A critical (theory) data literacy: tales from the field

Abstract

Purpose

This paper aims to explore how critical digital and data literacies are facilitated by testing different methods in the classroom, with the ambition to find a pedagogical framework for prompting sustained critical literacies.

Design/methodology/approach

This contribution draws on a 10-year set of critical pedagogy experiments conducted in Denmark, USA and Italy, and engaging more than 1,500 young adults. Multi-method pedagogical design trains students to conduct self-oriented guided autoethnography, situational analysis, allegorical mapping, and critical infrastructure analysis.

Findings

The techniques of guided autoethnography for facilitating sustained data literacy rely on inviting multiple iterations of self-analysis through sequential prompts, whereby students move through stages of observation, critical thinking, critical theory-informed critique around the lived experience of hegemonic data and artificial intelligence (AI) infrastructures.

Research limitations/implications

Critical digital/data literacy researchers should continue to test models for building sustained critique that not only facilitate changes in behavior over time but also facilitate citizen social science, whereby participants use these autoethnographic techniques with friends and families to build locally relevant critique of the hegemonic power of data/AI infrastructures.

Originality/value

The proposed literacy model adopts a critical theory stance and shows the value of using multiple modes of intervention at micro and macro levels to prompt self-analysis and meta-level reflexivity for learners. This framework places critical theory at the center of the pedagogy to spark more radical stances, which is contended to be an essential step in moving students from attitudinal change to behavioral change.

Citation

Markham, A. and Pronzato, R. (2024), "A critical (theory) data literacy: tales from the field", Information and Learning Sciences, Vol. 125 No. 5/6, pp. 293-320. https://doi.org/10.1108/ILS-06-2023-0087

Introduction

Datafication, everyday surveillance processes, and automated decision-making are such a taken-for-granted part of everyday routines that they are no longer even noticed but simply reproduced. From a Frankfurt School critical theory perspective, this means that the extractive practices and domination of corporatized interests are ever-more obscured for everyday users of digital technologies, in all areas of work, learning, and everyday life. By now, a strong tradition of scholarship focuses on how critical literacies might be fostered, the sort of literacies that move beyond competence to critique and deep analytical understanding of the politics, ethics, and social implications of digital transformations. As part of this community of scholars, we have been testing methods to facilitate strong critique from a critical theory standpoint, not only to build awareness of hegemonic and other forms of power, but to spark more sustained transformations of attitude and behavior. How to achieve these more sustained levels of critical literacy remains one of the most significant current discussions in learning and information research.

This contribution focuses on a set of critical pedagogy experiments conducted by the Authors in three countries, informed by Markham’s (2019, 2022) conceptual and pedagogical frameworks. Specifically, we showcase the elements of this overall pedagogical model that highlight a deliberate critical theory stance. Drawing on our experiences from working with students across many in-classroom activities aimed to facilitate critical digital and data literacy with young adults (primarily aged between 18 and 30), we discuss the significant value of combining guided autoethnography (Markham, 2022) with critical theory and creative play to facilitate a process whereby students can identify and critique structures of power and control in their everyday digital media use.

Building critical consciousness – of the sort advocated in the work of critical theorist Antonio Gramsci (1937/1971) and critical pedagogy activist and educator Paulo Freire (1970) – requires digging under the surface level of data-saturated experiences to identify and then critique hegemonic ideologies and power imbalances. In this context, we show how participants can be helped move from initial critique to critical theory style analysis, and through iterative loops of self-reflexive analysis, shift from attitudinal awareness to behavioral shifts, which we believe has stronger potential to lead to sustained critical consciousness.

This paper focuses on the curricular model and pedagogical underpinnings, beginning with a discussion of how the power of computational systems becomes hegemonic. We review current approaches to data literacy. We then introduce the tenets, experimental design, core elements, phases, and building blocks of our critical theory based digital literacy pedagogy framework. In the latter parts of the article, we reflect on the extent to which the framework promotes critical awareness and behavioral change.

The social power of big tech

In the last decade, such terms as “big data,” “algorithms,” “digital platforms” and, more recently, “artificial intelligence” (AI), have become common in media outlets and public debates as broad sweeping terms to encapsulate the central features of computational technologies, most often produced by big tech companies and materialized through software programs that are then experienced through an interface. These terms have also come to be used by scholars to refer to complex sociotechnical assemblages and, more specifically, to “the computational networks in which they function, the people who design and operate them, the data and users on which they act, and the institutions that provide these services” (Gillespie, 2016, p. 26).

Computational systems and their underlying datafication processes significantly shape and delimit contemporary communication processes. Algorithmic systems function as interlocuters; they adhere strongly to specific protocols for symbolic exchanges and simultaneously participate in everyday interactions as conversational partners, as when the auto-predict function of Google Search responds to a person’s query continuously even while the person is still formulating a search query. Generative AI systems such as ChatGPT are even more overtly conversational, which further obscures the delimiting and shaping aspects of the machinic processes. Although humans are not without agency in these human–machine interactions, the legacy of critical theory teaches us that habitual interaction in systems with rigid procedures and invisible or bureaucratized processes can naturalize these persuasive processes as “neutral” infrastructures, dismissible as “just the way things work,” which over time obscures the ideologies and corporatized interests that secure and maintain power through these processes and structures.

The datafication power of platforms, increasingly entangled in every potential realm of contemporary social life, can be situated as a part of a long-standing social dynamic whereby computational processes intervene and structure social life to favor the extraction of value, producing “the social for capital” (Couldry and Mejias, 2019, p. 339). These dynamics are in turn embedded within long-term colonialist trajectories, based on the expansion of an “apparatus of measurement” (Beer, 2016, p. 3), which functions, not least, to reproduce neoliberal rationalities and “imperialist modes of accumulation” through various forms of extraction (Sadowski, 2019, p. 3).

It is well established that algorithmically driven data modeling and analytic processes are based in, generate, and reify particular and asymmetrical power relations and dynamics. These can be both obvious and nonobvious, and it is the latter type of instantiation that buries power in the operations of what might be dismissed by general tech users as merely part of the features or default parameters of use (c.f., van Dijck, 2013). For example, as Amoore (2020) notes, machine learning algorithms are not only responsible for the recognition of individuals and objects “in the sense of identifying – faces, threats, vehicles, animals, languages,” but also for the reification of a particular system and logic of recognizability, in that “they decide what or who is recognizable as a target of interest in an occluded landscape” (p. 69). These “regimes of recognition” (Amoore, 2020) can become understood as impartial, yet result in diverse forms of discrimination. This aligns with earlier critiques of algorithmic biases (e.g. O’Neil, 2016; Noble, 2018; Eubanks, 2018), detailing how already marginalized groups suffer the most from the amplification of injustices, automated by datafication procedures/software programs used by the state or by private companies.

