For example, If I have to understand the relation between the age and dependent variable (Likert scale data) and my sample size is uneven across the age categories (like in the age category 20-30 years old, there are 10 individuals while in the age category 30-40 years old, there are 20 individuals), what can be done in this case?
The standard assumption in correlation and regression is that variables such as age are normally distributed, although it is rare to check this assumption in actual practice.
David L Morgan Yes, that is where I was stuck in the analysis. I thought that my data is non-normal and the sample size is uneven then how do I go head with my analysis
There is very little likelihood that any reviewer will require a test for nomaliity on an age distribution, but you can always list it as a "limitation" in your data.
If you have the actual ages and not just the categories of age you might try treating age as a continuous variable. Remember it's the residuals that need to be normally distributed not the independent variables themselves.
John
- Badges
- Science topic