13.10 Lab: Data

  • Data and files available under the link given in Section 1.3.

The dataset is fouirnaies_hall_financial_incumbency_advantage.dta, the data provided by Skovron and Titiunik (2015) in their best practices paper. The data and the original code (on which the code below is based) can also be found in the replication files provied by Skovron and Titiunik (2015) for which you can find a link in their paper. Importantly, for the illustration Skovron and Titiunik (2015) use a subset of the original data in which they filter out U.S. House legislature and focus on state legislatures:

  • “We focus only on Fouirnaies and Hall’s analysis of state legislative elections, using 32,670 races at the state legislative district level. The score [variable] is the Democratic margin of victory at t in the district (the vote percentage obtained by the Democratic party minus the vote percentage obtained by its strongest opponent), and the treatment is winning election t. The Democratic party wins the election when its margin of victory is positive and loses when it is negative, so the cutoff is zero.The outcome we analyze is the Democratic share of total contributions in the district at t + 1” (Skovron and Titiunik 2015, 29)

  • cov_statelevel (subset variable): Dummy for state level elections (we use only this subset of elections!)

  • x_score_victorymargin (running variable): Democratic margin of victory at t in the district (the vote percentage obtained by the Democratic party minus the vote percentage obtained by its strongest opponent)

  • y_donationshare (outcome variable of interest): Outcome variable share of donations flowing to the incumbent’s party in percent

  • cov_total_race_money (Covariate as outcome): Total money in race

  • cov_total_votes (Covariate as outcome): Total votes in race

  • cov_dem_inc (Covariate as outcome): Democratic incumbent

  • cov_rep_inc (Covariate as outcome): Republican incumbent

  • cov_total_group_money (Covariate as outcome): Total group money

References

Skovron, Christopher, and Rocıo Titiunik. 2015. “A Practical Guide to Regression Discontinuity Designs in Political Science.”