No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Algorithms at Work: The New Contested Terrain of Control
Abstract
The widespread implementation of algorithmic technologies in organizations prompts questions about how algorithms may reshape organizational control. We use Edwards’ (1979) perspective of “contested terrain,” wherein managers implement production technologies to maximize the value of labor and workers resist, to synthesize the interdisciplinary research on algorithms at work. We find that algorithmic control in the workplace operates through six main mechanisms, which we call the “6 Rs”—employers can use algorithms to direct workers by restricting and recommending, evaluate workers by recording and rating, and discipline workers by replacing and rewarding. We also discuss several key insights regarding algorithmic control. First, labor process theory helps to highlight potential problems with the largely positive view of algorithms at work. Second, the technical capabilities of algorithmic systems facilitate a form of rational control that is distinct from the technical and bureaucratic control used by employers for the past century. Third, employers’ use of algorithms is sparking the development of new algorithmic occupations. Finally, workers are individually and collectively resisting algorithmic control through a set of emerging tactics we call algoactivism. These insights sketch the contested terrain of algorithmic control and map critical areas for future research.
... To date, it is undeniable that the rise of such AI-enabled automation already transforms organizations' routines, especially when AI takes over tasks that were formerly performed by humans (e.g., Raisch & Krakowski, 2021). In this regard, existent discussions mainly deal with achievable cost savings and error reduction with AI-enabled automation (e.g., Kellogg et al., 2020), shifting humans to other 'higher-value' roles (e.g., Brynjolfsson & Mitchell, 2017), or emerging social challenges such as ethical AI (e.g., Rhue, 2019). Only recently, discussions began to stress the great importance of the reciprocal interplay between humans and intelligent machines for their coordination and its consequences within organizations (e.g., K. Leavitt et al., 2021;Murray et al., 2021;Rai et al., 2019;Schuetz & Venkatesh, 2020). ...
... Only recently, scholars began to stress the great importance of the reciprocal interplay between humans and AI (e.g., K. Leavitt et al., 2021;Murray et al., 2021;Rai et al., 2019;Schuetz & Venkatesh, 2020). While a handful of researchers have begun to examine how humans affect AI and vice versa and how organizations may coordinate this relationship, related research remains scarce and emphasizes the need for further analyses (e.g., Grønsund & Aanestad, 2020;Kellogg et al., 2020;Lindebaum et al., 2020;Lyytinen et al., 2021;Murray et al., 2021;Seidel et al., 2019;Sturm, Gerlach, et al., 2021). Especially, work on the impact of ML on organizational learning is still limited and mainly provide only basic insights into potential setups and hypothetical consequences (e.g., Ransbotham et al. (2020) propose possible learning modes with varying human-machine autonomy, Balasubramanian et al. (2022) theorize and simulate potential consequences of ML). ...
... ML systems are also very efficient learners as they can process large amounts of data very quickly and can thus make new knowledge rapidly available as soon as new data exists (e.g., Kellogg et al., 2020;Lindebaum et al., 2020). Yet, having such increased rationality, ML systems may also alleviate foolishness. ...
To achieve great performance and ensure their long-term survival, organizations must successfully act in and adapt to the reality that surrounds them, which requires organizations to learn effectively. For decades, organizations have relied exclusively on human learning for this purpose. With today’s rise of machine learning (ML) systems as a modern form of artificial intelligence (AI) and their ability to autonomously learn and act, ML systems can now also contribute to this vital process, offering organizations an alternative way to learn. Although organizations are increasingly adopting ML systems within a wide range of processes, we still know surprisingly little about how the learning of humans and ML systems affects each other and how their mutual learning affects organizational performance. Although a significant amount of research has addressed ML, existing research leaves it largely unclear whether and when humans and ML systems act as beneficial complementarities or as mutual impediments within the context of learning. This is problematic, as the (mis)use of ML systems may corrupt an organization’s central process of learning and thus impair the organizational adaptation that is crucial for organizational survival.
To help organizations facilitate useful synergies of humans and ML systems, this dissertation explores humans’ and ML systems’ idiosyncrasies and their bilateral interplay. As research on organizational learning has demonstrated, the key to managing such dynamics is the effective coordination of the ones who learn. The studies that were conducted for this dissertation therefore aim to uncover virtuous and vicious dynamics between humans and ML systems and how these dynamics can be managed to increase organizational performance. To take a holistic perspective, this dissertation explores three central levels of analysis.
The first level of analysis deals with performance impacts on the individual level. Here, the analysis focuses on two essential issues. First, the availability of ML systems as an alternative to humans requires organizations to rethink their problem delegation strategies. Organizations can benefit the most from the relative strengths of humans and ML systems if they are able to delegate problems to those whose expertise and capabilities best fit the problem. This requires organizations to develop an understanding of the problem characteristics that point to problems that are better (or less) suited to being solved by ML systems than by humans. Using a qualitative interview approach, the first study identifies central criteria and procedural artifacts and synthesizes these into a framework for identifying and evaluating problems in ML contexts. The framework provides a theoretical basis to help inform research about delegation decisions between humans and ML systems by unpacking problem nuances that decisively render problems suitable for ML systems. Building on these insights, a subsequent qualitative analysis explores how the dependency between a human and an ML system with respect to the delegated problem affects performance outcomes. The theoretical model that is proposed explains individual performance gains that result from ML systems’ use as a function of the fit between task, data, and technology characteristics. The model highlights how idiosyncrasies of an ML system can affect a human expert’s task execution performance when the expert bases her/his task execution on the ML system’s contributions. This study provides first empirical evidence on controllable levers for managing involved dependencies to increase individual performance.
The second level of analysis focuses on performance impacts on the group level. In contrast to traditional (non-ML) information systems, ML systems’ unique learning ability enables them to contribute independently to team endeavors, joining groups as active members that can affect group dynamics through their own contributions. Thus, in a third study, a digital trace analysis is conducted to explore the dynamics of a real-world case in which a group of human traders and a productively trading reinforcement ML system collaborate during trading. The studied case reveals that bilateral learning between multiple humans and an ML system can increase trading performance, which appears to be the result of an emerging virtuous cycle between the humans and the ML system. The findings demonstrate that the interactions between the humans and the ML system can lead to group performance that outperforms the individual trading of either the humans or the ML system. However, in order to achieve this, organizations must effectively coordinate the knowledge transfer and the roles of the involved humans and the ML system.
The third level of analysis focuses on performance impacts on the organization level. As ML systems increasingly contribute to organizational processes in all areas of the organization, changes in the organization’s fundamental concepts are likely to occur, and these may affect the organization’s overall performance. In a fourth study, a series of agent-based simulations are therefore used to explore the dynamics of organization-wide interactions between humans and ML systems. The results imply that ML systems can help stimulate the pursuit of innovative directions, liberating humans from exploring unorthodox ideas. The results also show that the alignment of human learning and ML is largely beneficial but can, under certain conditions, become detrimental to organizations. The findings emphasize that effective coordination of humans and ML systems that takes environmental conditions into account can determine the positive and negative impacts of ML systems on organization-level performance.
The analyses included in this dissertation highlight that it is precisely the unique differences between humans and ML systems that often seem to make them better complements than substitutes for one another. The secret to unleashing the true potential of ML systems may therefore lie in effectively coordinating the differences between humans and ML systems within their bilateral relationship to produce virtuous cycles of mutual improvement. This dissertation is a first step toward developing theory and guidance on coordinating the dynamics between humans and ML systems, with the aim of helping to rethink collaboration theory in the era of AI.
... Through this, Uber influences consumers' decisions and the drivers' work practices. Like Uber, many online labor platforms also adopt algorithms to monitor and control the platform workforce and optimize the efficiency of matching processes [1,2]. Relevant research has termed this new form, algorithm management [1,3,4]. ...
... Like Uber, many online labor platforms also adopt algorithms to monitor and control the platform workforce and optimize the efficiency of matching processes [1,2]. Relevant research has termed this new form, algorithm management [1,3,4]. Möhlmann et al. [2] refers to algorithm management as the, "large-scale collection and use of data on a platform to develop and improve learning algorithms that carry out coordination and control functions traditionally performed by managers" (p. ...
... While algorithm management can generate some benefits for platform companies, prior work suggests it could cause work environment tensions that may result in frustration or confusion among platform workers [1,3,5,6]. On the other hand, platform workers tend to think of the algorithms as highly opaque "black boxes" that interrupt their understanding of the inner workings [1,3,[7][8][9]. For example, algorithmic opacity may cause Uber drivers to experience uncertainties in reward compensations and ride assignments [2]. ...
While many online labor platforms have adopted algorithms to monitor or control workforces as a new form of algorithm management, there is no academic attempt to empirically examine how the algorithmic control of platforms influences platform workers’ behaviors in a platform context. In this study, we consider how algorithm management affects the platform workers’ response from a Digital Taylorism perspective. Digital Taylorism involves management’s use of technology to monitor workers by assigning and tracking work. Therefore, this study examines how algorithm control influences the platform workers’ response by mediating the tension of work compensation in an online labor platform context. Survey data collected from 216 food delivery riders in South Korea are used to test the model using partial least squares analysis. Our results show that algorithm control affects platform workers’ responses by mediating tensions of platform work compensation. Based upon our empirical findings, we can provide a theoretical perspective to relevant researchers who seek to find a theoretical mechanism of algorithm management. Moreover, we can offer practical insights to practitioners who are interested in algorithm management.
... Three years and a COVID-19 pandemic later, the world of work has evolved dramatically, as work from home became mandatory in many countries during confinements, and then was normalized in a range of jobs and workplaces where previously it was considered unfeasible. While the COVID-19 pandemic may be an opportunity to reinvent work organisation and make it more flexible in durable ways (Ollier-Malaterre, 2021), it has heightened the blurring of the boundaries (Allen, Merlo, Lawrence, Slutsky, & Gray, 2021) and accelerated the trend towards the quantification and digitalisation of organisational control (Kellogg, Valentine, & Christin, 2020). ...