The role of everyday social narratives cannot be downplayed here, as “[t]he way that data and analytics are imagined shapes their incorporation and appropriation into practices and organizational structures” (Beer, 2018, p. 465). Asymmetries of power, whereby certain stakeholders benefit at the expense of others, tend to get reinforced even as they are neutralized as “inevitable” in everyday explanations (Markham, 2021a), as people use, talk about, share practical advice about, and otherwise discuss contemporary digitalization and datafication. The discursive process of “closing off alternative” ways of thinking can be suitably illustrated by thinking about what happens at the surface level of communication or communicative actions, and under the surface, at deeper levels where premises and interpretive frames for sensemaking inform communicative actions, as elaborated by various critical organizational scholars (c.f., Mumby, 1988; Deetz, 1992; Heracleous, 2006; Markham, 2021a).

Scrutinizing these imaginaries requires awareness of how seemingly banal communicative actions generate or reinforce frames of reference that are difficult to sidestep. Analyzing the metaphoric frameworks around the concept of data, Markham (2019, p. 757) notes that the first step of combating the way datafication gets neutralized and naturalized is to shift the debate away from data as “objects” or “evidence” and “toward data as ideology.” Markham (2021a) further notes that the power of such terms as AI and “big data” revolves around the rhetorical construction of technological and social futures that are monolithic and inevitable, which builds a frame of reference whereby users should simply accept this as the way things are.

Digital literacy responses to digital and data disruptions

As highlighted by Buckingham (2007) and many since, digital literacy is frequently framed in public discourse through a “competency-based approach,” within which the term “‘literacy’ is used merely as a vague synonym for ‘competence’, or even ‘skill’” (p. 43). These approaches do not typically recognize the ideological entanglements constituting digital technologies and therefore present a restricted and instrumental view of information as something that “can simply be assessed in terms of its factual accuracy” (p. 46).

A decade later, Pangrazio (2016) reiterated and extended this criticism, highlighting the discrepancy between “technical mastery” and “critical mindsets” that risks undermining the cultivation of “a critical disposition in a context in which technical proficiency is prioritized” (p. 163). Just as media literacy was blended with digital literacy in the 1990s, the recent decade blends digital literacy with data literacy. Here, the interest is on datafication processes and data analytics. One could easily notice the same absence of criticality at the outset of “data” literacy discussions, especially among large-scale entities such as the UN or research funding bodies such as the European Research Council, who, as Gray et al. (2018) noted, promoted literacy as technical competency, the goal such as “being able to access, analyze, use, interpret, manipulate and argue with datasets in response to the ubiquity of (digital) data in different fields” (p. 2). As ethical crises in the mid-late 2010s continued to highlight the biases and problematic sociotechnical systems underlying the so-called advancements in big data aggregation and machine learning, there has been a groundswell of critical responses to how data literacy should be framed and taught.

This shift in thinking about how to facilitate digital literacy acknowledges the larger sociotechnical contexts for digital transformations, addressing the complications of datafication, digitalization, automated decision-making, and AI within larger ecosystems. Recent literacy scholarship highlights that being an effective citizen requires the capacity to contextualize the production and adoption of data artifacts (e.g. Pötzsch, 2019; Sander, 2020) or to understand the larger situations within which certain datafication or digitalization processes exist (c.f., Crawford and Joler, 2018). This ecological turn in media/digital/data literacy studies intentionally expands the focus of analysis to help users consider how specific experiences, technological affordances, device features, or norms are situated in larger media or sociotechnical ecologies (c.f. Hobbs, 2022, drawing on the legacy of critical media theorist Neil Postman; Milner and Phillips, 2020; Paris et al., 2022a, 2022b).

Within this ecological framework, however, there remains a significant challenge of shifting users from comprehension to critique, as Sefton-Green and Pangrazio (2022) note. Markham (2019) addresses this by connecting critical pedagogy to critical theory, clarifying that the “absence of criticality” is better understood as an absence of a critical theory mindset, and advocates taking a deliberate and strong “critical theory stance” in the classroom (p. 757). She notes that the power imbalances are deeply hegemonic, meaning people are operating in systems of what Gramsci would detail as “control through consent.” Because algorithmic and data analytic processes of platforms are often so helpful, and the platforms themselves are portrayed as neutral portals for information flow, countering their pleasant hegemony requires an overt emphasis on their “wrongness.” To “question everything from the perspective of ‘who benefits and who loses?’” (Markham, 2019, p. 757) compels people to step back from their use to observe as a critical theorist would, “to investigate the who, what, where, when, and how of this wrongness” (p. 757). This suggests that effective critical literacy training would require multiple steps, a chain-link process of effects to yield sustained critical awareness of the hegemonic situation, an ability to imagine otherwise, and a willingness to act otherwise. This is not intended to prompt users to reject the technologies, but to visibilize and counteract the neutrality being presented to deepen their analytical perspectives, which nudges them to “supersede their magic or naïve perception of reality by one that [is] predominantly critical” (Freire, 1974, p. 40) or more firmly grounded in the actualities of the larger ecosystems of power (c.f. Chomintra, 2023).

Within critical data studies and critical data literacy studies, the goal and methodology described below aims to find effective tools to facilitate a critical literacy that shifts from “awareness” to “critical consciousness,” to “critical theory critique” and then to “behavioral change.” In the next paragraphs, we explain the methodology and the pedagogical science underlying our activities.

Overall experimental design

Background, setup, and variations

Markham’s domain expertise and teaching and learning experience in critical digital studies, forms the background for this combination of pedagogy and participatory research design. She has spent ten years (2012–2022) building guided autoethnography as an approach for facilitating a critical theory-based digital literacy pedagogy. This includes an annual autoethnography project in classrooms, involving over the years more than 1,500 local and international students at the bachelor’s and master’s level in the USA, Canada, and Denmark. As part of courses related to digital media, digital culture, or research methods, students have been trained in multiple approaches and tools of inquiry to analyze their own relationships with digital technologies. A series of deliberately sequenced prompts and exercises are used to guide a user’s phenomenological and autoethnographic deep dive, as well as critical theory analysis of their own lived experience with digital media. Using this methodology, Pronzato has conducted similar exercises in 2021 with students in Italy. The techniques have been developed and tested with students from a range of different countries. While we cannot ignore the students’ level of education, socioeconomic background, location, and other factors, the patterns of responses among students have been remarkably similar across these years. In this article, the more salient features of this process are the sequencing of events, phrasing of prompts, and repetition of certain elements.

The guided autoethnography was originally structured as an eight-week in-class exercise, and after several years of testing, condensed variations have been adopted. For example, Pronzato’s project included using some exercises in a bachelor level course in Italy followed by a one-week guided autoethnographic project, with students who elected to become participants after the course concluded (2021). The procedures were informed and guided in consultation with Markham. Other variations have been adopted. In 2022–2023, the research team in a European funded project (TRAVIS) tweaked the format to use a variation on this set of exercises with participants in Austria, England, Estonia and Finland. In Denmark, Rehder et al. (2019) discuss their use of a variation on this set of exercises to train Danish teachers through VIA University and University College Copenhagen. Each variation uses slightly different frameworks and underpinnings; we mention them here to highlight that this model can be adapted. Our discussion herein draws on the authors’ own experiences. Any data donated by participants after the courses were finished has been used with permission, with institutional or noninstitutional informed consent. We describe the ethical parameters more in the next section.

