Explore chapters and articles related to this topic
Quantitative Modeling of Dynamic Human-Agent Cognition
Published in Michael D. McNeese, Eduardo Salas, Mica R. Endsley, Contemporary Research, 2020
James Schaffer, James Humann, John O’Donovan, Tobias Höllerer
Participants in both studies were recruited on Amazon Mechanical Turk (AMT). AMT is a web service that gives tools to researchers who require large numbers of participants and are capable of collecting data for their experiment in an online setting. AMT has been studied extensively for validity; notably Buhrmester, Kwang, and Gosling (2011) found that the quality of data collected from AMT is comparable to what would be collected from laboratory experiments (Hauser & Schwarz, 2015). Furthermore, since clickstream data can be collected, satisficing—the act of rapidly “tab-clicking” through study questionnaires—is easy to detect.
The physicality of digital public space
Published in Naomi Jacobs, Rachel Cooper, Living in Digital Worlds, 2018
Some have suggested that remote accessing of robots could be used to outsource menial tasks such as housework, which are still extremely difficult to automate. Services such as Amazon’s Mechanical Turk currently allow work such as translation or image identification (which is difficult for computers but easy for humans) to be done by workers in developing countries, who work at much lower rates of pay. If telepresence might allow workers to control a robot that can perform housework, this might undercut prices and lead to a robotic worker revolution. There would be economic and social implications for such a move, but also, potentially, privacy ones. Early pilot studies of such a system revealed anxiety in those who did not want an anonymous presence in their household, since it was impossible to connect with the user or users of the robots in any meaningful way. Systems are therefore in development for ways in which the tasks could be carried out while also using filters and other technological solutions to obscure personal details such as private documents, or indeed any object which does not need to be identified to carry out the task at hand (Harris, 2014).
Network-Based Interaction
Published in Julie A. Jacko, The Human–Computer Interaction Handbook, 2012
Networking has made remote working possible for many years both telecommuting from home and also more nomadic teleworkers such as sales representatives on the road or in hotels (Denbigh 2003). These are largely traditional working relationships, simply freed from the constraints of the office desk. However, networks have made possible a number of more radical remote working styles. Amazon’s Mechanical Turk (Amazon 2010) and similar marketplaces allow small pieces of work to be requested, executed, and paid where the person wanting the work and the person doing it have no contact or knowledge of each other except through the website.
The Effects of Feedback and Goal on the Quality of Crowdsourcing Tasks
Published in International Journal of Human–Computer Interaction, 2021
Jae-Eun Lim, Joonhwan Lee, Dongwhan Kim
Amazon’s Mechanical Turk is an online marketplace that allows individuals or organizations to post simple tasks and have them completed by large crowd workers (Kittur et al., 2008). As crowdsourcing is becoming an increasingly popular tool for user studies at low cost, many studies have been conducted on how to obtain efficient and reliable results from the general public. A number of studies raised quality control as one of the most significant issues for crowdsourcing (Ipeirotis et al., 2010; Kinnaird et al., 2013; Yu et al., 2014).