Key question New
AI-processed · Author-approved
What design requirements guide the design of open research data platforms to support journalistic data reuse?
Loads Spotify's player; Spotify may set cookies. This device remembers your choice.
Key findings
- Surfaced five core barriers to research data reuse: fragmented data access, technical format inconsistencies, insufficient methodological documentation, scarce verification support, and temporal update delays.
- Established two guiding meta-requirements: providing shared data structures for reliable technical exchange and enabling contextual translation for shared understanding.
- Defined ten platform design requirements covering both technical mechanisms (machine-readable formats, cross-source search, and newsroom tool integrations) and interpretive mechanisms (direct researcher-journalist clarification channels, narrative fields explaining assumptions and uncertainty, and example visualizations).
- Highlighted the importance of designing open-data platforms around the entire data reuse process, rather than treating data publication and access as the endpoint.
What it means for you
CIO / IT Executive
Allocate budget to integrate contextual translation features and direct communication channels into the institutional open-data platform for non-academic stakeholders.
IT Manager
Configure the open-data repository API to output datasets in machine-readable formats accompanied by standardized methodological documentation fields.
Business Strategist
Conduct a stakeholder mapping exercise to identify external non-academic user groups for a targeted open-data platform value-proposition pilot.
Researcher
Draft a plain-language narrative field explaining core dataset assumptions and uncertainties to publish alongside the raw research files.
Policymaker
Revise open science grant compliance criteria to mandate interpretive usability standards, such as example visualizations and methodology documentation, alongside raw data publication.
Summary
This study examines how open research data platforms can be designed to support data journalists in reusing complex scientific datasets. Following a design science research approach informed by boundary theory - which examines how knowledge is transferred and understood across communities with different expertise - the authors gathered requirements through semi-structured interviews with six data journalists, a co-creation workshop with 15 data journalists, and a formative evaluation with 13 information systems and research communication experts.
Problem
As open science policies increasingly encourage or require researchers to make their data publicly accessible, ensuring meaningful reuse remains a challenge. Open-data platforms often emphasize data publication and technical accessibility, while providing limited support for interpreting research data outside its original scientific context. This creates particular challenges for data journalists, who must understand and assess unfamiliar datasets before using them in their reporting.
Transcript
Expert: Hello, Anna. Across universities and research institutions, open science policies increasingly require data to be made publicly accessible. However, making data available online does not mean it is usable. Most current repositories focus almost exclusively on data publication and technical storage. They rarely provide the context needed for professionals outside academia—such as journalists—to interpret what the numbers actually signify.
Host: And journalists work under tight deadlines, so navigating raw, uncontextualized files creates friction. How did the authors investigate this problem?
Expert: The study was conducted by Christina Speck, Judith Erbacher, Thimo Schulz, and Christof Weinhardt, published in the proceedings of the International Conference on Information Systems in 2026. They applied a design science research approach informed by boundary theory, which studies how knowledge moves across groups with different expertise. To gather requirements, they conducted interviews with six data journalists, held a co-creation workshop with 15 data journalists, and performed a formative evaluation with 13 information systems and research communication experts.
Host: What were the primary obstacles the journalists reported when trying to reuse research data?
Expert: The study identified five core barriers: fragmented data access across isolated portals, technical format inconsistencies, insufficient methodological documentation, scarce verification support, and temporal update delays. In short, journalists struggle to locate datasets, convert them into usable formats, verify findings with researchers, and understand the assumptions or limitations behind the data.
Host: To address those barriers, the researchers developed design requirements based on boundary theory. How do those work in practice?
Expert: Boundary theory differentiates between syntactic boundaries—which are technical—and semantic boundaries, which involve meaning and interpretation. Correspondingly, the study established two meta-requirements. The first is providing shared data structures for reliable technical exchange. This includes machine-readable formats, unified cross-source search, and integration with standard newsroom analysis tools.
Host: And the second meta-requirement addresses the interpretive side?
Expert: Exactly. It focuses on contextual translation for shared understanding. The study proposes specific mechanisms like narrative fields where researchers explain their core assumptions and the uncertainty in the data. It also suggests embedding example visualizations and direct communication channels so reporters can contact the researchers to verify conclusions before publishing.
Host: Looking at practical implications, what does this study mean for organizations outside of academic research?
Expert: The lessons extend directly to corporate data management and public-sector transparency initiatives. Many enterprises build internal data lakes or open data portals assuming that access alone drives adoption. This study demonstrates that data reuse fails without contextual translation. For business leaders, treating data as a product requires designing around the entire consumption lifecycle. That means pairing technical pipelines with narrative documentation, explicit error margins, and clear channels for domain-expert consultation.
Host: So the core takeaway is shifting the definition of success: data publication is not the end of the project, but the beginning of a translation process between technical producers and practical users.
Expert: Precisely. Without structured interpretation, open data remains an underutilized asset.
Host: Alex, thank you for breaking down this study. And to our listeners, thank you for joining us on A.I.S. Insights, powered by Living Knowledge.
data journalism, open data platforms, data reuse, boundary spanning, design science research, open science