A context sensitive and “beyond regulations” approach to ethical research design

The overall aim has been to develop fruitful scaffolding techniques for digital literacy in higher education settings. Collectively, the outcomes of these experiments could also be described as a ten-year study of the lived experience of young adults in contexts of constant connectivity and ubiquitous digital media. However, research was not conducted with the goal of collecting or analyzing user data. The data was collected by users for their own autoethnographic explorations. At the same time, these large and mixed-media materials were submitted as course assignments and have been viewed by instructors. The materials are often sensitive, and third parties included in screenshots or screen recordings are often not initially adequately anonymized. This research and curriculum design has required careful and proactive ethical considerations that go well beyond regulatory guidelines or limitations.

The classroom exercises and the overall study have been guided by ethical best practices more than regulatory approvals, because institutional oversight was not often required. At the outset of the courses, students have been informed that after the course ends, they will be invited, but not required, to add as much or little of their raw data, analyses, and/or ethnographic accounts to the researchers’ archives as they like. With permission, some of these materials have been used to identify and discuss larger patterns of digital technology use for the past decade, or in the case of this article, to illustrate some of the participating students’ work. Because there are considerable ethical considerations, we elaborate in detail below.

In the classes, students are required to complete the autoethnographic activities, because they are part of the course curriculum; that is, learning tools for inductive and adaptive qualitative research methods, conducting empirical research to examine their own digital media use, learning to study digitally saturated social contexts, and examining their relationships with algorithms. Significant effort has been made in each iteration of this curriculum to distinguish “classroom” activities from the later possible participation in a “research study.” In all three countries, students have been informed in advance that the exercises were designed as part of a larger project focused on digital literacy methods, and as a side benefit, the materials have helped the teachers understand more about digital lived experience over time. They are informed repeatedly that they are not participants in this study and will not be invited to participate until after the course ends and they have received their final evaluations/grades. Even with this disclaimer, we feel it is crucial to ease any pressure students might feel about contributing sensitive fieldwork materials to what may seem like some unknown, larger “study.” Students are therefore assured on several occasions during the course of the exercises that “coursework” they submit is not included or archived for use by the professor/researcher.

Students are also not required to submit personally identifiable information, most noticeable through images, screenshots, screen captures, or other audio/video materials. They are given the option to submit field-gathered materials or ethnographic analyses in writing, rather than audio/visual form. Students also undergo rigorous ethical training to use methods to protect the privacy of their friends and family, whose activities or images may be captured in the student’s fieldwork.

After the courses end, students are invited to donate their data to the archive of materials or to be considered “participants” in the larger study. Once they receive their final grades for the course, they are sent an email invitation along with a project information sheet and consent form. If they opt in, we archive materials they elect to share. This does not guarantee their materials will be used, but allows for possible participation in further studies and any who opt in may be contacted. Over the years, however, minimal contact has been made after the courses end, because, as mentioned above, the objective is primarily pedagogical.

Archived data is secured by the primary researchers following the European Union’s general data protection regulations (GDPR) rather than institutionally specific data protection protocols, because institutional and regional models vary. Data are not anonymized because the visual and vocal data are inextricable from personally identifiable information. Almost all material shown to subsequent students or in publications is anonymized or altered to disguise the source. In some cases of showing examples to future students, materials have been unaltered, but used with express permission. Demographic details and personally identifiable content is anonymized by altering content, e.g. generating composites from multiple participants, following Markham’s advice for fabrication as ethical practice (2012).

Institutional and noninstitutional ethical oversight.

To clarify the institutional oversight by ethics review committees: In 2012, when this course was conducted in a US-based higher education institution, there was no requirement to submit a proposal to an institutional review board (IRB) for normal autoethnographic fieldwork conducted as part of research methods courses. There was also no plan to collect data as part of a larger study. Later, Markham and her 2012 co-teacher sought and received IRB approval at Loyola University Chicago to ask the former students to donate their autoethnographic materials for further analysis. Students were recruited for participation after the course ended, with institutionally approved informed consent procedures. From 2013 to 2021 in Denmark, when the course was conducted in Danish higher education institutions, ethics review from external or institutional committee was neither required nor sought, for either the classroom exercises or the overall conduct of the study. In the Danish and university system during this time, responsibility for ensuring that individual projects comply with research ethics or data protection regulations (such as GDPR) generally lies with the researcher. Ethical guidance may be sought to obtain advice for research that is considered problematic, but until quite recently (circa 2021), ethical oversight or approval committees do not exist to regulate researchers. Markham proactively used best practice and regulatory guidelines to inform her decision that students would not be automatically included in the larger study, but invited only after the course ended, with a formal informed consent process, drawing on procedures common in the USA. In Italy in 2021, ethical approval was not required for conducting autoethnographic analysis for purposes of classroom training. Informed consent is required to recruit participants for PhD-level research. Following institutional guidance and best practice, Pronzato waited until after the course ended to recruit former students as participants and used an institutionally approved informed consent procedure.

Throughout this project, best practices and conservative ethics design have been followed to gain informed consent, ensure voluntary participation, enable withdrawal of consent, and store/use donated data in ways that conform to GDPR standards. This includes the multiple guidelines developed by the Association of Internet Studies for researching digital contexts (Ess and The Association of Internet Researchers, 2002; Markham and Buchanan, 2012; Franzke et al., 2020). Further guidance can be found in the details of the GDPR documentation, national statements such as the Australian code for the responsible conduct of research (NHMRC, 2018), or guidance from governmental funding bodies such as the U.S. National Science Foundation, which provides useful interpretations of the US “common rule” through vignettes and scenarios (NSF, 2023).

For the research design described herein, which cuts across multiple countries, uses data produced in classrooms, and includes a multi-year research study, the ethical parameters must be researcher-driven, proactive, and extend beyond extant regulatory limits or norms. Researchers must be aware of laws that govern consent, privacy protection, and data management. This requires considerable attention to how ethical situations evolve and a more conservative approach that what might be legally required. As Markham (2018) has noted earlier about this, in times when technologies or regulatory environments are continuously changing, a conservative approach that goes beyond most regulatory guidelines is warranted. This stance is less about limit what sort of data might be collected, and more about using extreme caution in how and when consent is discussed and obtained, how data is stored, and how it is presented.

Regarding the materials used in this article: All instances are labeled. All materials are anonymized or altered, have been donated by participants, and are used with informed consent permission, through both noninstitutional and institutional informed consent processes.

