ANOVA Diagnostics Module. Stats Design & Analysis Web Guide (DAWG), University of Tennessee

ANOVA Diagnostics

Start 6 of 8

Check for outliers, equal variance, normality, and model validity

D. Check model validity.

Step D are diagnostic checks that are needed if you have a covariate or blocking in your model. If you don't, go to Step E.

Step D1: If you selected Covariate in choosing my design, then you need to go Here to check covariate model, and then you will be brought back to this page.
Check covariate model

Step D2: If your experimental design is not a CRD, then you have blocks in your model. Go Here to verify that you have no block*treatment, and then you will be brought back to this page.
Verify that you have no block*treatment

When you have completed these checks, go to Step E

next >> ( Step E: Finish the analysis).

H I N T S :

Remember, data transformations can fix apparent outliers.

If you rerun your SAS analysis, always check all diagnostics. For example, removing some outliers may cause
other outliers to be identified.

Do not be concerned if you have to rerun your SAS analysis many times. Correction of diagnostics may take
several tries.