Greetings to the wonderful Stata community,

I have a question regarding about interpretation of dummy variables. I am doing an analysis on homicide rates in developing countries. One of the explanatory variables in my model is religion. The countries in my analysis have 4 major religions. Hinduism, Christianity, Islam and Atheism (I know atheism is not a religion, but none the less it is an associated belief to not believe in God).

Basically I wish to see the impact of religion on homicides on the countries in my dataset. So I come up with 4 Dummy variables, one for each religious belief. Now my question is when I run the regression (by the way I am running a random effects model), how can I get the results of all of Dummy variables? Because STATA excludes the last Dummy variable due to Co linearity. That means I have interpretation available only on three religion dummies with the fourth being my reference variable that is omitted. Can someone please guide me on how to estimate results for all 4 dummy variables? Otherwise it represents bias in my research for purposely excluding one religion, while interpreting the impacts other 3 religions have.

I await your response.