An indirect approach to facilitating critical theory thinking

The courses have not been focused directly on critical approaches to digitalization or datafication, but rather, to build skills in mixed-methods, multimodal, inductive, and emergent qualitative approaches to the study of sociotechnical contexts. One can see this in some of the course labels, such as digital ethnography, qualitative research design, digital identity, digital culture, and social science research methods. The methods presented are generally grounded in an ethnographic epistemology, and the goal was to help the students build a toolbox of techniques to defamiliarize situations, reverse engineer structures and outcomes to see processes and decisions, and bring a forensic attitude to bear on situations, to zoom in on particular elements or processes in the overall ecology. This would help students identify what was performing with agency or agential force to influence the situation. This in turn would help them identify stakeholder interests that may be exerting power in situations.

The focus on building “critical theory” thinking among participants is deliberately subtle, blended with “critical pedagogy” techniques. In the classroom, we might talk about “critique,” “social change,” and “reflections on what is possible and desirable in the future,” rather than use the phrase critical theory , a technique earlier elaborated by Markham (2019). The critical theory underpinnings are guided by the premises of late-20th century critical organizational scholars who draw on Frankfurt School critical theorists such as Adorno and Althusser to talk about how people are interpellated or “called” into subjective positions by media – and later, on Foucault, to discuss how power is manifested in structures of thought or power/knowledge systems. The late-20th century of critical organizational studies emphasized how power dynamics are made invisible because they are enacted, and normalized, in the premises for action, rather than actions themselves (Lukes, 1974), operating at “deep structure” levels of discourses or communicative actions (c.f., Mumby, 1988; Deetz, 1992).

These concepts are not frontloaded in the curriculum or discussions. Although the vocabularies of relevant critical theorists are not emphasized, such as Althusser’s “interpellation,” Lukacs’s “reification,” or Gramsci’s “hegemony,” these concepts are explored indirectly, by considering how an everyday user’s ideas about how the world works are built and reinforced through both repetition and neutralization. Gramsci (1937/1971) plays a central role in the pedagogical design, in that we are deeply concerned about hegemony, whereby over time, through consensual participation in these systems, a powerful loop of invisible control is reinforced that privileges corporatized and capitalist interests and retains control through consent that is buried in infrastructures, defaults, and everyday banal procedures of being social or accomplishing personal tasks through platforms (see also Pronzato and Markham, 2023).

In this, it is important to note for critical theory informed readers that we are much more concerned with the application of pedagogical methods to build critical interrogations than the development of students’ theoretical expertise in critical theory. This follows Markham’s recommendation to “soft pedal” critical theory to make it more accessible and usable. Rather than discussing theories of bases, functions, and technologies of power, for example, we provoke students into realizing how, in algorithmically driven platforms, more likely than not, some other entity in this system has more power over their choices than they do themselves, and to consider how they might further explore and critically analyze this situation. In other words, while the techniques draw on classic Frankfurt School critical theories, these are more applied and performative than discussed as concepts. Later, as we reflect with students on how we/they might interpret their autoethnographic findings, we can begin to connect to classic and contemporary critical theory approaches to digital transformations in societies.

Prompting mixed methods to yield complex interpretations

The overall pedagogical design follows the principles of inductive approaches to ethnography, which seek to understand unique patterns of experience in naturalistic environments. This involves observing and document their social behavior and online activities at the granular level as if they are foreign to themselves, which includes documenting online activities in detail, reflecting on these activities through a range of written and visual techniques.

The exercises vary but generally have included creating automated or manual logs of activities, recording video or audio reflections, as often as needed and sometimes in response to specific prompts, writing freeform diary entries as “self-directed introspective elicitation” (Markham, 2017, np), making many versions of situational maps, either through styles of mind mapping, concept mapping, relational mapping, and allegorical mapping (following techniques of, e.g. Clarke, 2003; Jones and Harris, 2016; Markham, 2022), and building interpretations through writing of code memos or scene-setting vignettes.

All the exercises are designed to function as breaching experiments (Garfinkel, 1967) to disrupt the everyday use of digital platforms and to highlight epistemological and ontological frames of reference; those structures and relationships underlying and guiding particular use patterns. The material that emerges in written, audio, and visual form provides fuel for reflection, both as one analyzes these materials as ethnographic field notes and also as one engages in different levels of meta-analysis, whereby participants reflect on their reflections of their own feelings, attitudes, and assumptions during the research process.

The exercises, individually and as a whole, invite reflexive examination of one’s lived experience and personal involvement with digital platforms. When focused on making sense of datafication processes, these exercises build more complex contextualizations of datafication structures and their intertwining with everyday practices, which critical data literacy scholars highlight as essential. Another way to put this is that reflexive accounts emerge as students observe, document, reflect, and reexperience situations through an analytic lens, focused on themes and patterns they see emerging.

Careful attention to how exercises and prompts are framed and sequenced

Building analytical skills requires attention on what happens below the level of “method,” as one engages in practical habits of thought or applies a certain mindset to the task at hand. Thus, while one ends up teaching techniques of ethnography, ethnomethodology, and phenomenology, these terms are not mentioned until much later in the process, as they require a lot of explanation. More centrally, these concepts are far more obfuscatory than helpful at this stage and raise barriers to entry for the young adults we are working with.

To get students to work at the level of tool versus method or methodology, Markham (2022) has designed the phases, exercises, and prompts in such a way to foster the ability to adopt a forensic attitude, useful for defamiliarizing and then reverse engineering moments, encounters, actions, or situations (see Figure 1).

This granularity helps students build material that favors the identification of what system level (ecological) processes and dynamics might have occurred to produce the “situation,” which can be further specified as the “outcome of interactions.”

The exercises are conceptualized as sparks to generate action, as “[i]t is precisely in the microscopic that we find the so-called seeds of change” (Markham, 2021b, p. 915). Indeed, directing attention toward tacit, microscopic properties of interactions between the self and the platform, and the following connections with massive-scale global dynamics, can enable “the microscopic-as-the-whole, or the whole as a way to make sense of the granular” (Markham et al., 2021, p. 759). In this context, moving “past observation and even beyond sensemaking” can help turn reflexive engagements with others into action, thereby helping “small and local sensibilities enter the larger symbolic interaction playing out across global networks” (Markham, 2021b, p. 923).

The specific phases and building blocks

Phases

Although awareness is crucial, moving from awareness to critique and from attitude changes to behavioral changes in regards to datafication can be considered the aim and ambition of the critical data literacy approach we detail below. The “critical theory” angle shifts the ethnographic lens from “What is going on here?” (focus on thick description) to “Something is going wrong here” (focus on critique).

Practically, the methodology is based on the following basic phases through an autoethnographic study, conducted in somewhat overlapping and iterative fashion:

• fieldwork carried out by participants, in the form of observations of their everyday activities when connected and observations of lived experience when disconnected;

• collection of material evidence of digital communication practices, in such forms as logs, diaries, artifacts, and field notes produced by the participants;

• self-elicitation of participants’ own perceptions and attitudes regarding their activities and expectations with/in digital media; and

• first-level self-analysis and second-level meta (reflexive) analysis of participants’ own observations and materials through various multimedia formats (written, audio/visual recording, drawing, and mapping).

