Replication Material used in

The Competitive Road to Proportional Representation: Electoral Regime Change under Increasing Party Competition. World Politics, 2009, 61(2):254-295.

 

NOTE: You will need to install R 2.6 or higher and Winbugs.

1.  Stata 9 file with seat-vote data (geneurope_v3.dta), as described in the article.

 

2. If you are interested in loading the data in R 2.7 and replicating the results reported in the article use this replication code. 

To run the replication code, create a folder “C:\calvo" and place the Stata File "geneurope_v3.dta" in folder. (you can use a different directory but you will have to also change the working directory in the code).

 

Copy and Paste the code from the replication file all at once (the code will not run if WinBugs is not installed).

 

    3. To simulate Geographically Induced Partisan Biases use the following code.

         Copy and paste the simulation code in R 2.7.

   NOTE: To graph the simulations you need to install from CRAN the libraries SP and SPATIAL in R 2.7.

        library(sp)
        library(spatial)

   The simulation may take a couple of minutes to run and report the graphs.

 

 

     What will the Simulated Geographically Induced Partisan Biases show?

The simulations will show that simple plurality rules will produce severely biased allocation of seats when there are three parties and systematic mean votes differences across two regions.

Some simple things that you can try:

• Try different vote share to see how votes translate to seats (make sure votes add up to 1).

                   

• Try changing the mean vote share (using for example the descriptive voting data reported in Table 1 of the paper.