1. First, run a model with your dependent variable, your key independent variable and *all* the control variables. Please see the do file I emailed y'all for example codings of all the control variables

2. You should operationalize your independent variables as you think most appropriate. Note that I offer a number of different operationalizations for a few variables; you should decide which operationalization you think is best. Best might be a function of theory. Alternatively, you might try running a model multiple times, each time using a different operationalization of the same variable to see which operationalization appears to be most appropriate. To determine which is most appropriate, you can look at a variety of factors such as the adjusted R2 of the models, the individual and/or joint significance of the relevant variables, etc....

3. Put your results into a Table.

4. Interpret your results (do this carefully!). If you dummied out into multiple variables one or more "conceptual" variable, do the coefficients for the variable you dummied out suggest that the variable has a non-linear relationship with the dependent variable? Use an F-test to determine if the dummied out variables of a specific factor are jointly significant

5. Also discuss the model's Adjusted R-square and F-value.