This design strives to facilitate autoethnographic observations, iterative sensemaking during analysis phases, and layered or bricolage ethnographic accounts. This autoethnographic design is premised on the idea that sustained critical literacy is more easily achieved if we begin focusing on the self and then extend the lens outward. The initial sparks of insights, from an ethnographic perspective, emerge through defamiliarization, reverse engineering, and a forensic attitude, or distancing oneself from one’s everyday and taken-for-granted activities to observe oneself as a foreign object.

This implies oscillating from close or intimate levels of knowing to more distant levels of observing, with the goal of bringing some embodied and naturalized behaviors, attitudes, and values to the surface while documenting them.

Regardless of the type of exercise or adaptations in the overall experiment, as noted above, we have learned that prompts should be carefully phrased and sequenced, which has at least two benefits: first, if clearly articulated, prompts will guide participants toward a specific angle of focus, which helps build depth of analysis within specific instances. Second, if prompts are given individually in a sequence, rather than lumped together in a single exercise, this can help participants separate description, emotion, analysis, and reflexivity (meta-analysis).

Because the phases tend to overlap and become iterative, Markham (2022) has observed that these can become muddled for students. To provide clearer (or perhaps simpler) scaffolding, the actual exercises are not aligned to the phases but rather, organized in building blocks, or sets of exercises and prompts. We describe the core building blocks of this pedagogy below, and more details about how these are framed as in-class assignments can be provided on request from the corresponding author.

Building block 1: Tracking

Tracking one’s online activities is a key first step of the ethnographic experiments. This takes place at early stages and involves tracking one’s own types and frequency of usage across or within platforms and devices for approximately 48 h, although in shorter variations, one could monitor during a single day. This might seem distant from what is generally considered ethnographic fieldwork, but this is precisely the beginning steps of any ethnographic observation in a new environment. Students are given instructions to be obsessively persistent in this “close level observation,” to generate detailed data regarding their own movements, locations, purposes and actions on digital media. This is deliberately phrased “close level observation” rather than “ethnography” or “participant observation” for reasons mentioned elsewhere in this article.

Before they begin the actual tracking, we suggest they build a logistically oriented “tracking plan.” By imagining and then articulating the steps involved in the process, students begin to consider how this intensive tracking and logging through bodily actions will be accomplished. Then, they can start generating data about when they check their devices, where, how long, which contents they consume, how often and so on. They are prompted to be repetitive, over the allotted timeframe, which produces not only more detailed data for later analysis, but a data set that they might notice changes as they become more attuned to the activity and their own observational tendencies. The typical outputs of these exercises are screenshots, screen recordings, auto-loggings, or text notes (see Figure 2 for some example screenshots), which can all function like self-elicitation tools later. Importantly, the tools or techniques are not preselected or standardized, as participants produce data that is more meaningful to them when they are using tools that work for their unique preferences, tendencies, and situations. They share their tracking plans in small groups, however, which broadens their scope of options. Thus, while some will use automated tracking tools, others will choose more manual options such as writing diary-style entries, keeping a log in a small physical notebook they carry around with them, or making notations in excel spreadsheets at timed intervals.

After tracking their own digital activities for 48 h, they are given a limited timeframe in which to reflect on these materials through the production of three short (less than 5 min) vlogs all focused on a single moment or interaction that is typical but also stands out in their fieldwork for some reason. Each vlog responds to a different question about the same situation. They can choose between audio narration over video clips or images, a talking head video, or written narrative essay. Questions are shown in Figure 3.

Taking at least three different angles on the same small piece of field data is an iterative repetition that helps defamiliarize the situation and also foster a forensic attitude whereby one is focusing on the micro-level details of actions as parts of larger sequences of actions and responses of various entities within digital ecosystems. The exercise fosters reverse engineering the outcomes of interactions to identify and link decisions, actions, and outcomes, and to connect these in turn with the entities operating with agential force in the situation.

Trial and error have shown that the sequence of these three questions is important; the initial question is meant to yield detailed and mundane descriptions that are informative. For example, a student we can call Participant A notes:

According to my logs, on Friday, for most of the tracking I was sitting in class, which is why my computer usage is quite high. I was mainly working in Word and Google Docs, but have also been researching a lot through Google Search and Wikipedia to get more knowledge about the topics discussed in class. It is also noticeable that there are some applications I used, which I know are mainly for leisure time. lnstagram for example I mostly use in the morning or in the evening when I come home to relax. Facebook and email I use for communicating with others about class. Spotify I use while running.

…I post or reply mostly in Whatsapp (average 130 times each day), followed by Instagram (average 20 times every day, and average 100 posts and comments each day). I checked into Facebook only 4 times on average and did not post any content, but liked 10 posts on average each day. (Participant A, modified to anonymize, used with permission via data donation)

The second question is designed to analyze details from the first question, adding information about what is missing. Often in this second step, participants begin to notice perceived or technical affordances of platforms and devices, and notice the way habits, norms, or other aspects of culture seem to delimit as well as guide actions. For example, the same participant A continues, building a cultural or country-specific link between their preference to use Instagram versus Facebook:

Why do I do X instead of Y? This is a good question. I notice I do not post very much personal information in English and definitely not on Facebook, since arriving in this country. The things I see in my Facebook feed are mostly not from my home country friends but from people from other countries. It seems to me that in my country it might be not that usual to keep status updates on Facebook or get personal there. Of course, this is quite a generalization and it might also just be connected to my closer circle or to my age group. However, lnstagram seems to me more like a platform where you have an inner circle, where things are quite private and where you are just with your friends - at least, the way I use it. Facebook, however, seems more public to me.

…I’m beginning to realize platform use is extremely cultural and I didn’t ever think about this before. Need to think more about age and gender differences as well! (Participant A, modified to anonymize, used with permission via data donation)

The third question is designed to build more reflexive layers in their self-oriented analyses. Focusing on emotions provides a low barrier to entry for students to figure out that sensemaking is a subjective and layered practice, which helps them build analytical complexity. Reflexivity builds over time and repetition. Even so, staying with the same participant as above, one can notice traces of meta-level reflections by Participant A in describing what they felt was extremely high usage of WhatsApp:

How do I feel about my extremely high Whatsapp usage? I rely on this quite a lot, and when I did not have access to it, I was quite stressed. This is an important form of communication for me, and I always answer right away, even when it is nothing important. I feel like I need to always be in communication and I am a bit inflexible when this is not possible because I prefer to immediately engage.

…WhatsApp is making this very easy, especially since it is possible to use it on the computer because it makes the typing so much faster.

…How do I feel about this? I am starting to realize that there are a lot of habits we just internalize. For me, it is this ‘between thing’ that I need to do- not just in WhatsApp. To check my phone after doing something or I wake up and I have to look at my phone and stuff like that, it is just really internalized. It does not really follow a purpose but this is just what I am used to doing. (Participant A, modified to anonymize, used with permission via data donation)

These samples from Participant A help illustrate that these tracking exercises can yield detailed data about one’s own use that is surprising. It also showcases how a small pattern like number of times one uses Whatsapp versus Instagram versus Facebook over a two-day period can yield an analysis that notices culturally specific behaviors, subtle distinctions in public/private spheres according to platform norms, technical affordances of Whatsapp, and reflections on identity.

