Research

The technical development of social media, video surveillance cameras, and body cameras on the one hand, and the relevance of artificial intelligence on the other, are currently increasing the diversity and availability of video data, and this trend is set to continue in the future. Despite individual reflections on the empirical use of video data as research data from a Sociology of Scientific Knowledge perspective, there has been no targeted methodological comparison that critically examines the existing schism between qualitative and quantitative approaches. The planned scientific network will engage in dialogue with mixed and multimethod researchers to examine and compare the methodological and epistemological assumptions that lead to the production and use of video. Through joint exchange, the network will reflect on the claims to truth and validity as well as the quality criteria that are relevant for the collection, further processing, and analysis of the data. On the one hand, the network is based on the MMMR approach, i.e., Mixed Methods and Multi-Method Social Research. This is based on a broad understanding of method integration, which also includes the linking of methods within qualitative or quantitative approaches. On the other hand, the network draws upon perspectives from Science and Technology Studies and considers the production and use of video not in terms of a concept of science as science-as-made, but rather in terms of science-in-the-making, as first proposed in the laboratory studies of Bruno Latour and Steve Woolgar. The classification into quantitative versus qualitative is critically questioned in so far as the planned project asks how quantitative and qualitative data are generated and which methodological criteria are applied for this purpose. Finally, these perspectives are complemented by Critical Data Studies, because much video data is currently available as Big Video Data. The proposed network will explore in joint dialogue whether and to what extent (audio)visual data not only represent phenomena, but instead transform knowledge and thereby produce something new. This also includes the question of what constitutes the handling of research data from natural situations (in Garfinkel's sense) and how this may be changed by deepfakes of video data.

To achieve these goals, first, an annual workshop with invited national and international speakers will be held, followed by an international conference for mixed and multimethod researchers working in the field of video. Second, joint research projects will be initiated between network members, focusing either on meta-reflection or the application of integrated approaches with video. Third, the results of the network will contribute to optimizing the storage and reuse possibilities for video in existing repositories.

The social production of video data is closely tied to the context of its collection and processing. In a seminal study, Charles Goodwin and Marjorie Goodwin (1997) prominently demonstrated how an amateur video of police violence was contextualized and decontextualized differently in two court cases, leading to diametrically opposed judicial decisions. The workshop first addresses the question of which contextualizations are relevant for subsequent analysis. By what quality criteria is the relevance of contexts reliably determined? Conversely, under what conditions are contexts deemed irrelevant? The workshop focuses specifically on the methodological and epistemological premises used for the contextualization or decontextualization of video data and does not content itself with a simple reference to pragmatism, as is often the case in mixed methods (for a critique, see Gobo 2023). What exactly is designated as context in the various methodological perspectives? In ethnomethodology, for example, context is generated by the filmed participants referring to it. These can also be cross-situational contexts (see, e.g., Meyer 2017; Coenen/Tuma 2022). To what extent can multimethod approaches support the identification of relevant contexts?

Judith Schoonenboom (2023) has convincingly demonstrated that qualitative data are context-dependent, whereas quantitative data do not (or no longer) include this contextual reference. The workshop therefore addresses, secondly, the question of how the distinction between qualitative and quantitative data is established through contextualization or decontextualization. (Audio-)visual data differ from conventional research data in that visuality is particularly complex. Unlike language or text, visual perception is characterized in particular by synchronicity (see, e.g., Simmel 1993). (Audio-)visual research data comprise the totality of many synchronous pieces of information. The workshop aims not only to examine which contextual information is deemed relevant under which conditions, but also, in particular, how this generates qualitative or quantitative attribution. Schoonenboom has emphatically pointed out that a quantitative analysis can only be conducted with quantitative data, and conversely, a qualitative analysis only with qualitative data. The workshop explores the question of whether and to what extent the methodological schism has its origins in the contextualization or decontextualization of data. Against the backdrop of the complexity of (audio-)visual research data, the handling of the framing of research data is critically reflected upon, thereby contributing to a fundamental questioning of the binary distinction between qualitative and quantitative. 

