1. What question are you trying to answer? Why is it important? Or why is it interesting?
2. What is your dependent variable?
3. What are your independent variables (must have at least two, and the sample size must be at least 50)?
4. How do you model their relationship? i.e. what functional form will you use? You have many choices: OLS (linear, quadratic, dummy variable, log-linear, log-log), IV regression, Differencein-Difference, Fixed effect, Probit, and so forth). Running a simple linear regression model is totally acceptable for the course.
5. Where will you get the data? After you get the data, organize it so that it is ready for regressions. (Come to my office hours if you need help for data management)
1. Search for similar works
2. Describe what these studies find using your own words. (copy and paste will be treated as plagiarism)
3. How is your research different? What is your contribution? (even if you only make a marginal contribution)
1. Report regression results in a table. Include coefficients, standard errors, significant levels, number of observations, and R-square estimates.
2. Interpret your coefficients.
3. What is your conclusion based on the results? What are the possible problems with your findings?
1. Do not try to be ambitious. As long as your paper economically make sense, employing a simple linear regression model is fine for the course.
2. If you do not want to write an original paper, you have another choice: replicate the empirical results of an existing study or updating them by applying a different econometric model or using different data sets (updated data or data from a different country/state etc.) Then you need to report to me whether you successfully replicated the paper (100% replication is very difficult). If not, what are the reasons for the different results?