As society increasingly relies on the Internet for work, education, health care, recreation, and many other aspects of daily life, the prevalent and persistent inequity in people’s ability to access, adopt, and use the Internet is more evident than ever. In the wake of the COVID-19 pandemic, these inequities have become apparent at the global, national, municipal, and neighborhood scales. For example, at the start of the pandemic in the City of Chicago, more than 110,000 children under the age of 18 lacked access to broadband Internet connectivity, and in certain neighborhoods, nearly half of children under 18 lacked Internet connectivity. These gaps become even more pronounced in times of crisis and natural disaster when communities rely almost exclusively on Internet communications for many critical aspects of daily life. The issue has not gone unnoticed by policymakers, who have developed a variety of local efforts to provide schoolchildren with Internet service and Internet-connected devices, and recently passed federal legislation committing nearly $65 billion to improve Internet access, affordability, and literacy. Stakeholders who are making decisions about where and how to direct these investments need better information along the following three dimensions:

  1. Measurement techniques and datasets that directly address unknown questions and evaluate the effectiveness of different interventions;
  2. Data-driven collaborations with communities that are underserved by current Internet infrastructure to develop and test different options for infrastructure investments, the effectiveness of which can depend critically on the specific characteristics and needs found in different communities;
  3. Better data and analysis about how Internet connectivity relates to the social and individual conditions that contribute to whether and how the Internet actually improves people’s lived experience.

Answering these questions requires developing entirely new approaches—from the types of data we gather, to the choices we make regarding where and how to gather it, to the techniques we develop to inform decisions. Towards this goal, the Data Science Institute (DSI) has formed a collaborative research partnership at the  University of Chicago between the Department of Computer Science and the Crown Family School of Social Work, Policy, and Practice to establish the Internet Equity initiative (IEI).

The IEI brings together experts in computer science, social science, data science, and public policy to take a data-oriented approach to close the “digital divide”,  focusing on the following three areas:

Problem-driven measurement, data, and analysis.

Currently available data on the state of internet access and functionality in the U.S. offer a skewed and sometimes erroneous portrait of our current Internet infrastructure, capabilities, and service options. Effective technical, policy, and regulatory solutions require data and analysis that are more responsive to the specific questions that Internet stakeholders are posing. A significant shortcoming of current datasets, for example, is that their samples significantly under-represent specific geographies, especially those that lack good connectivity and have fewer Internet subscribers. Tackling this problem requires a completely new approach to internet measurement, gathering data at all “levels of the stack,” from physical infrastructure to quality of experience, and at varying granularities, including hyper-local. Our initial efforts are focused on developing a proof-of-concept across the City of Chicago that can be replicated and deployed nationwide (and eventually globally).

Models to evaluate diverse connectivity options.

As connectivity technologies continue to progress, the variety of technical deployment options for providing connectivity within a community has expanded. In any particular setting, the best (and most cost-effective) connectivity solutions have to solve at least two problems: how to get connectivity to a building where people live or work and how to distribute connectivity within the building, such as to individual apartments. The most appropriate combination of components and technologies for any particular setting ultimately depends on myriad factors, from neighborhood topography and building materials to the architecture and configuration of each building, to the governance and trust dynamics between building inhabitants, owners, and other community members. Ultimately, every deployment scenario is different; thus, new approaches are needed to provide scalable models that can help property owners, community organizations, and other stakeholders design and evaluate the combination of technologies that can provide the best connectivity to people across diverse contexts.

Social effects of infrastructure investments.

Ultimately, the value of new investments in Internet infrastructure hinges on whether these investments lead to improved outcomes in society. One of the significant challenges is thus determining how to measure and characterize the downstream social effects of infrastructure investment, both to evaluate the effectiveness of these investments and to understand how different population groups are benefitting from these investments. Towards this goal, the IEI will investigate the individual and social effects of public and private attempts to promote Internet inequity. One current project, a partnership with Chicago Public Schools and the Chicago Consortium on School Research, evaluates how Chicago Connected, a free Internet connectivity program made available to all public school students soon after the onset of the pandemic, has benefited groups of students from different socioeconomic backgrounds and residential locations. Another is a partnership with UChicago Medicine, where we are evaluating whether the presence of Internet connectivity and Internet-connected devices in homes—such as voice assistants—can improve health outcomes.