... The premise of such quantified monitoring is that people analytics can make management more efficient than human decisions, which are cast as subjective and biased (Zuboff, 2019). Tireless and predictable algorithms become a "fetish" and a magical solution (Burrell & Fourcade, 2021) to orchestrate to the "6 Rs" of management, i.e. restrict, recommend, record, rate, replace and reward (Kellogg et al., 2020). Such quantification entails serious ethical issues and risks for workers, managers, and organisations (Giermindl, Strich, Christ, Leicht-Deobald, & Redzepi, 2021;Tursunbayeva, Pagliari, Di Lauro, & Antonelli, 2021). ...
... However, people may regulate their privacy in many ways: they may refrain from entering personal information online when it is not mandatory, avoid using universal logins such as their Google or Facebook logins and avoid storing personal photos on clouds. Moreover, some workers become "algoactivists" (Kellogg et al., 2020): they alter monitoring equipment and manipulate algorithms (Martin et al., 2016). Recently, two DoorDash drivers figured out that turning down the lowest-paying deliveries raises pay rates; they started the #DeclineNow forum to encourage others drivers to trick the algorithm (Akhtar, 2021). ...
The COVID-19 pandemic has revealed and accelerated two trends that are now fully part of the “new normal” of work. First, the erosion of boundaries between work and life has become very salient with the normalization of work from home. Second, the quantification of organizational control, which was already present in monitoring devices and algorithmic management, has reached news levels with electronic monitoring of employees through “bossware” and Internet of Behaviours devices. This essay chapter analyses these trends and argues that active regulation of technology and its implications at work and outside of work is now an integral part of work for workers in many occupations. Specifically, the new normal of work routinely includes devising and adapting rules and behaviours around three major challenges: (a) constant connectivity (when and where workers are connected and available to work); (b) self-presentation (disclosures on video conferences, social media, and other online spaces); and (c) privacy (protecting personal information despite monitoring software, trackers, and algorithmic work). Colliding worlds and quantified algorithmic control are deep-rooted trends that must be addressed by workers, employers, unions, public policy makers, and scholars, if we are to build a new normal sustainable workplace.
... Commonly referred to as 'algorithmic management', such HR practices automate or semiautomate managerial decisions related to working conditions and workers' control. To achieve this, algorithmic management utilises AI and feeds it with big data (Adams-Prassl, 2019; De Stefano, 2019;Kellogg et al., 2020;Mateescu and Nguyen, 2019) collected from numerous sources, such as workers' CVs, keyboard and mouse movements, call logs, screenshots, webcams, application logging activities, wearable devices, and wellness programmes. AI-empowered HR decision support tools pose some direct risks that affect large parts of the workforce, regardless of occupation and sector of employment and skill levels. ...
... Regarding surveillance, the literature suggests that algorithmic management tools are designed to increase employers' capabilities to monitor and control workers (Crawford et al., 2019;De Stefano, 2019;EPSC, 2018;Kellogg et al., 2020;Servoz, 2019). This raises two issues. ...
... The first is the blurring of boundaries between work and private life, with serious implications for workers' right to privacy (Adams-Prassl, 2019; Moore, 2019a;Servoz, 2019). The second issue is related to the diminishing possibilities of workers' voice (Kellogg et al., 2020;Moore, 2019a) because algorithmic management tools render 'the sites of everyday resistance facilitated by worker-to-worker communication penetrable by management' (Moore, 2019a: 126). Employers use new worker monitoring tools to measure previously unmeasurable aspects of work, such as attitudes, tiredness, mental well-being and stress (Moore, 2019a). ...
This article discusses the risks that artificial intelligence (AI) poses for work. It classifies risks into two types, direct and indirect. Direct risks are AI-induced forms of discrimination, surveillance and information asymmetries at work. Indirect risks are enhanced workplace automation and the increasing ‘fissurisation’ of work. Direct and indirect risks are illustrated using the example of the transport and logistics sector. We discuss policy responses to both types of risk in the context of the German economy and argue that the policy solutions need to differ according to the type of risk. Direct risks can be addressed by European and national regulation against discrimination, surveillance and information asymmetries. As for indirect risks, the first step is to monitor the risks so as to gain an understanding of sector-specific transformations and establish relevant expertise and competence. This way of addressing AI-induced risks at work will help to improve the prospects of decent work, fair remuneration and adequate social protection for all.
... Companies today leverage digital technologies and increasingly intelligent algorithms in various forms to direct, evaluate and discipline employees in their everyday work. This phenomenon, known as algorithmic control (AC), can be mainly observed in platform organizations such as Uber, Instacart, Amazon Mechanical Turk, and increasingly also in traditional organizations such as Amazon Warehouses (Kellogg et al., 2020). There is reason to expect this type of control to become more widespread in the future (Anthes, 2017). ...
... To understand this new form of control, researchers have drawn on the organizational control literature and related theories (Wiener et al., 2021). However, recent research suggests that AC has several unique characteristics that distinguish it from traditional control forms Kellogg et al., 2020;Pregenzer et al., 2020;Veen et al., 2019), questioning the 'simple' transferability of prior research. For example, compared to a human controller, algorithms can process vast amounts of data in real time and make omnipresent control attempts at any time. ...
... By incorporating these unique characteristics of AC, several novel AC frameworks have emerged. These consider different dimensions of AC, such as the target of control, the scope of control, or the forms of control (Kellogg et al., 2020;Möhlmann et al., 2021;Parent-Rocheleau & Parker, 2021). However, the existing AC framework landscape now faces two key challenges. ...
Algorithmic control (AC) refers to organizations’ use of increasingly intelligent algorithms and related digital technology to steer worker behavior. While previous studies have identified and conceptualized various forms of AC in both platform and traditional work contexts, the presented conceptualizations lack measurability. This key shortcoming hampers further empirical research on the current use of AC and its manifold consequences. In this study, we report on the item development process for a scale for measuring perceived AC from a workers’ perspective. Following well-established approaches, an initial item pool was developed. The items were discussed and refined with the support of five academic experts and three AC workers. A subsequent rating study with 98 workers was conducted to ensure the content validity of all items. On this basis, the study at hand presents a comprehensive set of items for both AC in general and its seven sub-dimensions.
... The term "algorithmic management" was coined to express how algorithms assume managing roles in organizations and enable organizations to essentially control a large and dispersed workforce [20]. In their review of algorithmic management, Kellogg et al. [21] highlight six mechanisms through which algorithms assert control over workers: directing workers by restricting and recommending, evaluating workers by recording and rating, and disciplining workers by rewarding and replacing. Further evidence illustrates that the unique features of algorithmic management such as persistent surveillance [22], continuous performance assessment [23], automated decision-making [12], little human-to-human contact [24], and poor transparency of algorithmic decisions [25] facilitate important power asymmetries between workers and management [26]. ...
... 30], algorithmically managed workplaces are examined from a perspective of conflict between workers and management-implemented-algorithms [2,31,32]. Nevertheless, the addition of instantaneous, opaque, interactive and comprehensive algorithmic control mechanisms into the management process [21] in turn elicits more varying degrees of worker resistance. As a result, the concept of individual and collective resistance to such algorithmic control -"algoactivism" -has emerged [13]. ...
... Interviews were audio recorded, automatically transcribed and transcripts were manually checked for accuracy. Interview guide drawing on Kellogg et al.'s [21] conceptualization of algorithmic management practices was used. Data were thematically analyzed by one researcher, following a structured approach which systematically moved from open coding to axial and theoretical coding. ...
In many countries, on-demand food delivery platforms (e.g. Deliveroo, Wolt, Uber Eats) have become an inseparable part of the hospitality and tourism ecosystem. A key area of interest in technology research has been how platforms algorithmically manage the interaction between task requesters (e.g. customers, tourists) and task fulfillers (e.g. restaurants and delivery couriers). However, there is a lack of research on how such algorithmic management practices impact workers and what strategies workers adopt to counteract the algorithm. To that end, this qualitative study explores forms of expressing algoactivism in the context of on-demand food delivery platforms by conducting interviews with delivery couriers (n = 5) and restaurant workers and managers (n = 7). It is found that both couriers and hospitality employees adopt specific behaviors to optimize and game the platforms’ algorithms, and that some algorithmic management practices are perceived more negatively than others. Implications for e-tourism management and research are discussed.
... Other lesser known theories have also explained the research on human-machine collaboration. These include role theory (Tang et al., 2021), conservation of resources theory , behavioral decision theory (Lawler and Elliot, 1996), theory of planned behavior (Marler et al., 2009), labor process theory (Charlwood and Guenole, 2022;Kellogg et al., 2020;Fleming, 2019), signaling theory (Mccoll and Michelotti, 2019;Mirowska and Mesnet, 2022) and social cognitive theory (Mancha and Shankaranarayanan, 2020;Suseno et al., 2022). ...
... On employee assessment, the organization has begun to deploy fingerprint check-in, office video surveillance and other measures. The use of digital devices helps managers to better understand employee work status (Kellogg et al., 2020). The application of such smart devices can help managers to monitor employees in real time, even though some employees may not wish it (Bhave et al., 2020). ...
Purpose
This study aims to systematically map the state of work on human–machine collaboration in organizations using bibliometric analysis.
Design/methodology/approach
The authors used a systematic literature review to survey 111 articles on human–machine collaboration published in leading journals to categorize the theories used and to construct a framework of human–machine collaboration in organizations. A bibliometric analysis is applied to statistically evaluate the published materials and measure the influence of the publications using co-citation, coupling and keyword analyses.
Findings
The results inform that the research on human–machine collaboration in the organizational field is targeted at four aspects: performance, innovation, human resource management and information technology (IT).
Originality/value
This work is the first exploratory piece to assess the extent and depth of research on human–machine collaboration.
... Worker strategies regarding algorithmic management The literature has already described how platform business models have created new organisational patterns based on 'algorithmic management' (Lee et al., 2015;Zuboff, 2019;Kellogg et al., 2020). Meaning that algorithmic systems replace some organisational functions traditionally performed by managers and labour relations face new managerial frameworks where algorithms play a key role. ...
