Replication Data for: "Office Insecurity and Electoral Manipulation" (doi:10.7910/DVN/VGI93T)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Replication Data for: "Office Insecurity and Electoral Manipulation"

doi:10.7910/DVN/VGI93T

Harvard Dataverse

2015-07-20

1

Rozenas, Arturas, 2015, "Replication Data for: "Office Insecurity and Electoral Manipulation"", https://doi.org/10.7910/DVN/VGI93T, Harvard Dataverse, V1

Replication Data for: "Office Insecurity and Electoral Manipulation"

doi:10.7910/DVN/VGI93T

Rozenas, Arturas (New York University)

Harvard Dataverse

Rozenas, Arturas

Rozenas, Arturas

2015-06-21

https://doi.org/10.7910/DVN/VGI93T

Social Sciences

Data and replication files for "Office Insecurity and Electoral Manipulation" paper.

<a href="http://creativecommons.org/publicdomain/zero/1.0">CC0 1.0</a>

Label:

analysis.R

Text:

Replicates results in "Data and Measurements" section and tables in "Results" section as well as the table in Online Appendix 2.

text/plain; charset=US-ASCII

Label:

data.csv

Text:

Data file from which replications of the analyses can be conducted.

text/plain; charset=US-ASCII

Label:

Data.zip

Text:

The archive contains source dataset and four code files to pre-process and merge different datasets contained in sourcedata.zip file. The output of these files is the master data file data.csv, which is made available as a separate file. The easiest way to replicate pre-processing and merging is by extracting archives Data.zip and sourcedata.zip and running "Level 1" replication of the master replication file "replication.R."

application/zip

Label:

helper_functions.R

Text:

Contains helper functions. This file must be sourced before proceeding with all subsequent replications.

text/plain; charset=US-ASCII

Label:

IVanalysis.R

Text:

"Instrumental variables analysis" section of the paper and the related Online Appendix 5.

text/plain; charset=US-ASCII

Label:

matchout.Rdata

Text:

Contains results of the genetic matching procedure (for each of the three treatments). To replicate Figure 5 (matching), load this data and run the second section of the matching.R file. The contents of this file can be replicated (up to stochastic variation) running file matching.R.

application/x-rlang-transport

Label:

mediation.R

Text:

Mediation analysis in Online Appendix 3.

text/plain; charset=US-ASCII

Label:

nonlinear.R

Text:

Monotonicity of the effects in Online Appendix 4.

text/plain; charset=US-ASCII

Label:

replication.R

Text:

This is master replication file that sources other replication files. Replication can be conducted at two levels: 1. Level 1: conditional on data “data.csv”, one can replicate all of the analyses reported in the paper and Online Appendix, but this does not include data pre-processing: estimation of latent manipulation scores, computation of Coup Threat parameter “lambda”. 2. Level 2: replicates the creation of the master dataset “data.csv” from the source data files. To do that, one first needs to extract the folder “Data.zip”, and then extract the archive “sourcedata.zip.” See the comments inside this file for further information.

text/plain; charset=US-ASCII

Label:

robustness.R

Text:

Replicates results of "Alternative measures and explanations."

text/plain; charset=US-ASCII

Label:

selection.R

Text:

Replicates section "Endogeneous election-timing" and Online Appendix 6.2 related to this section. Note that to run this file, one needs to source the contents of Data.zip folder.

text/plain; charset=US-ASCII