Machine-readable libraries

Election commissions publishing their data pay next to nothing attention to the ease of analysing large data sets. The most common solution: hierarchical HTML pages. In such cases, pages need to be preprocessed via data scraping.

Countries with higher levels of falsification also stand out in terms of access to data. For example, the Russian CEC tries to close data sets from machine processing by using captchas, while the Belarusian CEC does not publish them at all, limiting itself to aggregate results by region.

But for the most part, organising commissions strive for open data. The data are open but not ready for analysis. In such cases, researchers can either make the data of the commissions machine-readable by themselves or use data made publicly available by other researchers and specialists. In such cases, it is necessary to check the correctness and completeness of the data (verification with official data).

Some commissions publish data in study-friendly formats. For example, the Armenian CEC publishes data in the form of Excel spreadsheets.

The following are several sources of machine-readable election data.