... Meaning that algorithmic systems replace some organisational functions traditionally performed by managers and labour relations face new managerial frameworks where algorithms play a key role. One of the most distinctive features of this new form of labour force management is its application on a mass scale, mediated through automated and digitalised processes which enable labour platforms to direct, evaluate and exercise disciplinary power over large numbers of platform workers (Kellogg et al., 2020;Wood, 2021). ...
Workers and organised labour are being challenged by the increasing expansion of digital labour platforms in most countries worldwide. Such digital platforms and their algorithms create controversial forms of work relationships and undermine traditional labour organisation, leading to extensive public and scientific debate. […] Through our empirical fieldwork and cross-country analysis, we hope to bring new insights and uncover new ground in this emerging field of study
Download here: https://novaresearch.unl.pt/en/publications/digital-labour-platforms-representing-workers-in-europe
... In traditional (physical) work settings such as large warehouses, using robots to deliver items to human packers with machine-driven efficiency increases workload and job repetitiveness. 'Algorithmic management' (Kellogg et al., 2020) and task fragmentation has similar impacts on platform micro-workers and gig workers (Cedefop, 2021d). ...
... Apart from directly reducing job satisfaction, workers may also become less satisfied with their job because of increasing use of algorithmic management and employee surveillance(Kellogg et al., 2020;Baiocco et al. ...
... Marking a fundamental shift in workforce management, companies are turning to intelligent algorithmic tools to direct, evaluate, and discipline employees [1]. Although management by algorithm is providing efficient tools for controlling dispersed workforce, the consequences of new algorithmic management strategies for hospitality workers remain a disputed subject. ...
... Although management by algorithm is providing efficient tools for controlling dispersed workforce, the consequences of new algorithmic management strategies for hospitality workers remain a disputed subject. Some positive outcomes of algorithmic work have been evidenced, such as perceived greater work flexibility [2], positive affective experience due to gamification elements [1] and enhanced organizational learning. However, not all socioeconomic categories reap the same benefits [3] -an indication of gendered and racial discrimination. ...
Employee turnover has been one of the main concerns facing the hospitality industry. This issue seems to be aggravated in artificial intelligence (AI) environment, where AI implementation is associated with pressure, job alienation, and labor replacement, increasing workers’ desire to quit their job. To analyze the relationship between AI awareness, job alienation, discrimination, and turnover intention, an online survey was distributed to hospitality employees ( n = 450). From a series of independent-samples T-tests and regression analyses, this study found employees’ turnover intentions are significantly associated with employees’ concerns of being replaced by AI, perception of job alienation, and workplace discrimination. Importantly, current algorithmically managed workers tend to feel more powerless and discriminated against, and thus have higher turnover intentions. Recommendations for practice and future research are provided.
... Essentially algorithmic systems helps managers to take decisions in overabundance of data and information (Jarrahi and Sutherland,2019). Although it has been used to reduce the burden of work and analysis for managers, it has been also used by the companies to administer and manage the human resources (Kellogg et al. 2020). ...
... Despite the recent increase in the use of terms like sustainable HR and sustainable employment, suggesting that PCs are generally fulfilled, several authors have stated that this is not the case for large groups of employees working in precarious employment contexts such as menial jobs (e.g., Bal and Brookes, 2022;Griep et al., 2019). This is evidenced by numerous recent press articles that have discussed problematic employment conditions in supermarkets (Harounyan, 2017), logistics (Kellogg et al., 2020), the travel industry (Hansen and Vugts, 2022), and the hospitality industry (Bresson, 2021). The consequences of these problems are manifold. ...
This qualitative research explores the psychological contract (PC) of a sample of self-initiated expatriates (SIEs) working in the French hospitality sector, focusing on PC evaluation as well as reactions to PC breach and feelings of violation. The authors found evidence of a psychological contract type not discussed before in empirical studies. The employer in this research intentionally disrupts the exchange relationship, creating a destructive PC. In these cases, it is assumed that employees would exit such an employment relationship, but instead the study found a mix of dysfunctional behavior in the form of neglect, workplace deviance and revenge cognitions. Accounting for the limitations of the study the authors highlight the implications of the findings for theory, practice and future research.
... A gestão mediada por algoritmos estrutura seu controle sobre os processos de trabalho por meio de sua capacidade de mapear, subdividir e monitorar digitalmente atividades simples e complexas (KELLOG et al., 2020;DANAHER, 2016). Em outras palavras, significa dizer que este formato de gestão limita a interação humana, provocando mudanças na dinâmica laboral (KELLOGG et al., 2020;KITCHIN;DODGE, 2011). E nesta lógica, as "novas" formas de controle sobre o trabalho são fundamentadas em um distanciamento entre os funcionários, mas sujeitam os trabalhadores a um "novo" controle, mais acentuado e mais centralizado (VIDIGAL; KROST, 2020), no qual os algoritmos são projetados para controlar e intensificar os ritmos, tempos e movimentos da força de trabalho (ANTUNES, 2020). ...
O uso dos algoritmos é cada vez mais comum nas organizações. Esta pesquisa teve como objetivo apresentar um panorama da produção científica crítica, da base de dados Web of Science, em relação ao gerenciamento e controle algorítmico do trabalho, por meio de uma pesquisa bibliométrica quantitativa. No tocante a Lei de Lotka, verificamos a autora Min Lyung Lee, com 5 artigos, representando 9,43% do total do corpus, sobre a Lei de Bradford, constatamos que os periódicos que têm publicado sobre o tema são diversificados e com relação a Lei de Zipf, observamos três palavras com maior ocorrência, algorithmic management, labor e gig economy.
... Translation between machine outputs, users, and clients might require new types of work and roles of doing such interpretations: 'algorithmic brokerage' work (Kellogg et al., 2019). For example, intelligence officers brokered between data scientists and police officers in implementing a predictive policing program (Waardenburg et al., 2022). ...
Professions continue to be the primary means through which societies institutionalize expertise. Recent analyses and narratives predict that artificial intelligence (AI) will make meaningful inroads into non-routine reasoning about complex cases, threatening the authority of professions. These predictions, we argue, draw on substantialist understandings of expertise as an intellectual possession, a mental achievement, or a cognitive state performed-by humans or machines-to achieve effects. A synthesis of empirical studies shows that expertise is more accurately conceptualized as relationally constituted-generated, applied, and recognized-through interactions. Relational expertise creates challenges of opacity, translation, and accountability for the development and deployment of AI technologies in the context of professional work. A relational understanding of expertise disrupts notions that professions may be augmented with, subordinated to, or dismantled by AI technologies. Instead, AI technologies are embedded in the network of interactions through which the relational expertise of professions is constituted.
... Building on Zuboff (2019) Schwarz here offers an original account of the multifaceted consequences of digital technologies on work and labour, by carefully disentangling the two notions. What is missing in this (already very rich) chapter is probably a focus on work automation and the algorithmic control of workers (Delfanti 2021;Kellogg et al. 2020), which are important aspects of the "increased governability" (p. 123) characterizing digital societies. ...
Sociological theory is often perceived as the semi-obsolete heritage of 19th and 20th century thinkers: good enough to make sense of power, social structure and face-to-face interactions, but substantially inadequate to interpret the now overwhelming technological mediation of social life. Perhaps for this reason, the social sciences see a proliferation of midrange theories of “the digital” following the hype around the technological trend of the moment—e.g., AI, crypto, blockchain, metaverse. “New technologies reshape society; therefore, brand new concepts and theorizations are needed to make sense of it” appears to be the doxa guiding recent scholarship. Yet, is this always true? Does a digital society necessarily require “digitally native” social theories?
... In terms of the theoretical perspective, the experts of our panel and members of the breakout session expressed the expectation that the field will continue to study aspects related to organizational control, one of the most fundamental topics of organizational scholarship (Kellogg et al., 2020, Wiener et al., 2021, but also will start to explore more adjacent areas scrutinizing how algorithmic management may influence workers' autonomy, resistance, and voice (Gal et al., 2020, Meijerink & Bondarouk, 2023, Möhlmann et al., 2022, Tarafdar et al., 2022. Here, the workshop participants stressed the need to examine both the desirable and undesirable outcomes of algorithmic management to workers and go beyond the deterministic assumptions on the effects of algorithmic management. ...
In recent years, the topic of algorithmic management has received increasing attention in information systems (IS) research and beyond. As both emerging platform businesses and established companies rely on artificial intelligence and sophisticated software to automate tasks previously done by managers, important organizational, social, and ethical questions emerge. However, a cross-disciplinary approach to algorithmic management that brings together IS perspectives with other (sub-)disciplines such as macro-and micro-organizational behavior, business ethics, and digital sociology is missing, despite its usefulness for IS research. This article engages in cross-disciplinary agenda setting through an in-depth report of a professional development workshop (PDW) entitled "Algorithmic Management: Toward a Cross-Disciplinary Research Agenda" delivered at the 2021 Academy of Management Annual Meeting. Three leading experts (Mareike Möhlmann, Lindsey Cameron, and Laura Lamers) on the topic provide their insights on the current status of algorithmic management research, how their work contributes to this area, where the field is heading in the future, and what important questions should be answered going forward. These accounts are followed up by insights from the breakout group discussions at the PDW that provided further input. Overall, the experts and workshop participants highlighted that future research should examine both the desirable and undesirable outcomes of algorithmic management and should not shy away from posing ethical and normative questions.
... In Australia, the Productivity Commission (2020) pointed out that in workplaces with high job demands and low employee controls, a gap between effort and reward and a low level of organisational Our results extend these benefits to mental health, as these technologies can help monitor anxiety. However, the use of wearable or emotion-recognition technologies is also extremely controversial given that these technologies can be used as new tools for surveillance and authoritarianism (Doberstein et al., 2022;Kellogg et al., 2020) and may represent new forms of dominance. Moreover, their use in the workplace raises important privacy and data protection issues (Doberstein et al., 2022;Eurofound, 2022;McStay, 2020) In the case of the neurodiversity initiatives analysed in this paper, participants pointed out that VR is used for training to help workers cope with situations that generate anxiety and favour their inclusion. ...
