Tool streamlines data from various consumer surveys

Robbert works as a researcher at Wageningen Social & Economic Research (WSER) at the interface between science and IT. Translating the needs of scientists into technological solutions, that's where his heart lies. To find out exactly what scientists need, he works closely with them. The data science & innovation team of which Robbert is a part was tasked with coming up with a solution to a problem that is essentially a standardisation issue: "One of our research groups focuses on food and consumers. These studies often deal with similar subjects. But because surveys are set up differently and focus on different scales, data from those surveys cannot simply be linked."

To solve the problem of the different types of questionnaires, Robbert and his colleagues set up a 'library' of standard questions. With these questions, which are in line with the protocols of COMFOCUS, researchers can now build their questionnaires. This has developed into a tool with which they can also design their questionnaire. This questionnaire can be put into the Qualtrics survey tool via an import-export function, with which the researcher deploys the questionnaire and collects data. The data is then stored in a secure database in an automated structured manner, without the researcher having to have technical skills for this.

Link with data warehouse

The core of this WCDS project is about the link that allows research data to be included in Adagio, WSER's data warehouse, at the touch of a button. The process consists of two steps. First, the questionnaire is automatically processed into a data model and the database is generated by software. The researcher can then load the collected data into his own piece of database. Because the researcher can build the questionnaire with questions from the data library, this saves the researcher a lot of work, according to Robbert: "Frequently used questions are ready for you, which makes it easy to 'click together' a questionnaire. Subsequently, all collected data is automatically FAIR: findable, accessible, interoperable and reusable. The data can be accessed via a standardized interface. Of course, the condition is that researchers set up appropriate access authorisations so that the data is 'as open as possible, but as closed as necessary'."

Close to the domain

According to Robbert, by harmonising, researchers can gain much deeper insights from existing datasets. It is important that researchers know how the tool works. Robbert: "Projects like this are often approached from a technical angle. With the risk that a solution is found that users do not understand or that does not fit well with what they need. Our advantage is that we are very close to the domain of food and consumer. The cooperation with the EU programme COMFOCUS also helped. The aim of this programme is to harmonise research methods in this domain. There was a great need to be able to reason beyond individual datasets. Based on that need, we designed an infrastructure that is partly based on existing tooling, such as EQT, Adagio, and iRODS. This allowed us to offer a working system relatively quickly."

Future

In the meantime, Robbert and his colleagues are working on an automatic link that will allow researchers to no longer even have to export and import data files themselves. And to make the data flow even easier, researchers will be able to generate a DOI for their dataset themselves in the future. "This allows the researcher to comply with the FAIR principles even better"

If it were up to him, many more research domains would use this infrastructure in the future. "The current set-up is suitable for any researcher who uses questionnaires. That is why we want to make the infrastructure suitable for use by other research groups. Regardless of which technology stack they work with. Our current implementation runs on a Microsoft solution, but we are currently in talks to create a platform-independent version."