Technology Use, Exposure to Natural Hazards, and Being Digitally Invisible: Implications for Policy Analytics

Abstract

Policy analytics combines new data sources, such as from mobile smartphones, Internet of Everything devices, and electronic payment cards, with new data analytics techniques for informing and directing public policy. However, those who do not own these devices may be rendered digitally invisible if data from their daily actions are not captured. We explore the digitally invisible through an exploratory study of homeless individuals in Phoenix, Arizona, in the context of extreme heat exposure. Ten homeless research participants carried a temperature-sensing device during an extreme heat week, with their individually experienced temperatures (IETs) compared to outdoor ambient temperatures. A nonhomeless, digitally connected sample of 10 university students was also observed, with their IETs analyzed in the same way. Surveys of participants complement the temperature measures. We found that homeless individuals and university students interact differently with the physical environment, experiencing substantial differences in individual temperatures relative to outdoor conditions, potentially leading to differentiated health risks and outcomes. They also interact differently with technology, with the homeless having fewer opportunities to benefit from digital services and lower likelihood to generate digital data that might influence policy analytics. Failing to account for these differences may result in biased policy analytics and misdirected policy interventions.

Big Data for Policymaking: Great Expectations, but with Limited Progress?

“…Watts () has argued, along similar lines, that social science should be more solution oriented, trialing different approaches to problem solving in different settings as a means to maximize the possibilities for successful implementation of knowledge. A 2017 special issue of the journal Policy & Internet on “Data for Public Policy” was initiated at the inaugural 2015 “Data for Policy” conference held at the University of Cambridge (Meyer, Crowcroft, Engin, & Alexander , p. 4), which highlighted emerging efforts to use “data to inform public policy with actual examples of successes and lessons from failures.” Articles in this special issue address topics as diverse as local government service provision (Malomo & Sena, ), the use of microdata to model unemployment (Guerrero & López, ), using open data to understand how people in a community have the capacity to act on their own behalf (Piscopo, Siebes, & Hardman, ), and the problem of policy that misses people—such as the homeless—who are not generating data (what the authors call being “digitally invisible”; see Longo, Kuras, Smith, Hondula, & Johnston, ).…”

The evolution of citizen and stakeholder engagement in Canada, from Spicer to #Hashtags

“…Digital engagement tools allow for micro‐efforts or “tiny acts” (Margetts et al ) in which participation is quantitatively different than previous versions. Additional device inputs such as wearable technology, Internet of Things sensors, and mobile smartphones can allow for citizen input through monitoring of actions rather than explicit, conscious contributions, providing a basis for a big data approach to engagement (Longo et al ).…”

Constructing a Public Narrative of Regulations for Big Data and Analytics: Results From a Community-Driven Discussion

“…For example, the generally nomothetic and positivistic epistemologies of Big Data projects approach social issues through a meta-analytic lens, thereby limiting consideration of the individual. This limited perspective of statistics-informed narratives must be weighted in policy design, particularly as they relate to colonized and marginalized communities prone to technological invisibility (Longo, Kuras, Smith, Hondula, & Johnston, 2017; O’Neil, 2017; Robinson et al, 2015), yet is often overlooked in the rush for the next panacea (Kitchin, 2014b). Moreover, automated means of data creation can operate in an abstract manner, divorcing Big Data from the members of the public for whom it represents (Driscoll & Walker, 2014).…”