1

I have a query on how to interpret the result for multiple regression with categorical variables. I have categorical variable called Stay_In_current_city_years which has 5 levels. After running the linear model I get an output with 4 different p-values.

How do I interpret this result? Does it mean that Stay_In_current_city_years1 with p-value=.0746 is not a significant predictor and Stay_In_Current_City_Years2 with p-value=.005 is a significant predictor? If so how would remove the variables which are not statistically significant? enter image description here Thanks

User 3400
  • 11
  • 3

1 Answers1

0

Can you specify the levels of the stay_in_current_city_years variable? One of the categories becomes the reference category and that's why you have 4 p-values. The categories with p-values > 0.05 are considered to be not significantly different from the reference category with respect to their effect on the outcome variable.

NB21
  • 36
  • 4
  • Thank you for the quick response. The Stay_In_Current_City_Years variable has 5 levels where Stay_In_Current_City_Years0 is taken as reference category. So in this case how can I conclude that Stay_In_Current_City_Years variable as a whole is statistically significant, since the levels Stay_In_Current_City_Years1,Stay_In_Current_City_Years2, Stay_In_Current_City_Years4 is statistically insignificant. Does it mean that I have to remove the variable Stay_In_Current_City_Years while running a multiple regression to predict the response? – User 3400 Mar 24 '19 at 16:49