I used LM model with (categorical) predictor variables on my data in r like this (I have count variable as dependent/target variable):

# Fit linear model to total viewing:
fit <- lm(
  total_viewing_minutes ~ panelist + sex + internet + education,
  data = cln

If I understood correctly in r the first category is always the reference one so has by definition a coefficient of 0.

So by looking at the model result, is 68.3 minutes (intercept) average time viewing of

Panelist0 + sex1 + internet1 + education1  

and by taking into account the term Panelist1
the 68.4 + 17.6 is avg time viewing of

 Panelist1 + sex1 + internet1 + education1  

is that correct interpretation?

enter image description here


1 Answer 1


Your interpretation is correct. I would also add that if Panelist1 increases by 1 unit and the other predictors are held constant, then the average increase in the response variable is 17.557 minutes.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.