Building block 2: Disconnecting

Following this intensive tracking, students are asked to disconnect or take a “fast” from digital media for 24 h. The aim is to experience, observe, and document, through self-oriented elicitation activities, what it feels like to be disconnected.

Disconnecting is intended as deeply defamiliarizing “breaching experiment,” [1] i.e. students are able to bring to the surface aspects of digital experiences and situations that are usually obscured because they are naturalized, habitual, and tacit. Despite the conceptual idea that “there is nothing to disconnect from in the digital world” (Bucher, 2020, p. 610), as “there is no clear outside to machine learning algorithms” (p. 615), this activity is generally very difficult for young adults, and evokes powerful responses, which are important to address (through classic forms of debriefing) when the exercise ends.

Participants focus on their affective response to being disconnected and reconnected and to combine these reflections with a detailed report of one’s activities, through written or audio-visual materials produced just before, soon after beginning, and just after completing the “fast.” Specifically, students are asked to iteratively focus on two questions, either at certain times, or whenever it feels “right”: (1) “What is going on here?” – to be observational and descriptive regarding online devices and their own behavior – and (2) “How does this feel?” – to provoke an open-ended focus on affect and emotion (notice these are the same as Questions 1 and 3 in Figure 2 above). The students are asked to then make a final vlog (or equivalent written or audio response) on this activity approximately 24 h after they reconnect, to reflect on how they felt about this experiment.

Building block 3: Situational mapping

Visual mapping techniques are useful for students to explore situations using more embodied sensibilities than writing or talking. These are used to analyze situations as well as to generate additional layers of data for further analysis. This type of mapping, using space on large pieces of paper, drawing lines between nodes, placing objects or ideas in visual relation to other elements of situations, brings embodied knowing to the foreground, emphasizing the role of emotions, subjectivity, place, and connections in the context of online engagements, whether these engagements involve other persons, nonhuman or more-than-human agents, or multi-entity agential forces. Here, we mention two common procedures asking students to explore through mapping what they may have observed earlier in tracking. These mapping exercises focus on micro-moments, as a single moment quickly expands out as they consider the larger terrain or scapes within which their behaviors are occurring.

First, participants use Clarke’s (2003) situational analysis to identify as many of the human and nonhuman agents in a particular situation as possible. These maps are drawn like concept maps, with nodes and lines. A key part of this initial mapping exercise is that several maps are built, using the same, or tertiary or banal elements from the initial map(s). Each new map centers a different aspect of the situation and brainstorms anew, with a slightly different, self-selected question. Over the course of several maps, new and surprising elements emerge that would not be identified on only one or two maps. Figure 4 shows some of the variations of mapping that have been devised, including pen-and-paper concept mapping, post-it-note mapping that can be rearranged repeatedly, and a visual map focusing more on icons and emojis than words.

Second, participants are invited to draw allegorical maps of themselves while using digital platforms (see samples illustrated in Figure 5). Cartographic maps are part of a long history of attempts to understand not only the location of humans in natural territories and surrounding boundaries, but also the territories and boundaries of the Self. As argued by Braidotti (1994), drawing “camp sites,” i.e. maps, “traces places where” individuals “have been in the shifting landscape of [their] singularity” (p. 46). In 2021, Pronzato focused attention on the value of having students produce allegorical maps. This draws on the Victorian era practice of drawing maps of the sentiments as if they were territories or regions laid out cartographically (see Figure 6). Following Markham and Harris’s (2021) use of this technique, the exercises were designed to focus attention on what Jones and Harris (2016) would describe as the “shifting, temporary relationship between experience (a doing), the corporeal (embodiment), and the known (knowledge)” (p. 5).

The exercise works very well to foster rich and thoughtful allegorical explanations of experience, as Participant B’s comment in Figure 7 illustrates:

Producing multiple maps is an essential part of this exercise, as each pass through the situation yields a different “map,” promoting a sensibility that reverse engineering what might be involved in creating a situation never yields a single cause-effect sequence, but many possible confluences. Specific analytical decisions will produce a specific “path through meaning” (Markham and Gammelby, 2018, p. 453). It confronts some of the simplifications often used to explain situations and is good training for interpretive qualitative methods, as well as ecological thinking.

Building block 4: Analyzing through brain dumps

Throughout all the phases, individuals are asked to write in response to suggested or self-developed prompts. This is designed to be iterative and repetitive, to build additional material that can be studied, or to produce different levels and types of self-reflexive sensemaking. Whereas the initial observation writing (e.g. in response to Question 1 in Figure 2 above) focuses on what is happening, following the guidance of Emerson et al. (2011), later writing, especially in the reflections during Building Blocks 1, 2, and 3, works toward what Geertz (1973) would call “thick descriptions” connecting their lived experience to larger “webs of significance” (p. 5).

We use Markham’s (2017) “self-directed introspective elicitation” exercise, an adaptation of well-known techniques for brain dumps that encourage specific self-interview style prompts to draw out tacit intersections of the self, local experience, and larger sociotechnical ecologies. Using a timer, students respond to a prompt by writing whatever comes to mind, without stopping, reading, or self-editing (typically they write with white font on a white background, or otherwise obscure the screen or paper). As with other exercises, repetitions are key to produce further material allowing the analysis of one’s activities, choices and sensemaking processes.

Self-directed introspective elicitation visibilizes the junk turning over in one’s head. Once on a page, it is more available for inspection or introspection. In other words, moving (dumping) one’s thoughts to a blank page allows students to better analyze their own thinking process, showing the narratives and cognitive mechanisms through which structural arrangements are reproduced. These can help retrieve “the second part of the word ‘ethnography’, i.e. grapho, the self-reflexive narration of human experience within the context in which it unfolds” (Risi and Pronzato, 2022, p. 272).

Students report this as an enabling technique to build analytical materials. While here it is impossible to exemplify the process of how a participants’ self-directed analysis emerges or builds, three snippets are included to demonstrate some material from braindumps:

Nowadays without Facebook we are very incomplete person; you can live your life normally, but if you start to be present on this social network you MUST be present almost forever, otherwise you likely miss lots of things that for you are important. And with notifications you have a continuous reminder that tells you: “hey, I’m here, you have a life also here on me, OPEN ME!”, and you are going to obey to it EVERY SINGLE TIME. A little bit scaring in my opinion. (Participant C, unaltered, anonymized, used with permission via data donation)

I wanted to post a selfie. Now it’s 45 minutes later and I have finally finished it.

