Visualizing Open Access

The central question this summer school focused on was what if the results from publicly funded research were freely available to everyone?

The context is described as follows: “Open Access represents a vision of unrestricted access and use of scholarly publications and further research materials freely available worldwide without any legal, technical or financial barriers. Germany supports this idea at both the federal and state levels. Numerous research and higher education institutions have already issued open access policies and guidelines.” The objective was to visualize open access in Germany using the Open Access Bundesländer-Atlas, mainly to raise awareness of the changes and opportunities in academic publishing.

Partners

The summer school is a cooperation between the Urban Complexity Lab (UCLAB) at the University of Applied Sciences Potsdam, the Humboldt University Berlin, the Free University of Berlin and the Open Access Bureau Berlin acting on behalf of the Federal Ministry of Education and Research funded project open-access.network.

The Data

The pre-structured datasets (.excel / .csv) were provided on Google Drive along with a document in which the individual columns were specified in detail. This provided a great gateway into the exploative data analysis (EDA). With a legend for individual data points, we were able to find our way around the dataset more quickly and delve into the visual data analysis in a more comprehensible way.

Each team conducted the EDA on its own and shared the insights with all participants daily. This way, the different insights were accessible to all and at the same time we got an impression of how differently the same dataset could be mapped or visualized.

In terms of content, the multidimensional dataset consisted of educational institutions, so-called open access (OA) indicators, and additional OA information points.

The Challenge

The dataset we were provided contained the current (2021) research status on the open access situation at German educational institutions. A closer look showed that while information about the OA implementation at universities was widely researched, still little was known about the situation at Universites of Applied Scienes and the Arts (Hochschulen).

Thus we set out to answer the question of what is the OA situation at Hochschulen in Germany? Datavisualization-wise we aimed to create an interface that would be easily accessible and joyfull to use.

Team

Our small interdisciplinary team consisted of three people - Jonas Mirbeth (German Studies), our copy writer, Nick Haupka (Computer Science) our web developer and me (Interfacedesign), who focused on the concept and the visual language of the datavisualization.

Ideation

Content

In the ideation phase we gladly used the usual working techniques like brainstorming and clustering on Miro. In addition, we were guided by very useful slides from the Human-Centered Computing Research Group (FU Berlin), which gave us guiding questions such as: Who is the target group? What is the data abstraction? How does the interaction work? etc.

Visual

Visually, as design research, we looked extensively at the datavizproject website to get a feel for possible data visualizations. Then I sketched digital low fidelity wireframes to explore what kind of information architectures would be conceivable for our project.

Concept

The visual storytelling was inspired by the filtering interaction. The association of filter coffee came to mind and over time it turned out to be a viable metaphor – naturally, because at educational institutions coffee is drunk often and a lot when discussing ideas. Playing with this idea, we found out that the infotainment approach was perceived as low-threshold and joyful to use.

Data Visualization

The interactive datavisualization is meant for filtering multiple and different Open Access (OA) criteria. Each criteria is an input button on the left side of the layout.

The results are visualized on a Germany base map which is zoned into its 16 federal states. The figures on the map show the number of the implementing institutions in the respective state. The fill color gets darker, the higher the number of institutions is.

At the right side of the map the percentage of the Germany-wide implementation of the criteria is shown. Furthermore we decided to address the unknown data points explicitly. An unknown data point being a not yet researched one. This way the visualization communicates how much more research work still needs to be done and an additional insight is generated.

In the Roast by Hochschule section we mapped each coffee bean to one of the 190 contained institutions. The color denotes the implementation status of the chosen OA criteria or the combination of criteria.

Here a great many thanks to Prof. Marian Dörk for the always very helpful introduction to the basics of data visualization. The slides helped us a lot in our process.

Visual language

We elaborated the idea of filter coffee by choosing colors, fonts and icons associated with the world of coffee. The “Nobel” font, which we use in headlines for instance is adapted from the lavazza logo.

Prototype

Here you can check out our final website prototype for yourself.

Implementation

For the final implementation we had to process the initial data set several times. The 190 institutions were generated by filtering, adjusting and enriching several data points. Furthermore we converted yes/no statements into booleans. You can find the project code as well as the data set on the GitHub repository. Please note, that since we worked in a project based on a division of labor, I myself was not that involved in the final web implementation.

Conclusion

The two week summer school was a very enjoyable short but still deep dive into a completely new topic. I enjoyed the lectures and the challenge very much. The data visualization crowd is a truly great community and I am glad and honored to be part of it.