Coding lab 2024: Statistics Explained through literate programming 2024
Dissemination is an integral part of the whole statistical production workflow. Inspired by the reproducibility movement in Open Science and driven by the opening and sharing of all assets – making not only the data, but also the methods, tools, and software open – the EMOS Coding lab aims to promote an innovative bottom-up approach to dissemination that enables collaboration and increases participative and reproducible forms of statistical services’ design and sharing. Reproducibility along the different stages of the dissemination process supports the objectives of the European data strategy, that expresses a commitment to transparency, collaboration, and accessibility in the digital era, as well as several principles expressed in the European Statistics Code of Practice, most notably principle no. 15 on accessibility and clarity.
This project aims at replicating Eurostat Statistics Explained articles, statistical products composed of tables, charts and text. The ultimate goal is streamlining the production process for retrieving the data (through calling the Eurostat Application Programming Interface - API), making calculations, producing tables and visualisations, and more ambitiously integrating text in computational notebooks (e.g. with R Markdown). In practice, the Statistics Explained pages published on Eurostat websites will be reimplemented in R, following the layout guidelines, which need to be replicated in the programming language. First examples (and templates) of what the project aims for can be explored on the statistics coded page of Eurostat GitHub domain.
Six students from EMOS-labelled master´s programmes worked from March to June on selected Statistics Explained articles covering the domains of culture statistics, sports statistics, and quality of life indicators. During the closing event, that took place at the end of June, the students presented the process of recreating the articles, the challenges they faced, and the solutions they implemented. The participants gained a deeper understanding of Eurostat´s data and metadata, dissemination products, and the importance of standardisation and automation.