The transformation of the intelligence ecosystem associated with the digital transformation represents a critical juncture for diversity and inclusion (D&I). We present a multidisciplinary perspective on digital transformation and D&I that demonstrates that, in the context of automated decision making, where algorithmic biases and the standardisation of thought represent new risks, neurodiversity initiatives become a cornerstone for advancing D&I. Based on interviews with neurodiversity experts, we identify innovative ways to efficiently configure an inclusive organisational design targeting neurodiversity by leveraging technologies. We identify several properties of technologies that support D&I in neurodiversity initiatives: the neutralisation of biases during interviews, the development of digital support for physical and mental well‐being and the facilitation of different cognition modes. Finally, we critically discuss the risks and opportunities offered by various technologies in terms of performance evaluation, new forms of dominance, and design of a digital ecosystem for mental well‐being.
... However, the outcome of algorithmbased HR decision-making is usually quite ambiguous. It is hard to tell whether applying the algorithms to evaluate employees' performance yields better or worse outcomes for employees (Kellogg et al., 2020). Also, consumers normally have limited information on the consequences of algorithm-based HR decision-making adoption, further stoking outcome ambiguity. ...
The growing uses of algorithm-based decision-making in human resources management have drawn considerable attention from different stakeholders. While prior literature mainly focused on stakeholders directly related to HR decisions (e.g., employees), this paper pertained to a third-party observer perspective and investigated how consumers would respond to companies’ adoption of algorithm-based HR decision-making. Through five experimental studies, we showed that the adoption of algorithm-based (vs. human-based) HR decision-making could induce consumers’ unfavorable ethicality inferences of the company (study 1); because implementing a calculative and data-driven approach (i.e. algorithm-based) to make employee-related decisions violates the deontological principles of respectful employee treatment (study 2). However, this effect was attenuated when consumers had high (vs. low) power distance beliefs (study 3); the algorithm served as assistance (vs. replacement) for human decisions (study 4); or the adoption was framed as employee-oriented (vs. company-oriented) motivated (study 5). Our findings suggested that consumers are aversive to algorithm-based HR decision-making because it is deontologically problematic regardless of its decision quality (i.e. accuracy). This paper contributes to the extant understanding of stakeholders’ responses to algorithm-based HR decision-making and consumers’ attitudes toward algorithm users.
... The latter condition is therefore crucial for deciding whether and how IAS providers and operators are to blame. To be held morally responsible, a firm's employees need to have the necessary autonomy and control over actions done in the name of the firm in relation to an IAS (Kellogg et al. 2019). When it comes to using and designing information systems, there are established strategies that allow IAS providers and IAS operators to collectively divert and reject responsibility ascriptions (French 1984). ...
Intelligence Augmentation Systems (IAS) allow for more efficient and effective corporate processes by means of an explicit collaboration between artificial intelligence and human judgment. However, the higher degree of system autonomy, along with the enrichment of human capabilities, amplifies pre-existing issues of the distribution of moral responsibility: If an IAS has caused harm, firms who have operated the system might argue that they lack control over its actions, whereas firms who have developed the system might argue that they lack control over its actual use. Both parties rejecting responsibility and attributing it to the autonomous nature of the system leads to a variety of technologically induced responsibility gaps. Given the wide-ranging capabilities and applications of IAS, such responsibility gaps warrant a theoretical grounding in an ethical theory, also because the clear distribution of moral responsibility is an essential first step to govern explicit morality in a firm using structures such as accountability mechanisms. As part of this paper, first the necessary conditions for the distribution of responsibility for IAS are detailed. Second, the paper develops an ethical theory of Reason-Responsiveness for Intelligence Augmentation Systems (RRIAS) that allows for the distribution of responsibility at the organizational level between operators and providers. RRIAS provides important guidance for firms to understand who should be held responsible for developing suitable corporate practices for the development and usage of IAS.
... lavorativi contemporanei, dove sono stati utilizzati per la produzione, la catalogazione e l'elaborazione di grandi quantità di dati. Tale processo è stato promosso il più delle volte dal management con il tentativo di automatizzare parte dei processi produttivi e/o il controllo manageriale compiuto su quest'ultimi, rendendolo più capillare ed espandendolo a sfere della vita un tempo totalmente private ed esterne all'ambito lavorativo (Kellogg 2020). A caratterizzare le tecnologie algoritmiche di ultima logie digitali in grado di prendere decisioni e risolvere problematiche autonomamente, imitando e/o superando le modalità di ragionamento umano (Misselhorn 2018). ...
L'introduzione nei luoghi di lavoro di tecnologie digitali miniaturizzate è stata posta al centro di un ampio dibattito interdisciplinare che ha coinvolto diverse branche di studio. Il presente articolo si oc-e vengono adoperate nelle organizzazioni contemporanee. A questo scopo l'articolo presenta e discute alcuni dei risultati emergenti da una ricerca empirica svolta in due organizzazioni lavorative del nord Italia. Si è visto come l'algoattivismo dei lavoratori consista innanzitutto nell'appropriarsi della tec-nologia, adattando la tecnologia algoritmica al proprio background di credenze, alle rappresentazioni elaborate riguardo la tecnologia nel corso del suo utilizzo, nonché alle possibilità e costrizioni dettate dalla propria quotidianità. L'algoattivismo dei lavoratori consiste anche nel dare forma a processi so-cio-organizzativi più ampi atti a monitorare e a gestire il loro stato di salute (anche) grazie al supporto tecnologico. Il tessuto organizzativo emergente è stato, quindi, caratterizzato da una continua connes-sione e sovrapposizione tra gestione della salute, attività lavorativa, obblighi familiari e vita mondana.
... So actually, we need to argue with the algorithm and our arguing with the algorithm means understanding algorithms. (Labour Organisation: Director) The interviewee added that these algorithms, coded by the employers, needed to be better understood by the trade union movement (see Lenaerts et al. 2018;Shapiro 2018;Wood et al. 2019;Tassinari and Maccarrone 2020;Duggan et al. 2020;Kellogg et al. 2020 ...
Based on rich first hand empirical material, the chapter discusses the implications of the gig economy for workers and trade unions in the UK context.
... Both these high degree of separation designs, as presented in Table 3, are problematic from an organizational performance perspective. The first, algorithmic management, reduces employee control and thereby the possibility to improve the work (Kellogg et al. 2020). In algorithmic management, one can find AI solutions that perform, for instance, task allocation and scheduling (Schildt 2017), which we understand as operational regulation activities. ...
In this paper, we contribute to research on enterprise artificial intelligence (AI), specifically to organizations improving the customer experiences and their internal processes through using the type of AI called machine learning (ML). Many organizations are struggling to get enough value from their AI efforts, and part of this is related to the area of explainability. The need for explainability is especially high in what is called black-box ML models, where decisions are made without anyone
understanding how an AI reached a particular decision. This opaqueness creates a user need for explanations. Therefore, researchers and designers create diferent versions of so-called eXplainable AI (XAI). However, the demands for XAI can
reduce the accuracy of the predictions the AI makes, which can reduce the perceived usefulness of the AI solution, which, in turn, reduces the interest in designing the organizational task structure to beneft from the AI solution. Therefore, it is important to ensure that the need for XAI is as low as possible. In this paper, we demonstrate how to achieve this by optimizing the task structure according to sociotechnical systems design principles. Our theoretical contribution is to the under explored field of the intersection of AI design and organizational design. We find that explainability goals can be divided into two groups, pattern goals and experience goals, and that this division is helpful when defining the design process and the task structure that the AI solution will be used in. Our practical contribution is for AI designers who include organizational designers in their teams, and for organizational designers who answer that challenge.
... This may lead to cases where a single worker engaging in poor behavior, or effort, can lead to the deterioration of cooperation across the entire team. We therefore hypothesize that platforms must adopt some form of real-time accountability and dispute resolution aided by algorithmic governance [36]. ...
Existing literature and studies predominantly focus on how crowdsource workers individually complete tasks and projects. Our study examines crowdsource workers' willingness to work collaboratively. We report results from a survey of 122 workers on a leading online labor platform (Upwork) to examine crowd workers' behavioral preferences for collaboration and explore several antecedents of cooperative behaviors. We then test if actual cooperative behavior matches with workers' behavioral preferences through an incentivized social dilemma experiment. We find that respondents cooperate at a higher rate (85%) than reported in previous comparable studies (between 50-75%). This high rate of cooperation is likely explained by an ingroup bias. Using a sequential mediation model we demonstrate the importance of a sense of shared expectations and accountability for cooperation. We contribute to a better understanding of the potential for collaborative work in crowdsourcing by accessing if and what social factors and collective culture exist among crowd workers. We discuss the implications of our results for platform designers by highlighting the importance of platform features that promote shared expectations and improve accountability. Overall, contrary to existing literature and predictions, our results suggest that crowd workers display traits that are more consistent with belonging to a coherent group with a shared collective culture, rather than being anonymous actors in a transaction-based market.
... Indeed, as these and similar kinds of computation-based social sorting and management infrastructures continue to multiply, they promise to jeopardise more and more of the formative modes of open interpersonal communication that have enabled the development of crucial relations of mutual trust and responsibility among interacting individuals in modern democratic societies. This is beginning to manifest in the widespread deployment of algorithmic labour and productivity management technologies, where manager-worker and worker-worker relations of reciprocal accountability and interpersonal recognition are being displaced by depersonalising mechanisms of automated assessment, continuous digital surveillance and computation-based behavioural incentivisation, discipline, and control (Ajunwa et al., 2017;Akhtar & Moore, 2016;Kellogg et al., 2020;Moore, 2019). The convergence of the unremitting sensor-based tracking and monitoring of workers' movements, affects, word choices, facial expressions, and other biometric cues, with algorithmic models that purport to detect and correct defective moods, emotions, and levels of psychological engagement and well-being, may not simply violate a worker's sense of bodily, emotional, and mental integrity by rendering their inner life legible and available for managerial intervention as well as productivity optimisation (Ball, 2009). ...