References: 

Coenen, Ekkehard/Tuma, René (2022): Contextural and Contextual – Introducing a Heuristic of Third Parties in Sequences of Violence. In: Hoebel, Thomas/Reichertz, Jo/Tuma, René (ed.). Historical Social Research, 47 (1), Special Issue “Visibilities of Violence. Microscopic Studies of Violent Events and Beyond, 200–224.

Gobo, Giampietro (2023): Mixed Methods and Their Pragmatic Approach: Is There a Risk of Being Entangled in a Positivist Epistemology and Methodology? Limits, Pitfalls and Consequences of a Bricolage Methodology. FQS Forum: Qualitative Sozialforschung/Forum: Qualitative Social Research, 24(1), Art. 13, dx.doi.org/10.17169/fqs-24.1.4005 (download on 6.5.25).

Goodwin, Charles/Goodwin, Marjorie (1997): Contested Vision: The Discursive Constitution of Rodney King. In: Gunnarsson, Britt-Louise/Linell, Per/Nordberg, Bengt (ed.). The Construction of Professional Discourse. London, New York: Routledge, 292–316.

Meyer, Christian (2017): The Cultural Organization of Intercorporeality: Interaction, Emotion, and the Senses among the Wolof of Northwestern Senegal. In: Meyer, Christian/Streeck, Jürgen/Jordan, Scott (ed.). Intercorporeality. Emerging Socialities in Interaction, New York: Oxford Univ. Press, 143–172.

Schoonenboom, Judith (2023): The Fundamental Difference Between Qualitative and Quantitative Data in Mixed Methods Research. FQS Forum: Qualitative Sozialforschung/Forum: Qualitative Social Research, 24(1), Art. 11, dx.doi.org/10.17169/fqs-24.1.3986 (download on 7.5.25).

Simmel, Georg (1993): Soziologie der Sinne. In: Idem, Aufsätze und Abhandlungen 1901-1908, Band II, edited by Cavalli, Alessandro/Krech, Volkhard. Frankfurt/M.: Suhrkamp, 276–292.

In Science and Technology Studies, the term “embodiment” is used to describe the transcendence of traditional mind-body boundaries (see, e.g., Myers 2008; Mol 2002; Latour 2012). It demonstrates not only that states of consciousness manifest in the body – as in psychology or biology, for example – but especially how the body becomes an acting agent. While previously only human consciousness was attributed agency, in current discussions this applies equally to the body as well as to technology itself (for an overview, see e.g., Kissmann/van Loon 2019). Comparable debates in sociology draw on Merleau-Ponty’s phenomenology of the body, communicative constructivism following Berger and Luckmann, or practice theory, without, however, assuming a symmetry between the different entities and treating them as equal both linguistically and methodologically. The empirical research method of video occupies a unique position within these debates because it accounts for the corporeality of communication and examines the effects of gestures and facial expressions. The workshop aims to characterize the empirical expressions of embodiment and their theoretical foundations within the medium of video. On the one hand, it examines the claims to truth and validity of observed phenomena such as intercorporeality (e.g., Meyer/Streeck/Jordan 2017) or the corporeality of the sense of taste (e.g., Mondada 2019). What quality criteria are cited to establish these forms of corporeality as forms of sociality? On the other hand, the impact of specific forms of embodiment in social media is examined. Currently, so-called far-right memes are gaining significance as components of images and videos (see, e.g., Knopp et al. 2024). What effects of manipulation and influence do far-right memes utilize? What worldviews do they embody, and what opportunities for appropriation do they offer? Overall, the aim of the workshop is to discuss the various perspectives on embodiment as a basis for video as an empirical research method across methodological, disciplinary, and regional boundaries.

References:

Kissmann, Ulrike T./van Loon, Joost (ed.) (2019): Discussing New Materialism. Methodological Implications for the Study of Materialities. Wiesbaden: Springer VS.

Knopp, Vincent/Terizakis, Georgios/Denker, Kai/Groß, Eva/Häfele, Joachim/Pollich, Daniela (2024): Rechtsextreme Mime. Eine praxisorientierte Einführung für die Ausbildung in Polizei und Sozialwissenschaften. Bielefeld: transcript.

