covariance parameters are shown by default. There are no repeated measures, so there is a single variance estimate for the residuals.
Figure 2. Covariance Parameters view, random effects block 1 
- From the Effect dropdown of the Covariance Parameters view, select Block
1.
This is the variance estimate for the intercept of the random effect with school defining subjects, and is a measure of the between-school variation. The estimate is large relative to the residual variance and (looking ahead) the variance for Block 2, suggesting that most of the variability in test scores that is not explained by the fixed effects can be explained by school-to-school variation, but the standard error is also large, so there is a lot of uncertainty concerning the actual size of this effect.
Figure 3. Covariance Parameters view, random effects block 2 
- From the Effect dropdown of the Covariance Parameters view, select Block 2.
This is the variance estimate for the intercept of the random effect with school*classroom defining subjects, and is a measure of between-classroom variation. The size of the effect is roughly equivalent to the residual variance, but with a little more uncertainty in the estimate; that is, a larger standard error.