This chapter is concerned with setting up practical guardrails within the research activities and environments of Computational Social Science (CSS). It aims to provide CSS scholars, as well as policymakers and other stakeholders who apply CSS methods, with the critical and constructive means needed to ensure that their practices are ethical, trustworthy, and responsible. It begins by providing a taxonomy of the ethical challenges faced by researchers in the field of CSS. These are challenges related to (1) the treatment of research subjects, (2) the impacts of CSS research on affected individuals and communities, (3) the quality of CSS research and to its epistemological status, (4) research integrity, and (5) research equity. Taking these challenges as motivation for cultural transformation, it then argues for the incorporation of end-to-end habits of Responsible Research and Innovation (RRI) into CSS practices, focusing on the role that contextual considerations, anticipatory reflection, impact assessment, public engagement, and justifiable and well-documented action should play across the research lifecycle. In proposing the inclusion of habits of RRI in CSS practices, the chapter lays out several practical steps needed for ethical, trustworthy, and responsible CSS research activities. These include stakeholder engagement processes, research impact assessments, data lifecycle documentation, bias self-assessments, and transparent research reporting protocols.
... We know from the existing literature that digital platforms are not conventional organisations (Gawer, 2014) but rather multiactor, supraorganizational environments that facilitate the interaction of unrelated and constantly changing sets of participants who act in their own interests (e.g., Möhlmann et al., 2021). This environment invites the dynamic reshaping of control as a result of digital technology-enabled social interactions on platforms (Kellogg et al., 2020). For example, the distinctions between formal and informal means of control can be blurred when platform participants engage through digital media and when the formal controls that platform operators inscribe in the technology are reformulated, interpreted, and re-enforced by informal means on the platform (Alaimo & Kallinikos, 2021;Orlikowski & Scott, 2014;Plantin et al., 2018). ...
Digital platforms are supraorganizational entities that use digital technology to facilitate interactions between diverse actors, leading to novel forms of organisation and accompanying forms of control. The current Information Systems (IS) literature, however, struggles to describe control on digital platforms in a way that does justice to the dynamic character of the phenomenon. Taking this as an opportunity, we follow the enactment of control over time and across parties in a hybrid ethnographic study of the social commerce platform Poshmark. Specifically, we conceptualise the dynamics of control as changes in the means of control—formal or informal—and the sources of control—operator or participants—over time. Tracking these conceptual dimensions, we identify the distinct ways control has changed on Poshmark. Synthesising these findings into four dynamics of control, we show that control on digital platforms is rarely static due to aggregate effects arising from the operator and from participant interactions with each other through the digital features deployed on the platform. Based on these insights, our study contributes to the IS literature on control by broadening the conception of control on digital platforms. The theoretical and practical insights generated in this paper thereby lay the foundation for the systematic study of the dynamics of control that are unique to platform environments.
The Artificial Intelligence (AI) revolution has arrived, as AI systems are increasingly being integrated across organizational functions into the work lives of employees. This coupling of employees and machines fundamentally alters the work-related interactions to which employees are accustomed, as employees find themselves increasingly interacting with, and relying on, AI systems instead of human coworkers. This increased coupling of employees and AI portends a shift towards more of an “asocial system” wherein people may feel socially disconnected at work. Drawing upon the social affiliation model, we develop a model delineating both adaptive and maladaptive consequences of this situation. Specifically, we theorize that the more employees interact with AI in the pursuit of work goals, the more they experience a need for social affiliation (adaptive)—which may contribute to more helping behavior towards coworkers at work—as well as a feeling of loneliness (maladaptive) which then further impair employee well-being after work (i.e., more insomnia and alcohol consumption). In addition, we submit that these effects should be especially pronounced among employees with higher levels of attachment anxiety. Results across four studies (N = 794) with mixed methodologies (i.e., survey study, field experiment, and simulation study; Studies 1 to 4) with employees from four different regions (i.e., Taiwan, Indonesia, United States, and Malaysia) generally support our hypotheses.
Digital technologies induce organised immaturity by generating toxic sociotechnical conditions that lead us to delegate autonomous, individual, and responsible thoughts and actions to external technological systems. Aiming to move beyond a diagnostic critical reading of the toxicity of digitalisation, we bring Bernard Stiegler’s pharmacological analysis of technology into dialogue with the ethics of care to speculatively explore how the socially engaged arts—a type of artistic practice emphasising audience co-production and processual collective responses to social challenges—play a care-giving role that helps counter technology-induced organised immaturity. We outline and illustrate two modes by which the socially engaged arts play this role: 1) disorganising immaturity through artivism, most notably anti-surveillance art, that imparts savoir vivre, that is, shared knowledge and meaning to counter the toxic side of technologies while enabling the imagination of alternative worlds in which humans coexist harmoniously with digital technologies, and 2) organising maturity through arts-based hacking that imparts savoir faire, that is, hands-on knowledge for experimental creation and practical enactment of better technological worlds.
We theorize about human-team collaboration with AI by drawing upon research in groups and teams, social psychology, information systems, engineering, and beyond. Based on our review, we focus on two main issues in the teams and AI arena. The first is whether the team generally views AI positively or negatively. The second is whether the decision to use AI is left up to the team members (voluntary use of AI) or mandated by top management or other policy-setters in the organization. These two aspects guide our creation of a team-level conceptual framework modeling how AI introduced as a mandated addition to the team can have asymmetric effects on collaboration level depending on the team’s attitudes about AI. When AI is viewed positively by the team, the effect of mandatory use suppresses collaboration in the team. But when a team has negative attitudes toward AI, mandatory use elevates team collaboration. Our model emphasizes the need for managing team attitudes and discretion strategies and promoting new research directions regarding AI’s implications for teamwork.
Echoing past waves of transformation, the public sphere is awash with anxiety about automation now driven by the rise of intelligent machines. Emerging technologies encompass a wider and wider range of work, and the disruptions that will accompany the transformation of work involve pressing problems for research and practice. Communication scholarship is distinctively well equipped for the study of automation today because communication itself is increasingly the focus of automation, because the automation of work is a communication process, and because deliberations about automation will shape how we manage those disruptions. This article reviews scholarship in communication that focuses on automation, highlighting research that focuses on communication as the substance of automation, discourse about automation, and communicative practice of automation.
This chapter provides an overview of the e-government processes in Brazil. We highlight some recent initiatives, emphasizing development, challenges, the relationship with other levels of government, and best practices while providing a case for the country. Despite Brazil’s recent political, economic, and financial crises, e-government is an ongoing process. However, it can be discontinued if actions encouraging digital tool use among citizens and civil servants are wrongly set up. We argue that digital transformation requires more than just data and algorithms and is enmeshed in social relationships. Regulation serves as a foundation, but e-government is a dynamic process, much more than laws are required. By illustrating the evolution and highlighting some examples in this chapter, we hope to inspire practitioners, legislators, and policymakers to consider better public policies for participation and digital transformation to improve e-government processes.
The increasing workplace use of artificially intelligent (AI) technologies has implications for the experience of meaningful human work. Meaningful work refers to the perception that one’s work has worth, significance, or a higher purpose. The development and organisational deployment of AI is accelerating, but the ways in which this will support or diminish opportunities for meaningful work and the ethical implications of these changes remain under-explored. This conceptual paper is positioned at the intersection of the meaningful work and ethical AI literatures and offers a detailed assessment of the ways in which the deployment of AI can enhance or diminish employees’ experiences of meaningful work. We first outline the nature of meaningful work and draw on philosophical and business ethics accounts to establish its ethical importance. We then explore the impacts of three paths of AI deployment (replacing some tasks, ‘tending the machine’, and amplifying human skills) across five dimensions constituting a holistic account of meaningful work, and finally assess the ethical implications. In doing so we help to contextualise the meaningful work literature for the era of AI, extend the ethical AI literature into the workplace, and conclude with a range of practical implications and future research directions.
Dieser Beitrag skizziert die aktuelle Auseinandersetzung mit digitaler Transformation in der PR-Forschung als einen dritten, technologieorientierten Diskurs, der auf einen dialog- sowie einen nutzenorientierten Diskurs rund um den Einsatz digitaler Informations- und Kommunikationstechnologien (ICT) im Berufsfeld der PR folgt. Anschließend wird gezeigt, dass Digitalisierung immer schon alle drei, diesen Diskursen zugrunde liegenden Ideologien – deliberatives, liberales und elitäres Denken – gleichzeitig bedient hat und es wird empfohlen, diese stärker zusammenzudenken. Dadurch werden genuin digitalisierungsbedingte Spannungsfelder besser verständlich, die sich digitaler PR auf Ebene des Kommunikations-, Organisations- und Selbstverständnisses stellen und die Forschung wie Praxis stärker ins Zentrum rücken sollten.
This critical review of research on platform-mediated work argues that platform work studies are too focused on gig and remote work platforms. We introduce a framework that identifies three perspectives on how platforms reorganize work: narrow, broad, and systemic. This framework is used to examine the impact of platform-mediated work on four different aspects of work: management power, work processes, social protection and labor rights, and skills and career prospects.
Recent studies on precarity among gig workers has turned away from labour process factors to explore the role of the wider social, cultural and institutional environment. Existing western-centred studies in this aspect argue that platforms reproduce racialised and gendered hierarchies to leverage control over vulnerable populations. This study extends this literature by focusing on the migration factor in a non-western context. Using the case of Didi, drawing on ethnographic and interview data, it is argued that migrant drivers’ high tolerance for platform precarity should be understood as an imposed position, for they are actually trapped in the platform by China's state-led, tech-driven economic restructuring project, through a new mode of migrant labour differentiation comprising three factors – changes in the labour market, hegemonic gender norms and the reformed hukou system. It thus enriches our understanding of worker precarity in the gig economy by highlighting the impact of migration and the state.
