How can analytics data help us understand how digital tools are being used today? Beyond the pure business aspect, digital analytics also provides material for university research. At the University of Upper Alsace (Université de Haute-Alsace) in France, digital data is used for projects that unite hard science and social science. We asked Carsten Wilhelm, researcher at the CRESAT lab (Centre for Research on Economy, Society, Art and Technique), and associate professor in the University of Upper Alsace’s Information-Communication department, to tell us more.
What is digital analytics’ role at the University of Upper Alsace?
The University of Upper Alsace (based in two cities, Mulhouse and Colmar) is a multidisciplinary university with a major scientific and technical centre. The university also offers courses in innovative communications and digital culture. Our interest in digital methods at the CRESAT lab and in the Information-Communication department is twofold: Firstly, we decided to integrate these skills into the curriculum for students headed toward digital professions, in our master’s degree in communications and digital publishing, for example. It’s important for our future web project managers, future communications managers, and future agency consultants to know and understand the analyst’s profession. After earning their master’s degree, some of our alumni have continued in this sector to work for international companies. Professionals from within the digital industry play an important role in our educational offerings, and we work with several companies including AT Internet for our courses on analytics, e-reputation management, learning analytics, community management, and digital’s role within an organisation. To do this, we’ve developed numerous partnerships so that we can offer our students internships and apprenticeships in the professional world, in companies with digital culture.
Secondly, for several years, we’ve invested in research methods drawing from digital. Within this framework, we are particularly interested in methods for analysing web data for social science research, in order to really understand why this data matters for digital humanities.
Do you use digital analytics for research purposes?
Yes. It was completely natural for us at the University of Upper Alsace to initiate a collaboration between hard sciences and social sciences, in order to better understand the approach of each. With this goal in mind, an initial internal project was financed in 2013 to bring together data processing specialists and social science researchers.
When our researchers are working with material that’s omnipresent yet highly particular in how it’s processed – as is the case with web data and digital tracking – we want to be able to push the tools available to them even further. Our colleague Dominique Boullier talks about this – a new, third generation of social sciences research that follows on the heels of the first generation of research (at the start of the last century) based on government statistical data, and a second generation of research (post-war to today) based on empirical research methods almost like opinion polling. Today’s digital tracking, digital noise, and Big Data have changed the game; these are interesting yet complex-to-use elements that beg to be better understood by social sciences, and integrated into their processes. Unlike Chris Anderson, who suggested a few years back that the data deluge would make the scientific method obsolete, as massive amounts of data could provide answers without needing to be interpreted, we think that on the contrary, it’s very important for social sciences to understand how these data collection, aggregation and processing tools work, so that we can use them to our advantage. Even still, a solid approach in social sciences is still necessary in order to organise and interpret this type of data, as shown with the Algopol project led by an interdisciplinary team including Dominique Cardon. We therefore systematically cross-reference web data with other research methods and, when possible, combine digital data and qualitative research.
“We want to know how data analysis can enrich researchers’ understanding of human and social sciences.” – Carsten Wilhelm
What methods do you use to explore your data?
Alongside qualitative methods (observations, interviews, focus groups) and traditional quantitative methods (research by questionnaire), we also want to know how data analysis can enrich researchers’ understanding of human and social sciences.
Ideally, we cross-reference quantitative data with more traditional information and communication science methods. We even study the frameworks made available by tools like AT Internet’s Analytics Suite; our colleague Dominique Cardon speaks about why it’s particularly interesting for social science researchers to examine the approach, categories and methods used within these tools. These tools were designed for a very specific objective, and they reflect a thought process that’s particular to the professions of these tool users (such as analysts) – and they do this well. Using these tools requires us to question their architecture, and to question the resulting analytical conclusions. It’s very enriching.
Can you share some examples of projects you’ve implemented?
The first project beyond our walls that will benefit from AT Internet® tools will be a study of how the Forum des images’ online resources are being used, via tracking of web activity and more traditional sociological audience research. This project aims to better understand the behaviours of the site’s visitors and of the forum’s audience (which are not necessarily the same group of people) vis-à-vis the resources available for music-lovers (master classes, exhibit access, festivals…). Audience diversity cannot be taken into account using only audience measurement, digital analytics or questionnaire research. This type of research context and question requires a combined methodology.
This methodology will be particularly useful for our CUMEN (Culture of Digital Media) project, an Idex/Novatris project, which studies how media is used by the younger generation, from high school to university and beyond, with the particularity of working on international, multilingual grounds (France, Germany, Switzerland, United States), in collaboration with our international university partners. This project is already combining several methodologies (use of personal journal, interviews, focus groups, questionnaires). And we’re currently thinking about digital analytics’ role in this.
Do the results of projects like these influence your methods for teaching communication sciences?
Digital usage in itself has already strongly influenced our teaching practices. In our communication and multimedia degree programme, we’ve integrated work projects to propel students’ digital skills starting in their third year. Our master’s degree programme in communications and digital publishing integrates analytics into the curriculum; with each research project, we strive to put our students in the driver’s seat, in the role of researcher or analyst, up until the final report they must produce for our partners. In producing this report, our students are therefore confronted with two sides of applied research: answering the most typical questions about how digital tools are being used, and answering questions from researchers who interpret this same data. It enables students to take a managerial position and to adjust the approach where needed. We are particularly mindful about ensuring our students are proficient in this, via our digital literacy projects in the framework of the CUMEN Project, for example, and via the theses prepared by our students.
Many thanks to Carsten Wilhelm and the team of instructors at the University of Upper Alsace’s Information-Communication department for their contributions to this article.