Understanding control variables, and how to use them

[Quinn Fraser]

Hi everyone,

I'm having difficultly trying to understand what control variables are, and how to use them in my regression analysis.

For what it's worth, I'm studying support for income redistribution in South Africa.

To my understanding, my dependent variable is level of support for income redistribution. My independent variables are ones that I'm trying to test to see the effect on the dependent. So what does it mean to control for a particular variable? Is there a difference between a control and independent variable? Is this what different models are? I understand it's a variable that remains constant, but what exactly does that mean in STATA? Is there a difference in how I code control and independent variables in STATA?

I appreciate any help or support.

Here is a list of variables that I'm looking at for my study:
- sex (M, F)
- age (grouped by 20's, 30's, etc)
- religiousness (1-10 scale)
- political ideology (1-10 scale, right=1, left=10)
- income (1-10 scale, 1=poorest)
- population of city (grouped by tens of thousands)
- ethnicity (white, black, coloured),
- education level (no school, primary, secondary, post-sec)
- union membership (yes, no)
- employment sector (unskilled, semi-skilled, skilled)

[Maarten Buis]

There is no difference between independent variables and control variables as far as Stata is concerned. When you write up your results you will spent a lot more space on describing the effects of the independent variables (as those are the effects you are interested in) than on the effects of control variables.