Table 1.

A Summary of Papers in the Volume.

SectionAuthor(s)TitleResearch QuestionMethodsKey Findings/Contributions
Algorithmic organizing and data scientistsHopf, Joshi, Shollo, and StelmaszakData-based craft: How data scientists craft their data, models, and productsHow do data scientists iteratively shape “materials” (data), “tools” (algorithms), and “products” (analytics outputs) in a craft-like manner?65 in-depth interviews across 25 organizations; grounded theoryIntroduces “data-based craft,” showing how data scientists continuously rework data, refine algorithms, and deliver products that remain in flux – challenging static views of technical craft
 Valadão, Glaser, and HanniganAssembling Frankensteins: How data scientists stitch together provisional artifacts to generate novel insightsIn what ways do data scientists create and refine algorithms as evolving “assemblages” or “Frankensteins,” relying on protean tools and communal knowledge?Multi-method qualitative study (interviews, observation of industry events, archival data)Depicts how data scientists combine code snippets, libraries, and data sets in improvised ways, emphasizing the ad hoc, iterative nature of building algorithms, rather than a linear, finalized design process
Transforming organizations for algorithmic readinessPlesner and JustesenMaking organizations algorithm-ready: Algorithmic organizing through techno-organizational scriptsWhat redesigns and role shifts occur as an organization prepares for and integrates an algorithm, and how do “techno-organizational scripts” evolve in practice?Ethnographic fieldwork in a public debt collection centerShows that implementing an algorithm requires iterative “rescripting” of both the technology and existing roles, highlighting tensions between efficiency mandates and professional discretion
 WashingtonMachine-readable legitimacy: An ethnography of regulatory technologyHow do continuous data flows (RegTech) reshape financial compliance and the process of securing legitimacy from regulators?Longitudinal ethnography of industry associations and U.S. regulators (2013–2020)Develops “machine-readable legitimacy,” revealing that real-time data pipelines move compliance from a periodic, document-based event to an ongoing, algorithmic process negotiated among vendors, industry, and regulators
Opening up algorithmic encountersKostuj and Trittin-UlbrichMaking sense of glitches? Exploring cultural producers’ understandings of and interactions with the Instagram algorithmHow do cultural producers (journalists, artists, activists) interpret algorithmic “glitches,” and what do these moments reveal about platform power and user strategies?Online ethnography (240 hours), semi-structured interviews with cultural producersDemonstrates that perceived glitches (e.g., shadowbanning) prompt creators to adapt content, self-censor, and mobilize collective resistance, exposing hidden biases and power asymmetries in platform algorithms
 Sharma and AristidouHuman-AI coordination in extreme contexts: Overcoming trust and agency concernsHow can humanitarian “super-teams” integrate AI effectively in life-or-death crises while navigating trust (black-box fears) and agency (automation risks)?Archival data, multi-crisis comparative analysis (Gaza, Kenya, Nepal), process tracingIdentifies paradoxical tensions around trust and human agency; proposes “coordinative divergence” (broad inclusion) and “coordinative convergence” (unified goals) as twin practices that help AI teams function under urgent, high-stakes conditions
Advancing methodologies for the study of algorithmic organizingSpencer and KimInterpreting the inscrutable: Ethnographic approaches to studying the development of machine learning modelsHow can researchers use a “data work”-centered ethnography to uncover how ML models are shaped by local hierarchies, biases, and resource constraints?Comparative ethnography in two teaching hospitals (China and the Netherlands); observation of data collection, annotation, and recalibrationHighlights how organizational politics and iterative data labeling processes embed bias and tacit knowledge into ML models, offering a “data work” lens to demystify algorithmic inscrutability
 Timmer, Wrona, and ReineckeExploring algorithmic assemblages through multimodal inquiryWhy and how should researchers adopt multimodal methods (verbal, visual, embodied) to study algorithmic organizing and its sociomaterial “folding?”Proposal of a multimodal qualitative framework; illustration via an IT consultancy case; focus on non-verbal cues, code logs, and emotional dynamicsProvides a 10-principle methodology for capturing the often-hidden interactions (visual, embodied, emotional) that shape how algorithms are adopted, resisted, or transformed in organizational settings

or Create an Account

Close Modal
Close Modal