# We said the data is normally distributed, based on the raw data or residual? [duplicate]

I have a confusing regarding the assumption test for the data, in some theory were said that there are three assumption of data as we called as "good" data:

1. Independent
2. Normally distributed
3. Homogeneity

my question are: Referring to point (2) and (3),

1. We said the data is normally distributed, based on the raw data or residual? I have a case when i was conducting the ANOVA analysis, the assumption for ANOVA is normal and homogeneity, what i did until now is to check this through the residual (after i got the model ANOVA), is it true?

2. if those assumption come from raw data not residual, how I can check the homogeneity? bartlett is using residual that I know to check the homogeneity..

In general, we're referring to the distribution of the residuals in these models. The distribution of $X$ in a linear regression or ANOVA model may be highly irregular, or sampled in a sequential fashion (e.g. a stratified sample of 10 people within each age group defined as 10-19, 20-29, ...). When we fit such models, we think of the $Y$s as being conditional upon the $X$s.