This document outlines the steps needed for the reproduction files associated with "Should We Talk about the Weather? How Party Competition and Coalition Participation Influence Parties' Attention to Economic Issues" by Edward Goldring, Brandon Beomseob Park and Laron K. Williams, Party Politics. If you have questions please contact williamslaro@missouri.edu.
These files require both Stata and R. In R, you will need the grid, foreign, and ggplot2 packages from CRAN. The script file "Issue Attention--Replication.do" reproduces all summary statistics, estimates, substantive effects and robustness checks shown in the paper. The R file Issue Attention--Figures.R produces both figures. All needed datasets are provided and referenced in the script files. Users should change working directories as needed.
Although you can run all files in an interactive fashion, we recommend running each file as a batch and then reading the associated log file ("Spatial Issue Competition--Replication.smcl"). For convenience, we provide a PDF of the Stata log file in the event a user does not have that program.
The dataset "Germany.dta" is our primary dataset for the results and figures. These data are based on Sagarzazu and Kluver (2017). These data are used in various levels of aggregation and shape in order to make the figures, and each figure includes its own datasets ("Emphasis.dta" and "Spatial-Long-Term Effects.dta"). Note that we set a seed for starting points that will need to be maintained in order to reproduce the datasets as we provide them.
We also include several ancillary files needed related to the two weighting matrices. The file "Generate Weights Matrices.do" demonstrates how we generated the weights matrices ("Wcoalition_rs.dta" and "Wcontiguity_rs.dta").
The figures are all created in using the "Issue Attention--Figures.R" file in R. Each of the figures imports a dataset produced in the do file discussed above and in the paper.