PRESS RELEASE
Germany needs a research data infrastructure for the 21st century
- The German research data infrastructure lags behind that of its international peers. Important data are missing, restrictions frequently prevent data linking, and data access is user-unfriendly.
- Statistical legislation should be output-oriented, i.e., it should specify which informational needs official statistics should fulfil. However, it should allow freedom in the actual data collection in order to enable faster reactions to new developments.
- Research is fundamental for informed policy decisions. For this reason, a Research Data Act should be passed quickly, assigning research a high weight against other competing interests, including data protections. This is a prerequisite for independent research to be able to evaluate much of the collected data.
The data infrastructure in Germany lags behind that of its international peers. There are considerable deficits in the scope and timely availability of data as well as in the possibilities of data linkage. This makes empirical analyses, which are an essential basis for evidence-based policy advice and decisions in politics and public administration, difficult. Due to the poor data infrastructure, it is difficult to assess the consequences of planned measures or to evaluate the effectiveness and efficiency of measures already implemented.
“The Covid pandemic and the energy crisis have shown that a reliable and timely available data basis is indispensable for effective measures and their evaluation. Every improvement in German data infrastructure is an investment in better-functioning, more targeted and more cost-effective policy decisions,” emphasises Monika Schnitzer, chair of the council.
A multi-pronged strategy is needed to make the German research data infrastructure fit for the 21st century. This includes a realignment of legislation relevant for the official statistics. Instead of stipulating the exact data collection process, the law should define what information and services official statistics must provide. The statistical offices and other relevant authorities would then design the specifics of the data collection to meet those needs. Such output-oriented statistical legislation makes collection and provision of the required information much quicker, especially in times of crisis. In addition, the Federal Statistical Office should be given a research mandate and tasked with providing research data. The speed of response and the quality of official statistics should further be improved through an increase in human and financial resources.
To close existing data gaps, additional data collection is needed, for example in the areas of education and wealth, and a high-frequency household survey should be established. The greatest potential for improving the data offered by official statistics lies in a much more extensive use of administrative process data. These are available at low cost, in a more timely manner and without additional burdens for companies and households. Their other advantage is higher accuracy in comparison to the survey data.
A research data law that grants researchers extensive rights to use and link data should be passed quickly. It should assign research a high weight against other competing interests, including data protection. This is a prerequisite for independent research to be able to evaluate much of the collected data.. To improve the user-friendliness of the data usage, remote access should become the default access mode offered for research purposes. The public data offering of the Statistical Network should also be brought up to the latest technical standards.
