Hi Statalists,
I am trying to find the relationship between the change in labour productivity (dependent variable) and immigrants occupational level (independent variable). My unit analysis is by sector. From 2010-2016. Right now I am going to include dummy variable (manager, senior office and etc = is equal to 1 for high skilled occupations and vice versa) and interaction term sector*highskill into the regression. However the dummy is omitted because of collinearity as shown at the end of the post***. So my question is.....
If my focus is on the interaction term (i.e. how immigrants with different occupational level (high or low skilled occupations) affects labour productivity in each sector), how can I do that?
I've attached a picture of my final dataset which I use for my regression
Thanks in advance
Lisa
Related Posts with The problem with creating dummy variables
create a new var extracting part of string with different length- Variability over time of positive and negative values
- Difficulty aggregating variables to the unit of analysis without losing or altering data
- create a new var extracting part of string with varying length
- bysort gen n = _n with if condition 1/2/3/4 is true
Subscribe to:
Post Comments (Atom)
0 Response to The problem with creating dummy variables
Post a Comment