…I just hope that someone will like the picture. I see a celebrity who posted just before me, had 174,546 likes and 769 comments. Yeah. I doubt that I will get those. I mean–

…So now, i guess i can go on with my evening now, I have made myself. I make myself. I have posted a picture and now I have shared something that I have made. I have made myself. I have put myself out there. I have shown myself. Now i just hope that someone will see this and think of me and maybe –

…It‘s hard to explain what I wish I would get from this picture. I just want to make myself present. I just want to let people know that I'm there. That I exist. That i’m not just sitting here. (Participant D, transcribed from video and excerpted, details anonymized, used with permission via data donation)

I am shocked to find out how much time i waste (then again, not sure why i should be shocked). Really. Nothing I did in the entire hour I recorded myself actually […] accomplished anything. AT ALL. Seriously!? To think—I do this for hours a day.

…The final observation I made was how zombie-ish I look […] It is just a bit sad how glued I am to the computer. (Participant E, unaltered, anonymized, used with permission via data donation)

Building block 5: Writing ethnographic vignettes

The autoethnographic exercises result in a diverse and vast type of materials that are used to produce narrative style ethnographic accounts. Especially toward the end, to produce evocative pieces that connect relevant findings to larger patterns, they are asked to produce micro/snippet pieces to share or submit.

Analytical vignettes are short snippets depicting a scene, intended to evoke salient aspects or exemplify patterns in a situation. They are used here to elicit small units of lived experience on digital platforms that can be then added to larger ethnographic or analytical accounts, or used as the central part of an analysis. These are intended to evoke a feeling, depict the voice of the technological entity in a dialogue with the self or with another entity, detail a scene as in a screenplay, tell the biography of a micro-element of the situation, or describe a scene from a speculative future. At their best, they are rich, generative, and provocative, focused on depth and granularity rather than breadth. We are struck by the richness and poignancy of the vignettes that have been generated, as illustrated by these characteristic vignettes by participants F (quote below) and G (Figure 8 below):

I just had this fierce urge to snap a photo and share it. This view, it just popped out as I turned the corner. It was looking at me, wanting me to take a photo of it. But I don’t want to break my media fast. Phone off. No connection. No response. My thoughts feel stuck in a vacuum where nobody can reach them

…Media fasting has created an emotional vortex inside me today. As I am sitting on the cold floor in my living room, I am fighting with all these thoughts and feelings. Trying to discern what is real […] I removed the technologically- mediated share of my life for only one day and I feel like something is missing. What is it that I am missing? How can I say that I miss anything? Why would I say it? I can say that I miss a table and a chair because I am sitting on the floor while writing this […]. I am definitely missing something. The answers? I wonder to what extent devices possess and control me. If they control me, how do I reclaim the control? (Participant F, modified to preserve anonymity, used with permission via data donation)

In other versions of the experiment, we introduce and discuss specific critical theory concepts and prompt students to write small and speculative arguments, drawing on critical theory concepts if possible, or simply taking the perspective that something is wrong, and detailing how and why they think this is the case. We guide them to analyze how microscopic contexts are connected to larger infrastructures, the complex ecosystems in which online lived experiences are embedded. These essays are often prompted by questions of “In this moment, who benefits and who loses?” Sometimes, if a group of students expresses a noncritical orientation, a stronger question might be needed, such as “Given the agency and control dynamics in this situation, why can power never be equal?”

At this stage of the overall experiment, the development of analytical depth is crucial. By building granular descriptions and explanations of their own experience, students can learn useful things about themselves, how they relate with digital platforms and how their subjectivities are transformed into data for commercial and surveillance purposes. If the time and topic of the course permit, students can conduct further and more detailed qualitative coding at different levels (open, axial, and selective) on these materials.

These phases and building blocks, offered here as heuristic frameworks, are part of a scholarly activist approach to using critical pedagogy alongside autoethnographic tools to prompt critical consciousness and then using critical theory less as a direct topic and more as an underpinning framework to guide sensibilities and discussions. This perspective seeks to continue experimenting, to understand “How does critical literacy happen when it happens?” We reiterate that collecting data from students can be counterproductive to this endeavor, since as Markham (2022) has learned over many years of development of the framework, if the process is instrumentalized for data collection and comparative purposes, the fundamental goal of the enterprise is compromised, as the techniques, prompts, and facilitation will no longer be able to respond and adapt to the uniqueness of each situation.

Developing a new sensibility/critical consciousness

This last section shows how, in our experience, the framework works. We observe students moving beyond basic comprehension to analyze and then critique various affordances, interfaces, and ecologies influencing their digital media use. Notably, this is not about understanding machine learning and algorithms themselves, but by forensically analyzing the outcomes of these processes, identifying and situating plausible agents in larger ecosystems, and then situating their current and future selves in these same ecosystems, while considering issues of relative power, agency, and control.

Many students report they feel more critical because they now can see beneath the surface level of their attitudes and behaviors. This critical consciousness might be obvious, as when Participant H states directly:

I thought these reflections would be hard to write, but ended up being surprised at how hard it was to stop writing! I had so many thoughts and feelings that I wanted to share about my tracking experience. After paying so much attention to my media usage, I picked out a few things that really stood out for me, provoking me to seriously consider my media use and discuss it with others. I had no idea how much other entities are influencing my actions. Seriously no idea. (Participant H, unaltered)

Often, however, the reflections are more subtle, indicating some degree of increased awareness of the inordinate influence platforms exert over users:

By going from map to map I felt lighter, and more aware […], [the exercises] allowed me to understand what I really feel, what factors affect me. (Participant I, translated into English, unaltered)

This research also made me think a lot and convinced me to make a serious effort to be more aware of and less attached to social media from now on. (Participant J, translated into English, unaltered)

Many participants found that conducting these exercises helped them visualize the literal situatedness of their practices (e.g. the bed, while doing online class, etc.), the embodied intimacies (sleeping and waking to their devices and apps), how their bodies feel after hours spent using digital platforms. Defamiliarizing and then refamiliarizing through an allegorical map revealed new aspects of their relations with digital platforms. As participants J, K, and L note:

I am completely addicted to the evil red notification circle and I get a physical sensation every time I see one. I have to check, enter, and delete it. (Participant J, unaltered)

Mornings, especially just after waking up, should be a time for contemplation and relaxation, (instead I pick up the phone in my hand and scroll through social media). It has simply become an automated action difficult to control. […] [it] makes me angry. And it annoys me. (Participant K, translated into English)

When I pick up my phone, I automatically open these apps, I unconsciously open and look at Instagram stories, then I close the app and I don't even remember doing it. I realize that it's become so much a daily routine, that I've even lost track of the meaning of these apps. […] Everything has become automatic […], I have realized that my relationship with social media is […] dictated by boredom and habit. (Participant L, translated into English)

As for facilitating behavior changes based on deeper levels of critical consciousness, along the lines of what Freire would call political consciousness Freire, (1974), this occurs especially in the short term, but it is difficult to assess, because the research design deliberately does not measure behavior or attitude changes over time. Many participants say they have changed or will change their behaviors, either to regain control or to “break free” of the systems.