Collaboration is critical to organizations and difficult when work is distributed. Prior research has indicated that when individuals are distributed, organizations respond by structuring their work to decrease reciprocal interdependence, reduce the complexity of tasks that individuals perform, or accept moderate inefficiencies. Yet in an increasing number of organizations—location-independent organizations—employees are fully distributed, exist without a physical office, and engage in reciprocally interdependent work. To understand how these distributed organizations collaborate, I undertook an inductive multiple-case study. I identify two patterns of collaboration, an asynchronous orientation and a real-time orientation, and reveal the specific enabling practices for each, with a focus on asynchronous-oriented organizations. This research contributes to the distributed work literature by detailing three novel practices that enable effective collaboration for reciprocally interdependent work without geographic or temporal alignment and to the organizational design literature by identifying distinct approaches to distributed collaboration. This study also engages with the future-of-work conversation by providing empirical grounding that enhances our understanding of the theory, boundary conditions, and nuance of the phenomenon of distributed organizations, specifically location-independent organizations.
Drawing on a qualitative, multi-case study of three kinds of geographically tethered gig work—ride-hailing, delivery, and domestic services platforms—in the United States, I examine how workers anticipate the influences of metrics, live with metrics, and cope with algorithmic precarity. Data for this project include in-depth interviews with 50 gig workers about their efforts to interpret and manage metrics as part of their everyday work practices. The analysis reveals that participants were anxious about metrics primarily because of the disciplinary outcomes, that are, the threat of job loss and the valued job features. It also directs attention to how workers felt and experienced customer-sourced ratings and system-generated behavioral metrics variously across platforms. Information asymmetries and the perceived lack of control also intensified a sense of powerlessness among participants. While participants articulated strategies that aimed at managing the uncertainty of customer-sourced ratings—and more precisely, the work-related uncertainty created by “difficult customers”—throughout service interactions, their feelings of anxiety could not be resolved. Furthermore, the (in)visibility of metrics, the settings of platform-mediated worker–customer interactions, and workers’ platform dependence contributed to the varying disciplinary power of metrics. The study contributes to understanding how metrics as affective measures mediate the trilateral relationship between platforms, workers, and customers in the gig economy.
Robots are transforming the nature of human work. Although human–robot collaborations can create newjobs and increase productivity, pundits often warn about how robots might replace humans at work andcreate mass unemployment. Despite these warnings, relatively little research has directly assessed howlaypeople react to robots in the workplace. Drawing from cognitive appraisal theory of stress, we suggestthat employees exposed to robots (either physically or psychologically) would report greater job insecurity.Six studies—including two pilot studies, an archival study across 185 U.S. metropolitan areas (Study 1), apreregistered experiment conducted in Singapore (Study 2), an experience-sampling study among engineersconducted in India (Study 3), and an online experiment (Study 4)—find that increased exposure to robotsleads to increased job insecurity. Study 3 also reveals that this robot-related job insecurity is in turnpositively associated with burnout and workplace incivility. Study 4 reveals that self-affirmation is apsychological intervention that might buffer the negative effects of robot-related job insecurity. Our findingshold across different cultures and industries, including industries not threatened by robots.
Prior studies that examine how new expertise becomes integrated into organizations have shown that different occupations work to legitimate their new expertise to develop credibility and deference from other organizational groups. In this study, we similarly examine the work that an expert occupation did to legitimate their expertise; however, in this case, they were legitimating practices that they actually considered illegitimate. We report findings from our 20-month ethnography of data analysts at a financial technology company to explain this process. We show that the company had structured data analytics in ways similar to Bechky’s idea of a captive occupation: They were dependent on their collaborators’ cooperation to demonstrate the value of data analytics and accomplish their work. The data analysts constantly encountered or were asked to provide what they deemed to be illegitimate data analysis practices such as hacking, peeking, and poor experimental design. In response, they sometimes resisted but more often reconciled themselves to the requests. Notably, they also explicitly lowered their stated standards and then worked to legitimate those now illegitimate versions of their expert practices through standardization, technology platforms, and evangelizing. Our findings articulate the relationship between captive occupations and conditions wherein experts work to legitimate what they consider illegitimate practices.
Purpose
The author analyses the strategies developed by workers and unions to obtain representation and the successes and limitations of the strategies, in a context of platform work such as Spanish dominated by labour relations of employee workers.
Design/methodology/approach
The empirical material is the result of a series of in-depth interviews conducted between August 2020 and September 2021 with 41 workers, 15 of them union delegates, in addition to 4 union members and a labour lawyer. From these interviews, the author obtains a detailed account of the working conditions and the different phases that unionism has gone through in its objective of obtaining representation in a completely new sector.
Findings
The author found that employment in the relationship does not solve all the problems of platform work, especially those related to algorithmic control, but employment in the relationship provides advantages such as the right to representation. Workers play an important role in union strategies.
Originality/value
This study is the first in Spain, where platform work in passenger transport includes the employment relationship as a legal contracting mechanism.
Herrschaft bringt immer auch Widerständigkeit hervor. Demnach stellen Regelabweichungen, die sich aus unvollständig determiniertem Arbeitshandeln ergeben, ein strukturelles Merkmal im Arbeitsprozess dar. Die Formierung eines informellen Repertoires widerständiger Praktiken im Kontext betrieblicher Herrschaft ist dabei von der Arbeitssoziologie bisher vernachlässigt worden. Um diese konzeptionelle Leerstelle zu füllen, systematisieren die Beiträger*innen die Vielzahl der Praktiken und stellen verschiedene methodische, theoretische und empirische Perspektiven einer arbeitssoziologischen Widerstandsforschung vor.
This article investigates the precarious labour conditions of Chinese food-delivery drivers in the platform economy. Drawing on one year of ethnographic fieldwork where the author worked as a food-delivery driver in Shanghai, the three key forces producing precarity in the platform labour regime are explored: (i) the platform circumvents its employer responsibilities for drivers by outsourcing the labour services of food delivery to third-party labour-hires companies; (ii) predatory algorithmic management is leveraged by the platform to control the labour process for excessive exploitation; and (iii) the institutional deprivation of citizenship rights of the rural migrants converts drivers into urban denizens with a vulnerable socio-economic labour environment. These determinants combine to produce low-paid, insecure, uncertain, and dangerous working conditions which food delivery drivers have limited power to resist both at individual and collective levels. Building on these findings, this article argues that the peculiar intersection of bogus triangular employment relations, predatory algorithmic control, and the subservient citizenship of rural migrants, produces precarity in the platform labour regime. The article highlights the role of the state and management in producing the precarity experienced by Chinese food-delivery drivers and contributes to understanding the work precarity of the platform economy in the digital age.
Technology enables actor-to-actor experience co-creation leading to value creation for the parties involved in the process. This research presents the initial impact of a mobile application developed to foster technology enabled relationship in rural Lebanon. Results indicate that technology has the potential to positively impact both host and guest fostering relationships building in all the trip stages and leading to socio-economic development and transformative experiences. In fact, the paper shows that relationships created and strengthened with the support of technology are expected to have effects at personal, community and business level.
We explore how members of a community of practice learn new tools and techniques when environmental shifts undermine existing expertise. In our 20-month comparative field study of medical assistants and patient-service representatives learning to use new digital technology in five primary care sites, we find that the traditional master-apprentice training model worked well when established practices were being conferred to trainees. When environmental change required introducing new tools and techniques with which the experienced members had no expertise, third-party managers selected newer members as trainers because managers judged them to be agile learners who were less committed to traditional hierarchies and more willing to deviate from traditional norms. This challenged community members’ existing status, which was based on the historical distinctions of long tenure and expertise in traditional tasks. In three sites, the introduction of this illegitimate learning hierarchy sparked status competition among trainees and trainers, and trainees collectively resisted learning new tools and techniques. In the other two sites, managers paired the new, illegitimate learning hierarchy with the opportunity for trainee status mobility by rotating the trainer role; here, trainees embraced learning in order to exit the lower-status trainee group and join the higher-status trainer group. Drawing on ideas of status group legitimacy and mobility, we suggest that managers’ pairing of an illegitimate learning hierarchy with the opportunity for trainee status mobility is a mechanism for enabling the situated learning of new techniques when traditional expertise erodes.
Building on an emerging literature concerning algorithmic management, this article analyzes the processes by which food delivery platforms control workers and uncovers variation in the extent to which such platforms constrain the freedoms—over schedules and activities—associated with gig work. Drawing on in-depth interviews with 55 respondents working on food delivery platforms, as well as a survey of 955 platform food delivery workers, we find that although all of the food delivery platforms use algorithmic management to assign and evaluate work, there is significant cross-platform variation. Instacart, the largest grocery delivery platform, exerts a type of control we call “algorithmic despotism,” regulating the time and activities of workers more stringently than other platform delivery companies. We conclude with a discussion of the implications of the spectrum of algorithmic control for the future of work.
We advance the concept of platformic management, and the ways in which platforms help to structure project-based or “gig” work. We do so knowing that the popular press and a substantial number of the scholarly publications characterize the “rise of the gig economy” as advancing worker autonomy and flexibility, focusing attention to online digital labor platforms such as Uber and Amazon’s Mechanical Turk. Scholars have conceptualized the procedures of control exercised by these platforms as exerting “algorithmic management,” reflecting the use of extensive data collection to feed algorithms that structure work. In this paper, we broaden the attention to algorithmic management and gig-working control in two ways. First, we characterize the managerial functions of Upwork, an online platform that facilitates knowledge-intensive freelance labor - to advance discourse beyond ride-sharing and room-renting labor. Second, we advance the concept of platformic management as a means to convey a broader and sociotechnical premise of these platforms’ functions in structuring work. We draw on data collected from Upwork forum discussions, interviews with gig workers who use Upwork, and a walkthrough analysis of the Upwork platform to develop our analysis. Our findings lead us to articulate platformic management -- extending beyond algorithms -- and to present the platform as a ‘‘boundary resource” to illustrate the paradoxical affordances of Upwork and similar labor platforms. That is, the platform (1) enables the autonomy desired by gig workers, while (2) also serving as a means of control that helps maintain the viability of transactions and protects the platform from disintermediation.
