Sometimes it is sensible or even necessary to interlink certain data from different datasets in order to increase the knowledge gain. When the data already exists in separate datasets, this is referred to as output harmonization or ex-post harmonization, which enables the combination of data from different dataset sources. If the aim is to ensure data compatibility at the time of data collection stage, a pre-established categorization system serves as a tool for harmonized data collection in various national and/or cultural contexts. In this case, it involves input harmonization or ex-ante harmonization.

The following video clip (Chapter 00:22:22) provides an illustrative introduction to the topic of data harmonization:

The possibilities for harmonizing research data are complex and are subject to certain conditions. Nevertheless, harmonization is worthwhile, especially in order to close gaps in (one’s own) datasets, to increase the sample size in the data foundation, and/or to improve the robustness and reproducibility of research results.

The methods of data harmonization are promising tools for combining data from different datasets, however, they also represent complex procedures. In the section Tips & Checklists, you will find several helpful contributions and articles, which offer support for getting started with data harmonization.