######
###
### Information to replicate the analysis for "Paths to Power:  A new dataset on the social profile of governments"
### British Journal of Political Science
###
### Jacob Nyrup, Carl Henrik Knutsen, Peter Egge Langsæther, and Ina Lyftingsmo Kristiansen
###
######

We are using both R and Stata to conduct the analyses in the paper. 11 files are needed to replicate the analysis of the paper and the appendixes. These are two scripts in R, two scripts in Stata, two data files containing Paths to Power (PtP) in RDS-format, and 5 additional files with additional variables used in some of the analyses. In addition, we have uploaded both the individual-level dataset and the country-year level dataset in dta and xlsx-format. Below we briefly go through each script.

The individual-level dataset is found in "PathstoPower_individuallevel_v1.0.rds", while the country-year level dataset is found in "PathstoPower_countryyear_v1.0.rds" (you can also find them in csv and xlsx format).

The file "1-pathstopower_mainanalysis" creates all of the main analyses included in the paper, apart from Figure 4 which is created using "3-figure4.do". In addition to the data from PtP we also use "df_alex.rds", "df_lead", and "df_tmd.rds" which contains averages from both other datasets and PtP and are used to create Figure 1.

The file "2-pathstopower-appendix" creates all analyses in the appendices apart from Appendix G. The analysis in Appendix G is created in "4-appendixG.do". We use "backgroundvariables.rds", which contains information on GDP per capita and other background variables, in some of the appendices. The file "bothsamples.xlsx" contains the 400 ministers that were coded twice to test an intercoder reliability. Every variable has the subscript 1 and 2 indicating whether it was the first or second coder that coded the data.