After performing an ANOVA, if it fails to reject H0 how we re-arrange the model?

I think you should report the results as is. Maybe there is another model that is more suitable. But perhaps that should be left for a different analysis.


TS Contributor
The goal of an analysis isn’t to “make it significant,” I agree with @Buckeye that you should report it as is. If the omnibus test isn’t significant then that’s what you’re left with.