More than the sum of its parts: the amplified potential of educational data

Data is the basis for knowledge in that it contains potential information. A comprehensive and solid basis of data is therefore of fundamental importance for research on education and learning in Switzerland and for monitoring the Swiss education system. In order to obtain such a database, various aspects must be taken into account. Data must be harmonised in order to measure comparably; data must be documented according to a targeted standard to enable exchange. In addition, access to data must be comprehensible and as open as possible, and ultimately, it is important to be able to relate the data sets to one another. This in turn requires that permission to linkage has been given, that a meaningful assignment (e.g. through identifiers) is possible and that linking data sets makes sense for research from the perspective of quality and content. In this sense, this project aimed to facilitate the use of data for educational research and educational monitoring in Switzerland, thereby laying the basis for more comprehensive knowledge.

In Switzerland, there are already important organisations and significant studies for the collection and provision of data on the education sector. These include FORS, the Federal Statistical Office (FSO), the TREE study, the ÜGK study, universities and cantonal offices, to name just a few. At the same time, the data sets often only describe sections of the education system, which leads to fragmentation that in turn does not allow for a holistic picture. Access to the data is often not described in a useful manner, and documentation is incomplete and not standardised. This makes research and monitoring more difficult and thus leads to inefficient structures. As a further aspect, digital transformation can be expected to create even more data sources that need to be tapped into and made usable.

The project was launched with the aim of linking data sets and making data products on the Swiss educational area available on a sustainable basis. It quickly became clear that this was not feasible for an infrastructure project. Data may only be used and linked for research projects as far as possible. Even the FSO has to separate linked data sets for research projects after the research has been carried out and is not allowed to keep them for the long term. For the University of Applied Sciences of the Grisons, which has teaching and research but not data storage and is therefore not a research data centre, the physical linking and storage of data from different sources was therefore not possible. However, it also shows that it is of great importance for research and monitoring to identify the fundamental hurdles to a harmonised data landscape and to find ways to overcome them. This project subsequently focused on this approach.

Over the course of this project, significant additional initiatives were advanced and launched. The Open Data Initiative developed from a pure concept of ‘open’ to an understanding of ‘open’ as being accessible according to rules. This development was clearly demonstrated in the two workshops organised by this project and FORS. As a result, the guideline ‘as open as possible’ and access to sensitive data, as educational data often is, were included in the Open Data Initiative portfolio. New initiatives in the course of the project were the adaptation of the regulation on the Statistics Act, the discussion of the law on the secondary use of data and the discussions on data spaces. All of these initiatives could be addressed and criticised through this project. The outputs of the project provide important, well-founded descriptions of the situation in Switzerland for research and practice, as well as best practice and recommendations for action that can be adopted now or in the future.

Three main messages

1. Data as a source of knowledge must be protected and at the same time made available for analysis. The focus of the measures should be on data usage. In order to do this in a secure way, infrastructures for secure access and usage must be in place.
2. Even though the digital transformation produces more data, a good data landscape for research and monitoring of the educational sector must be maintained. It is evident that the data that is a by-product of digital transformation is not intended for research purposes. Procedures, protocols and control mechanisms need to be put in place to use and combine data from surveys and from the digital transformation.
3. Data flows in the field of education and learning need to be more visible. Only then trust can be built, the amount of data collected by surveys minimised, and a useful network of data as basis for knowledge established. This will benefit the educational system and everybody participating in it.

Virtual Educational Observatory