Drawing on interviews with 77 high-performing eBay business sellers in France and Belgium, this article investigates the power asymmetries generated by customers’ evaluations in online work settings. Sellers revealed a high degree of sensitivity to negative reviews, which, while infrequent, triggered feelings of anxiety and vulnerability. Their accounts exposed power asymmetries at two levels: the transactional level between sellers and customers and the governance level between sellers and eBay. Our findings highlight three main mechanisms underlying power asymmetries in this context. First, online customer evaluations have created a new form of employee monitoring in which power is exercised through the construction of visibility gaps between buyers and sellers and through an implicit coalition between buyers and the platform owner, who join together in the evaluation procedures. Second, by mediating and objectifying relations, algorithms reproduce power asymmetries among the different categories of actors, thereby constraining human agency. Third, online customer evaluations prompt sellers to exploit their practical knowledge of the algorithm to increase their agency. Through the lived experience of working for an algorithm, our findings contribute new understandings of power and agency in online work settings.
This chapter lays out a research agenda in the sociology of work for a type of data and organizational intermediary: work platforms. As an example, the authors employ a case study of the adoption of automated hiring platforms (AHPs) in which the authors distinguish between promises and existing practices. The authors draw on two main methods to do so: critical discourse analysis and affordance critique. The authors collected and examined a mix of trade, popular press, and corporate archives; 135 texts in total. The analysis reveals that work platforms offer five core affordances to management: (1) structured data fields optimized for capture and portability within organizations; (2) increased legibility of activity qua data captured inside and outside the workplace; (3) information asymmetry between labor and management; (4) an “ecosystem” design that supports the development of limited-use applications for specific domains; and (5) the standardization of managerial techniques between workplaces. These combine to create a managerial frame for workers as fungible human capital, available on demand and easily ported between job tasks and organizations. While outlining the origin of platform studies within media and communication studies, the authors demonstrate the specific tools the sociology of work brings to the study of platforms within the workplace. The authors conclude by suggesting avenues for future sociological research not only on hiring platforms, but also on other work platforms such as those supporting automated scheduling and customer relationship management.
Organizations increasingly rely on algorithm-based HR decision-making to monitor their employees. This trend is reinforced by the technology industry claiming that its decision-making tools are efficient and objective, downplaying their potential biases. In our manuscript, we identify an important challenge arising from the efficiency-driven logic of algorithm-based HR decision-making, namely that it may shift the delicate balance between employees’ personal integrity and compliance more in the direction of compliance. We suggest that critical data literacy, ethical awareness, the use of participatory design methods, and private regulatory regimes within civil society can help overcome these challenges. Our paper contributes to literature on workplace monitoring, critical data studies, personal integrity, and literature at the intersection between HR management and corporate responsibility.
Analogy—the ability to find and apply deep structural patterns across domains—has been fundamental to human innovation in science and technology. Today there is a growing opportunity to accelerate innovation by moving analogy out of a single person’s mind and distributing it across many information processors, both human and machine. Doing so has the potential to overcome cognitive fixation, scale to large idea repositories, and support complex problems with multiple constraints. Here we lay out a perspective on the future of scalable analogical innovation and first steps using crowds and artificial intelligence (AI) to augment creativity that quantitatively demonstrate the promise of the approach, as well as core challenges critical to realizing this vision.
This paper develops grounded theory to understand how and when experts in lower-power peripheral roles can develop influential relationships with higher-power line managers in core functions to regularly elicit their cooperation. I use data from an ethnographic study of experts in peripheral roles—mental health professionals hired by the U.S. Army to provide rehabilitative services to active-duty soldiers suffering from conditions such as post-traumatic stress disorder (PTSD)—and the line managers in core functions they depended on for their soldier-care recommendations to be followed—their soldiers’ direct commanders. I analyze relational histories of 56 commander–provider dyads and detail a three-phase process model I refer to as “rapid relationality.” Through the process of rapid relationality, mental health professionals capitalized on short windows of opportunity to quickly develop influential relationships with commanders, despite lacking formal authority over them, before a series of minor conflicts or a major conflict threatened their burgeoning influence. My analysis suggests it is not only what peripheral experts do that allows them to elicit cooperation from line managers but also when and how quickly they do it that matters. Speed and taking early action are important because experts cannot predict when conflict will occur and hence may have short initial windows of opportunity for establishing an influential relationship. I further find that some experts have fewer tactics available to them or must use more time-consuming and energy-intensive tactics than others to achieve the same relational influence based on their personal characteristics (e.g., gender). As a result, it may be harder for some experts to achieve relational influence before their window of opportunity closes.
Most users want their Twitter feed, Facebook page, and YouTube comments to be free of harassment and porn. Whether faced with “fake news” or livestreamed violence, “content moderators”-who censor or promote user-posted content-have never been more important. This is especially true when the tools that social media platforms use to curb trolling, ban hate speech, and censor pornography can also silence the speech you need to hear. In this revealing and nuanced exploration, award-winning sociologist and cultural observer Tarleton Gillespie provides an overview of current social media practices and explains the underlying rationales for how, when, and why these policies are enforced. In doing so, Gillespie highlights that content moderation receives too little public scrutiny even as it is shapes social norms and creates consequences for public discourse, cultural production, and the fabric of society. Based on interviews with content moderators, creators, and consumers, this accessible, timely book is a must-read for anyone who’s ever clicked “like” or “retweet.”.
This article evaluates the job quality of work in the remote gig economy. Such work consists of the remote provision of a wide variety of digital services mediated by online labour platforms. Focusing on workers in Southeast Asia and Sub-Saharan Africa, the article draws on semi-structured interviews in six countries (N = 107) and a cross-regional survey (N = 679) to detail the manner in which remote gig work is shaped by platform-based algorithmic control. Despite varying country contexts and types of work, we show that algorithmic control is central to the operation of online labour platforms. Algorithmic management techniques tend to offer workers high levels of flexibility, autonomy, task variety and complexity. However, these mechanisms of control can also result in low pay, social isolation, working unsocial and irregular hours, overwork, sleep deprivation and exhaustion.
A number of lawsuits in the United States are challenging the employment classification of workers in the platform economy. Employee status is a crucial gateway in determining entitlement to labor and employment law protections. In response to this uncertainty, some commentators have proposed an “intermediate”, “third,” or “hybrid” category, situated between the categories of “employee” and “independent contractor.”
After investigating the status of platform workers in the United States, the authors provide snapshot summaries of five legal systems that have experimented with implementing a legal tool similar to an intermediate category to cover non-standard workers: Canada, Italy, Spain, Germany, and South Korea. These various legal systems have had diverse results. There has been success in some instances, and unintended consequences in others.
Accordingly, the authors recommend proceeding with caution in considering the creation of a third category. That is due to the risk of arbitrage between the categories, and the possibility that some workers will lose rights by having their status downgraded into the third category. Cherry and Aloisi posit employee status as the default rule for most gig workers. The authors propose an exception for those working on a de minimis basis or those engaged in volunteerism for altruistic reasons.
2015 ACM. This paper introduces architectural and interaction patterns for integrating crowdsourced human contributions directly into user interfaces. We focus on writing and editing, complex endeavors that span many levels of conceptual and pragmatic activity. Authoring tools offer help with pragmatics, but for higher-level help, writers commonly turn to other people. We thus present Soylent, a word processing interface that enables writers to call on Mechanical Turk workers to shorten, proofread, and otherwise edit parts of their documents on demand. To improve worker quality, we introduce the Find-Fix-Verify crowd programming pattern, which splits tasks into a series of generation and review stages. Evaluation studies demonstrate the feasibility of crowdsourced editing and investigate questions of reliability, cost, wait time, and work time for edits.
History of gender and labor in British computing
The number of predictive technologies used in the U.S. criminal justice system is on the rise. Yet there is little research to date on the reception of algorithms in criminal justice institutions. We draw on ethnographic fieldwork conducted within a large urban police department and a midsized criminal court to assess the impact of predictive technologies at different stages of the criminal justice process. We first show that similar arguments are mobilized to justify the adoption of predictive algorithms in law enforcement and criminal courts. In both cases, algorithms are described as more objective and efficient than humans’ discretionary judgment. We then study how predictive algorithms are used, documenting similar processes of professional resistance among law enforcement and legal professionals. In both cases, resentment toward predictive algorithms is fueled by fears of deskilling and heightened managerial surveillance. Two practical strategies of resistance emerge: foot-dragging and data obfuscation. We conclude by discussing how predictive technologies do not replace, but rather displace discretion to less visible—and therefore less accountable—areas within organizations, a shift which has important implications for inequality and the administration of justice in the age of big data.
Technological advances and the big-data revolution have facilitated fine-grained, high-frequency, low-cost measurement of individuals’ work. Yet we understand little about the influences of such quantification of work on workers’ behavior and performance. This article investigates how and when quantification of work affects worker productivity. We argue that quantification affects worker productivity via auto-gamification, or workers’ inadvertent transformation of work into an independent, individual-level game. We further argue that quantification is likely to raise productivity in a context of simple work, where auto-gamification is motivating because quantified metrics adequately measure the work being performed. When work is complex, by contrast, quantification reduces productivity because quantified metrics cannot adequately measure the multifaceted work being performed, causing auto-gamification to be demotivating. To substantiate our argument, we study implementation of an RFID measurement technology that quantifies individual workers’ output in real time at a garment factory in India. Qualitative evidence uncovers the auto-gamification mechanism and three conditions that enable it; a natural experiment tests the consequences of quantification of work for worker productivity. This article contributes to the study of quantification, work games, technology, and organizations, and we explore the policy implications of further quantification of work.
In this pictorial, we explore how emergent menstrual biosensing technologies compound existing concerns for the everyday ethics of extracting and analyzing intimate data. Specifically, we review the data practices of a set of existing menstrual tracking applications and use that analysis to inform the design of speculative near future technologies. We present these technologies here in the form of a product catalog for a fictional company called Vivewell. Through this work, we contribute both a set of speculative design proposals and a case study of a design project that begins with the analysis of existing data policies.