These deeper levels of critical consciousness seem much more common when the experiment leaves ample time for participants to reflect on their own interpretations. For example, early on in the experiment, participants often blame themselves for being in troubling situations, citing media panics of internet addiction or other user pathologies (e.g. addiction, narcissism, laziness, boredom, and habits), as discussed by Tiidenberg, et al., 2017):

Realizing how used I am to constantly playing something on Netflix kind of freaked me out. I even thought I might have a problem.

…As I’ve become more aware of my own digital media habits, I’ve also become more inclined to ask others about theirs. That made me realize that we all have our guilty pleasures and addictions. Whether it is Netflix or constant use of Snapchat, our lives are all somehow intertwined with the digital, whether we like it or not. (Participant M, unaltered)

Later, as participants spend time reflecting on these situations as occurring within larger systems of power and control, aided by the introduction of some critical theory concepts, they can more easily identify external locuses of control:

The more I think about it, the more I think that it’s not all me. Everything about the interface is designed to keep me there. I don’t even need to get off my ass any more, since Netflix kindly keeps playing the next episode. I won’t stop using Netflix completely, but I will be strategically reducing my usage. I want to regain some control. (Participant M, unaltered)

Some go even further to depict systems that are controlling, extractive, powerful, and irresistible, as these snippet excerpts from participants reaching the end of the experiments illustrate:

The algorithm locks you up in a cell, a cell made up of playlists ready to be consumed by you, and this is where your freedom is lost. You feel that you have all the music in the world in your hand, but it relieves you of the effort of searching for new music by immediately offering you playlists made just for you. (Participant N, translated into English)

It’s crazy how all my knowledge is on Google, it’s even annoying. Everything: news and photos are provided to us by algorithmic media. I find this very frustrating. It’s like a gate: if you want access to all the beautiful things behind it, you have to give me your data, you have to let us manipulate you. It’s not fair. (Participant O, translated into English)

We are trapped by these red dots. And they keep us coming back, and we think it is only to get rid of the damn notifications, but then there we are, scrolling and adding data to their platform and then getting more relevant things to pay attention to. (Participant P, content anonymized).

Conclusions

Across these experiments, our goal, as teachers and researchers, has remained to invite students to reflexively analyze their behavior in environments of increasing datafication and surveillance. The framework we use enables students to identify and unpack hegemonic power, by looking at how the interactions between the self and the platform are a form of “control through consent.” Then, through different iterations of similar exercises, this pedagogical model seeks to enable participants to unveil how the continuous repetition of both consent and control becomes habitual, and how hegemony is buried in seemingly mundane activities. This series seeks to break into “deep structures” of power, building on Markham’s (2020) premise that “to be critical in any effective or sustained way requires deep understanding of the contexts within which digitalization or datafication is occurring” (p. 229).

A core component of this pedagogical framework is that it focuses on small moments before moving to large-scale analysis. Observation through fieldwork or social science data collection is followed by close-level thematic analysis of patterns in action-responses, utterances, or encounters. Early exercises focus on the more micro or momentary levels of materiality and communicative actions. Later, the interpretative lens turns outward toward agential influences in the larger ecological sense, which involves more deliberate application of critical theory concepts, focused on what seems to be just “wrong.”

Criticality is understood as a precondition for civic engagement (Carmi et al., 2020; Hintz et al., 2022). The methodology presented here can be considered an Frankfurt School style operationalization of Pangrazio and Selwyn’s (2019, p. 426) framework, which emphasizes five domains of “personal data literacies” that “not only involves technical skills and understandings, but should also include conceptualisations of the inherently political nature of the broader data assemblage” and “aim to build awareness of the social, political, economic and cultural implications of data, as well as cultivating the metaphorical ‘space’ to reflect critically on these processes.”

This also aligns with and specifies the general trend in critical data or digital literacy scholarship, such as that advocated by Paris et al. (2022a, p. 86), in which people can take a critical informatics approach to learn to “account for and accommodate the interconnected, dynamic complexity of the many systems and levels of analysis at play,” so that “situated knowledge and related practices can be cultivated as insider power that can be exercised over technological development, which when exercised with care, can upend unequal power relations and promote more ethical socio-technical relationships” (Paris et al., 2022b, p. 3).

Our operationalization of Pangrazio and Selwyn’s framework involves building skills for identification and understanding by framing the student’s engagement as one of learning useful research methods for reaching more granular levels of analysis through defamiliarizing, reverse engineering, and thinking like a detective with a forensic attitude. We then facilitate reflexivity through prompts that have students reflect at direct and various meta levels about their experiences, asking themselves “Why did I do ‘this’ versus ‘that’?” or “How do I feel about x or y?” Once they reach a point of self-reflexivity and have conducted enough infrastructural, situational, and reflexive analysis to have a moment of critical consciousness, we push this forward by inviting them to delineate dynamics of power, agency, and control in specific situations, using these concepts as a lens to analyze their own situations, especially those that that may seem innocuous and meaningless but foster particular ways of being that privilege certain interests over others, and “to question how larger structures come into being and end up dominating them in ways that may not be fair” (Markham, 2019, p. 757). Finally, we help them find tactics to think otherwise by providing the scaffolding to produce snippet reflections, evocative vignettes, and even larger ethnographically informed accounts of lived experience, useful for conveying what they’re thinking to others.

Every tool used will have a different impact depending on a number of factors, not least different cultural premises and learning styles, topics we do not discuss here. The outcomes vary. But in multiple moments and ways, these techniques provide tools for young adults to dive deeper into the infrastructures of systems as well as their own existential being, as it relates and is being co-created with powerful technological and corporate agents. The sequence invites them to begin to critique how the hegemonic layers of sociotechnical assemblages affect the self in everyday lived experience. When enough self-centered revelations reach a certain threshold, participants may move beyond attitudinal shifts or use of proposed critical tactics in the moment, toward more sustained behavioral shifts. Sometimes, they talk about changing the behaviors of their family and friends. We find this last bit the most exciting as critical pedagogues. When they start teaching others the techniques they have learned, this is a good indicator of building critically oriented literacies as lifelong practices in and as communities of practice.

Figures

Figure 1.

Three tools to build analytical richness through attention on granular details

Figure 2.

Common practices used by participants to conduct the tracking exercise

Figure 3.

Basic question prompts used to guide a participant’s attention to the phenomenon at different levels of autoethnographic observation and reflection

Figure 4.

Showcasing different techniques for conducting situational analysis mapping

Figure 5.

Allegorical maps produced by Italian participants in 2021

Figure 6.

Examples of actual Victorian-era allegorical maps

Figure 7.

Allegorical map produced by Italian participant in 2021

Figure 8.

Narrative account alongside screenshot of Grindr platform interface by Participant G

Note

1. Sociologist and ethnomethodologist Harold Garfinkel (1967) would have his students perform these. He would famously encourage them to perform oddly in common environments to reveal the invisible social structures or norms guiding everyday life, such as standing in an elevator “backwards,” or arriving at their family home one evening and knocking on the door like a stranger.