How do I cite the dataset?
If you use variables from Paths to Power dataset, you should cite: Nyrup, J., Knutsen, C. H., Langsæther, P. E., & Kristiansen, I. L. (2025). Paths to power: A new dataset on the social profile of governments. Forthcoming in the British Journal of Political Science.
If you are also using information from the WhoGov dataset, then please also cite: Nyrup, J., & Bramwell, S. (2020). Who governs? A new global dataset on members of cabinets. American Political Science Review, 114(4), 1366-1374.
How was the data collected?
The data was collected through a systematic effort involving more than 30 research assistants and external researchers who conducted internet searches for information on cabinet ministers in the period from 1966 to 2021. The sources included Wikipedia, LinkedIn, biographies, obituaries, and news articles.
Which month is the data from?
The data contains data on cabinet members holding office in every July 1966 to 2021, with specific adjustments for 1966 and 1970, where data was only available for September and January, respectively.
Can you combine the dataset with other datasets?
Yes, the PtP dataset can be combined with other datasets. It has variables that can be used to create aggregated country-year datasets. This provides more extensive analytical possibilities by linking PtP data with global (e.g., World Values Study), regional (e.g., Afrobarometer), or national survey data. The individual dataset can also easily be combined with the World Higher Education Database (WHED).
I found an error in the dataset. Who do I contact?
You can contact us at paths-to-power@stv.uio.no if you have any suggestions for corrections.
What is the relationship between WhoGov and Paths to Power?
Paths to Power utilizes an updated version of the WhoGov dataset to get the name of the ministers. WhoGov provides basic information on cabinet members, which serves as the foundation for identifying coding units in PtP. PtP supplements this basic information by including detailed educational, occupational, and social background information.
What are possible sources of error and bias?
Potential sources of errors are discussed in detail in the "Limitations and considerations" section of the article. Examples of possible sources include:
- Incomplete or unclear sources leading to difficulties in coding, particularly for class background data.
- Challenges in finding consistent information across various ministers, especially in less democratic contexts or poorer countries.
What is the coverage of Paths to Power?
PtP covers 141 countries, globally, with populations exceeding 10 million; all OECD and EU countries; and several other smaller countries. We collect information---whenever identifiable---on 44,789 cabinet members holding office across 1966-2021, regarding their level, type, and location of education, place of birth, family- and class origins, prior occupation, and political background
What type of information does the dataset include?
PtP contains detailed individual-level data on cabinet ministers including:
- Educational background (e.g., highest degree obtained, type of degree, education abroad)
- Social background (e.g., place of birth, class background, political family associations)
- Occupational background (e.g., main occupation before entering politics and prior political experience)
How does the dataset handle ministers with multiple degrees or occupations?
For ministers with multiple degrees of the same level, PtP codes only the last degree obtained prior to their emergence as cabinet members. Previous occupations are limited to the main occupation they held in the decade before entering politics, ensuring that only the most relevant experience is recorded.
What measures were taken to ensure coding consistency across countries and coders?
To ensure coding consistency, a detailed codebook was provided to coders, along with training sessions that emphasized consistent interpretations of data. Regular meetings facilitated discussions on difficult coding decisions, and internal communication among coders was maintained through platforms like Slack. External coders were paired with a research assistant on the project to maintain consistent coding.
Which variables have the most missing data, and why?
Variables such as social class background and some of the educational background variables (such as university name) often have higher rates of missing data due to the difficulty in obtaining comprehensive information, especially in poorer or less democratic countries.
Can the dataset be aggregated for country-level analysis?
Yes, the individual-level dataset can be aggregated to develop country-year datasets, allowing analysis at the country level, such as measuring the share of ministers with higher education or other derived measures interest.