The term “crowdwork” describes a new form of digital work that is organized and regulated by internet-based platforms. This chapter examines how crowdwork platforms ensure their virtual workforce’s commitment and control its performance despite its high mobility, anonymity, and dispersion. The findings are based on a case study analysis of 15 microtask and macrotask platforms, encompassing 32 interviews with representatives of crowdwork platforms, and crowdworkers, as well as an analysis of the platforms’ homepages and community spaces. The chapter shows that performance control on crowd platforms relies on a combination of direct control, reputation systems, and community building, which have until now been studied in isolation or entirely ignored. Moreover, the findings suggest that while all three elements can be found on both microtask and macrotask platforms, their functionality and purpose differ. Overall, the findings highlight that platforms are no neutral intermediaries but organizations that adopt an active role in structuring the digital labor process and in shaping working conditions. Their managerial structures are coded and objectified into seemingly neutral technological infrastructures, whereby the underlying power relations between capital and labor become obscured.
Most studies of technologies’ impact on occupational change focus on occupational groups’ adoption and use of particular technologies in a field or workplace. Drawing on an 18-month ethnographic study of a crime laboratory, I focus instead on “evaluative spillovers”: the comparisons that occupations encounter when technologies change the work of neighboring occupations in their field. I explore what happened when DNA profiling was held up as the “gold standard” of forensic evidence, resulting in scientific, public, and legal scrutiny of other forensic science occupational groups. Comparisons with DNA profiling challenged the working techniques and the values of firearms examiners, toxicologists, and narcotics analysts, but each group responded differently, either embracing or resisting changes to their work practices. Their responses were predicated on the institutional pathways that evaluative spillovers traveled through the field in locales such as professional association meetings and court proceedings. These three aspects of the occupational system—technique, values, and institutional pathways—influenced how workers negotiated the impact of technological change in the field of forensic science.
This paper examines the work practices involved in making data legible to machines and machine output legible to humans. The study is based on ethnographic research of a team of art experts at DNArt – a data classification system that features a growing database of art images, a classification scheme, a similarity matching algorithm, and a website that together serve as a consumer judgment device in an emerging online market for art. I analyze interactions from meeting observations, interviews, documentation, and online interaction data to show how non-technical art experts explain and repair sociotechnical breakdowns – when their expectations for similarity between art images and artists differ from the similarity relations produced by the algorithm. By repairing breakdowns, the art experts construct the algorithm anew, as a legitimate revealer of similarity in art. In doing so, the team's repair work is folded back into the black box of the algorithm, rendering it invisible and unacknowledged, sometimes even by the experts themselves.
Cambridge Core - Organisational Sociology - Automating Finance - by Juan Pablo Pardo-Guerra
In a talk in 2013, Karin Knorr Cetina referred to ‘the interaction order of algorithms’, a phrase that implicitly invokes Erving Goffman's ‘interaction order’. This paper explores the application of the latter notion to the interaction of automated-trading algorithms, viewing algorithms as material entities (programs running on physical machines) and conceiving of the interaction order of algorithms as the ensemble of their effects on each other. The paper identifies the main way in which trading algorithms interact (via electronic ‘order books’, which algorithms both ‘observe’ and populate) and focuses on two particularly Goffmanesque aspects of algorithmic interaction: queuing and ‘spoofing’, or deliberate deception. Following Goffman's injunction not to ignore the influence on interaction of matters external to it, the paper examines some prominent such matters. Empirically, the paper draws on documentary analysis and 338 interviews conducted by the author with high-frequency traders and others involved in automated trading.
This qualitative case study adopts a labour process analysis to unpack the distinctive features of capital’s control regimes in the food-delivery segment of the Australian platform-economy and assess labour agency in response to these. Drawing upon worker experiences with the Deliveroo and UberEATS platforms, it is shown how the labour process controls are multi-facetted and more than algorithmic management, with three distinct features standing out: the panoptic disposition of the technological infrastructure, the use of information asymmetries to constrain worker choice and the obfuscated nature of their performance management systems. Combined with the workers’ precarious labour market positions and the Australian political-economic context, only limited, mainly individual, expressions of agency were found.
Trained machine learning models are increasingly used to perform high-impact tasks in areas such as law enforcement, medicine, education, and employment. In order to clarify the intended use cases of machine learning models and minimize their usage in contexts for which they are not well suited, we recommend that released models be accompanied by documentation detailing their performance characteristics. In this paper, we propose a framework that we call model cards, to encourage such transparent model reporting. Model cards are short documents accompanying trained machine learning models that provide benchmarked evaluation in a variety of conditions, such as across different cultural, demographic, or phenotypic groups (e.g., race, geographic location, sex, Fitzpatrick skin type [15]) and intersectional groups (e.g., age and race, or sex and Fitzpatrick skin type) that are relevant to the intended application domains. Model cards also disclose the context in which models are intended to be used, details of the performance evaluation procedures, and other relevant information. While we focus primarily on human-centered machine learning models in the application fields of computer vision and natural language processing, this framework can be used to document any trained machine learning model. To solidify the concept, we provide cards for two supervised models: One trained to detect smiling faces in images, and one trained to detect toxic comments in text. We propose model cards as a step towards the responsible democratization of machine learning and related artificial intelligence technology, increasing transparency into how well artificial intelligence technology works. We hope this work encourages those releasing trained machine learning models to accompany model releases with similar detailed evaluation numbers and other relevant documentation.
Silicon Valley technology is transforming the way we work, and Uber is leading the charge. An American startup that promised to deliver entrepreneurship for the masses through its technology, Uber instead built a new template for employment using algorithms and Internet platforms. Upending our understanding of work in the digital age, Uberland paints a future where any of us might be managed by a faceless boss. The neutral language of technology masks the powerful influence algorithms have across the New Economy. Uberland chronicles the stories of drivers in more than twenty-five cities in the United States and Canada over four years, shedding light on their working conditions and providing a window into how they feel behind the wheel. The book also explores Uber’s outsized influence around the world: the billion-dollar company is now influencing everything from debates about sexual harassment and transportation regulations to racial equality campaigns and labor rights initiatives. Based on award-winning technology ethnographer Alex Rosenblat’s firsthand experience of riding over 5,000 miles with Uber drivers, daily visits to online forums, and face-to-face discussions with senior Uber employees, Uberland goes beyond the headlines to reveal the complicated politics of popular technologies that are manipulating both workers and consumers.
This article is the fifth in a series to celebrate the 70th anniversary of the ILR Review. The series features articles that analyze the state of research and future directions for important themes this journal has featured over many years of publication.
The decline in unionization experienced in the United States over the past 40 years raises a question of fundamental importance to workers, society, and the field of industrial relations: Have workers lost interest in having a voice at work, or is there a gap between workers’ expectations for a voice and what they actually experience? And if a “voice gap” exists, what options are available to workers to close that gap? The authors draw on a nationally representative survey of workers that both updates previous surveys conducted in 1977 and 1995 and goes beyond the scope of these previous efforts to consider a wider array of workplace issues and voice options. Results indicate that workers believe they should have a voice on a broad set of workplace issues, but substantial gaps exist between their expected and their actual level of voice at work. Nearly 50% of non-union workers say they would vote for a union, compared to approximately one-third in the two prior national surveys, which points to continued interest in unions as a voice mechanism. Additionally, the authors find significant variation in the rates of use of different voice options and workers’ satisfaction with those options. The results suggest that a sizable voice gap exists in American workplaces today, but at the same time, no one voice option fits all workers or all issues.
A solution to marketplace information asymmetries is to have trading partners publicly rate each other post-transaction. Many have shown these ratings are effective; we show that their effectiveness deteriorates over time. The problem is that ratings are prone to inflation, with raters feeling pressure to leave "above average" ratings, which in turn pushes the average higher. This pressure stems from raters' desire to not harm the rated seller. As the potential to harm is what makes ratings effective, reputation systems, as currently designed, sow the seeds of their own irrelevance.
Global online platforms match firms with service providers around the world, in services ranging from software development to copywriting and graphic design. Unlike in traditional offshore outsourcing, service providers are predominantly one-person microproviders located in emerging-economy countries not necessarily associated with offshoring and often disadvantaged by negative country images. How do these microproviders survive and thrive? We theorize global platforms through transaction cost economics (TCE), arguing that they are a new technology-enabled offshoring institution that emerges in response to cross-border information asymmetries that hitherto prevented microproviders from participating in offshoring markets. To explain how platforms achieve this, we adapt signaling theory to a TCE-based model and test our hypotheses by analyzing 6 months of transaction records from a leading platform. To help interpret the results and generalize them beyond a single platform, we introduce supplementary data from 107 face-to-face interviews with microproviders in Southeast Asia and Sub-Saharan Africa. Individuals choose microprovidership when it provides a better return on their skills and labor than employment at a local (offshoring) firm. The platform acts as a signaling environment that allows microproviders to inform foreign clients of their quality, with platform-generated signals being the most informative signaling type. Platform signaling disproportionately benefits emerging-economy providers, allowing them to partly overcome the effects of negative country images and thus diminishing the importance of home country institutions. Global platforms in other factor and product markets likely promote cross-border microbusiness through similar mechanisms.
As seen in Wired and Time
A revealing look at how negative biases against women of color are embedded in search engine results and algorithms
Run a Google search for “black girls”—what will you find? “Big Booty” and other sexually explicit terms are likely to come up as top search terms. But, if you type in “white girls,” the results are radically different. The suggested porn sites and un-moderated discussions about “why black women are so sassy” or “why black women are so angry” presents a disturbing portrait of black womanhood in modern society.
In Algorithms of Oppression, Safiya Umoja Noble challenges the idea that search engines like Google offer an equal playing field for all forms of ideas, identities, and activities. Data discrimination is a real social problem; Noble argues that the combination of private interests in promoting certain sites, along with the monopoly status of a relatively small number of Internet search engines, leads to a biased set of search algorithms that privilege whiteness and discriminate against people of color, specifically women of color.
Through an analysis of textual and media searches as well as extensive research on paid online advertising, Noble exposes a culture of racism and sexism in the way discoverability is created online. As search engines and their related companies grow in importance—operating as a source for email, a major vehicle for primary and secondary school learning, and beyond—understanding and reversing these disquieting trends and discriminatory practices is of utmost importance.
An original, surprising and, at times, disturbing account of bias on the internet, Algorithms of Oppression contributes to our understanding of how racism is created, maintained, and disseminated in the 21st century.