Latour, Bruno (2012): Technical Does Not Mean Material. Comment on Lemonnier, Pierre, 2012, Mundane Objects: Materiality and Non-verbal Communication, Walnut Creek, CA: Left Coast Press, HAU: Journal of Ethno-graphic Theory 4 (1): 507–510.

Meyer, Christian/Streeck, Jürgen/Jordan, Scott (ed.) (2017): Intercorporeality. Emerging Socialities in Interaction, New York: Oxford Univ. Press.

Mol, Annemarie (2002): The Body Multiple. Ontology in Medical Practice. Durham: Duke University Press.

Mondada, Lorenza (2019): Rethinking Bodies and Objects in Social Interaction. A Multimodal and Multisensorial Approach to Tasting. In: Kissmann, Ulrike T./van Loon, Joost (ed.). Discussing New Materialism. Methodological Implications for the Study of Materialities. Wiesbaden: Springer VS, 109–134.

Myers, Natasha (2008): Molecular Embodiments and the Body-Work of Modeling in Protein Crystallography. Social Studies of Science, 38/2, 163–199.

Big video data from sources such as surveillance cameras or police body cameras is stored and processed in information systems after it is recorded. Police officers in Germany, for example, transfer video recordings from their body cameras to analysis software. There, they have the option to flag individual video sequences for repurposing, so that they are not automatically deleted after 30 days (see the Police and Public Order Authorities Act). Subsequently, they can be used in a police case management system. The workshop explores the question of how researchers access their video data: How is the data collected, processed, and further developed? Which selection of video data is subjected to analysis? What premises are used for this? If big data is utilized, specific inquiries are made regarding the information and communication technologies used to produce it. Information systems have long been conceptualized in Science and Technology Studies as digital infrastructures whose operation is only possible through close interconnection with social practices (see, e.g., Karasti/Blomberg 2018; Clarke/Star 2008; Star/Ruhleder 1996). Publications of this nature emphasize in particular the relational character of digital infrastructures, because they become relevant only in situ and in relation to practices. Consequently, they can only be perceived and made visible when they enable activities. Otherwise, they remain invisible. While Science and Technology Studies highlight the relational character of infrastructures, Critical Data Studies focus on the performative power of the data itself. The latter assume that data do not represent a phenomenon, but instead generate something new and thereby transform knowledge (see, e.g., Ruppert et al. 2017). The common aim of research in these two traditions is to make visible the conditions of data use and thus the production of knowledge through data. Both traditions equally assume that the handling of data is political, because different data infrastructures would lead to different forms of knowledge. 

Digital infrastructures not only shape the social production of videos and thus ultimately the analytical process, but they also take on significant importance in the form of research data management systems. The workshop therefore also aims to make visible the digital infrastructures that enable the storage and reuse of videos. The workshop will examine which forms of video data and which contextual information – such as metadata – have access to the repository. How is a distinction made between quantitative and qualitative data? From an MMMR perspective, it is relevant to determine which form of video processing should be stored and made available for reuse. How is the encoding or transformation of video data accounted for? Which steps of the interim analysis are made transparent?

References:

Clarke, Adele E./Star, Susan L. (2008): The Social World Framework. A Theory/Methods Package. In: Hackett, Edward J./Amsterdamska, Olga/Lynch, Michael E. Wajcman, Judy (ed.). The Handbook of Science and Technology Studies. Cambridge, MA: MIT Press, 113–137.

Karasti, Helena/Blomberg, Jeanette (2018): Studying Infrastructuring Ethnographically. Computer Supported Cooperative Work (CSCW). 27(2), 233–265.

Ruppert, Evelyn/Isin, Engin/Bigo, Didier (2017): Data Politics. July–December, 1–7, DOI: 10.1177/2053951717717749.

Star, Susan Leigh/Ruhleder, Karen (1996): Steps Toward an Ecology of Infrastructure. Borderlands of Design and Access for Large Information Spaces. Information Systems Research. 7(1), 111–134.

At the end of the three-year funding period, a concluding conference will be held at the University of Kassel. While the three workshops are intended primarily for network members, invited guests, and individual interested outsiders, the concluding conference, through a call for papers, is open to all interested mixed and multimethod researchers who work with video as an empirical research method. Unlike the workshops, the conference will have an open theme.