Anoush Margaryan, Glasgow Caledonian University, Caledonian Academy: Research Centre for Technology-enhanced Professional Learning
The unfolding digitisation of our society has stimulated the development of new types of work practices collectively referred to as ‘virtual work’(Huws, 2014). These emerging work practices challenge traditional patterns of individual agency, organisation, power, stability, responsibility and learning (Littlejohn and Margaryan, 2013).
One type of virtual work is crowdwork ‘a form of employment in which a large group of otherwise disconnected people are brought together within Internet-based platforms for the purpose of performing a task. There are two type of crowdwork: microwork and online freelancing. Microwork (exemplified by Amazon Mechanical Turk, AMT) refers to projects broken down into microtasks that can be completed in seconds or minutes, such as image tagging or data entry. Distribution and monitoring of microwork takes place largely via algorithms. Online freelancing (typified by Upwork) is a form of crowdwork in which employers contract professional services, such as web development and graphic design, to distributed workers. In contrast to microwork, online freelancing focuses on more complex projects performed over longer periods of time ‘hours, days or months.
In this distributed, fragmented and unregulated type of work in which workers do not have access to learning and professional development opportunities available within traditional employment (training or access to experienced colleagues), how do crowdworkers go about organising their learning? What strategies do crowdworkers use to identify and fulfill their learning needs and connect with other to learn from?
The learning practices of crowdworkers have not been investigated empirically. There have been intervention studies in which researchers, mainly computer scientists, designed and evaluated various software tools to support aspects of learning within crowdwork such as giving and receiving feedback (eg Dow et al., 2012); however these interventions have been driven by the capabilities of technology rather than the goal of understanding crowdworkers’learning practices and they do not shed light on the actual learning practices involved. A recent small-scale exploratory study examined the challenges and opportunities of crowdwork for workers’employability (Barnes et al., 2015). This study does not explore how crowdworkers learn, nor does it explicitly examine the learning potential of crowdwork. Limited data presented in this study suggest that crowdworkers sometimes self-organise using social media to help each other develop skills; however, no description or analyses of such practices have been published to date.
To address the gap in the literature we devised a study to identify and describe the strategies that crowdworkers use to self-organise and self-direct their learning at work. The project examines the personal motivations, goals, agency beliefs and pathways underpinning crowdworkers’learning (what, why, how and with whom they learn within crowdwork platforms); as well as individual and environmental factors impacting upon this learning. The project also examines whether current forms of crowdwork support or hinder learning, developing empirically-based recommendations as to how platforms and practices could be reshaped and reorganised to increase their learning potential. The research questions are:
1. What learning strategies do crowdworkers use, and why and how do they use these strategies to organise and direct their learning within crowdwork platforms?
2. How do different types of crowdwork ‘microwork and online freelancing (structure, design and nature of work tasks offered) impact learning?
3. Which key individual and environmental factors are essential to productive learning within crowdwork?
4. How could work processes within crowdwork platforms be designed to foster learning?
5. What are the roles of different key stakeholders - the worker, the platform provider, the employer, and the policymakers ‘in fostering learning within crowdwork settings?
The study draws on a mixed-method design, including a survey followed by in-depth interviews. Data are being collected from two platforms: AMT (microwork) and Upwork (online freelancing). Data collection involves 3 stages. First, crowdworkers self-regulatory learning strategies, workplace learning activities and work tasks are analysed using an extant instrument, Self-Regulated Learning at Work Questionnaire, SRLWQ (Fontana et al., 2015). The survey is supplemented by interviews to ascertain crowdworkers’learning pathways, current learning goals and motivations. Third, publically available data and documentation about training provision and support for workers’learning that may be offered by AMT and Upwork is analysed and representatives of these platforms and selected employers are interviewed to contextualise the crowdworkers’perspectives.
Initial findings suggest that that there are qualitative and quantitative differences in learning practices within these two platforms. Specifically, the greater diversity and complexity of tasks available within Upwork and the higher educational and skill level of online freelancers appears to lead to Upworkers adopting quantitatively and qualitatively different learning strategies than those adopted by AMT workers. The learning-intensity of microwork appears to be lower than that of online freelancing-type crowdwork. This study is ongoing and the combined findings from all three phases will be presented at the conference.
Crowdwork is a growing type of employment, in both developed and developing countries. Improved understanding of learning practices within crowdwork would inform the design of crowdwork platforms; empower crowdworkers to direct their own learning; and help platforms, employers, and policymakers enhance the learning potential of crowdwork. Furthermore, understanding crowdworkers’learning practices is essential to the enhancement of the developmental economic potential of crowdwork (Lehdonvirta et al., 2011).
References
Barnes, S.-A., Green, A., & de Hoyos, M. (2015). Crowdsourcing and work: Individual factors and circumstances influencing employability. New Technology, Work and Employment, 30(1), 16-31.
Dow, S., Kulkarni, A., Klemmer, S., & Hartmann, B. (2012). Shepherding the crowd yields better work. Proceedings of the 2012 CSCW conference. New York: ACM.
Fontana, P., Milligan, C., Littlejohn, A., & Margaryan, A. (2015). Measuring self-regulated learning in the workplace: An instrument validation. International Journal of Training and Development, 19(1), 32-52.
Huws, U. (2014). Labor in the global digital economy. New York: Monthly Review Press. Lehdonvirta, V., & Ernkvist, M. (2011). Converting the virtual economy into development potential. Washington, DC: World Bank.
Littlejohn, A., & Margaryan, A. (2013). Technology-enhanced professional learning. London: Routledge.
