Frequently Asked Questions

Missing from merge, and missing from mismatch. More detailed information can be found in the methodology document.

• You can choose the option “Remove empty rows”. This option will remove all rows with only missing data from your customized dataset, but still indicate whether a missing observation is the result of the merge or missing in the original data.
• You can set the missing codes to true missing values, e.g. like this if you use R: df[df == -11111] <- NA

V-Dem experts give their ratings on an ordinal scale; the output from applying our measurement model to these ratings is on an interval scale. Hence, the ratings as such take on discrete values but the latent scores can be of any numeric value (often between -4 and 4).

The Demscore infrastructure only includes those variables from the QoG OECD and Standard Cross Sectional datasets that are NOT also included in the time series datasets. This is to avoid redundancy of the data. You can download all variables from the time series dataset in the QoG Country Output Unit in Demscore if you want to conduct cross-sectional analysis.

To choose the right Output Unit, you first need to decide in which format you want to retrieve the data. Demscore offers several formats which includes, but are not limited to, the following: Country-Year, Cabinet, Country/Regional, Conflict, and many others. If there are several output units available for that format, you need to decide which set of identifiers you want to use. This is important since different Modules use different country definitions and names, such as ISO, Gleditsch and Ward, or V-Dem country units. We account for that when merging and translating the data but please be aware that countries are not always defined similarly across different Modules. You can find a descriptions of all Output Units on our Data Download Page.

If you want to learn more about how Output Units are created, please consult the Demscore Methodology document.

The downloader ID makes it possible for others to replicate your download. If you share your downloader ID for instance with your colleagues or supervisors, they can retrieve the exact same dataset and reference documents as you did if they choose the option Download data by Downloader ID in the Demscore download interface.

You can find the downloader ID in your autogenerated customized codebook which is included in the zip file retrieved from the download interface.

Yes. The downloader ID works across versions, i.e., even if the version on the website in the future is e.g., Demscore v5, you will still be able to retrieve data from v1 if you have a downloader id.

Demscore does not offer the option to download a full dataset, since the purpose of the infrastructure is to merge and harmonize data from different sources. If a user is interested in a dataset in its original form, they can find the link to each project’s website (where the original dataset is available) on the Partner pages.

However, the user can recreate the original dataset through Demscore by selecting all variables or codebook sections from a dataset in their original Output Unit, and by removing both the unit columns and empty rows from the download.

The Demscore project does not have a formal citation of its own. Hence, when using Demscore, we recommend citing each respective project and dataset. We indicate how every dataset is to be cited in the autogenerated codebook you receive with your data download, both in the dataset description and the codebook entry for each variable. Most often you only need to cite the dataset a variable originates from, but sometimes there is a variable-specific citation listed in the codebook entry in addition to that. For these cases, please also add the variable-specific citation to the reference list of your publication. Full references are linked in the codebook entries of the variables and listed in the codebook’s bibliography.

Please also cite the Demscore Methodology when you use data retrieved through Demscore. We suggest the following citation: Gastaldi, Lisa, Melina Liethmann, Johannes von Römer, with Alma Eckhoff Owing, Samuel Sjögren, Steven Wilson, and Staffan I. Lindberg. 2024. “Demscore Methodology v4”. Gothenburg, Sweden: Demscore National Research Infrastructure.