Below in text and attached as a PDF are NDIA’s comments on a National Broadband Research Agenda. The Request for Comments was posted in the Federal Register on 9-9-2016 by the National Telecommunications and Information Administration (NTIA) and the National Science Foundation (NSF). We submitted our comments on 10-11-2016.

Summary of our comments:

  1. The Department of Commerce should find a way to make existing Census tract-level data on household Internet access available to researchers and communities now.
  2. NTIA should take immediate steps to collect and preserve the records of BTOP SBA and PCC projects.
  3. NTIA and NSF should give priority in the Broadband Research Agenda to investigating the impacts of “digitalization without inclusion”, especially through Federal actions and policies, on vulnerable populations and communities.

October 11, 2016

Department of Commerce, National Telecommunications and Information Administration
National Science Foundation
Docket No. 160831803–6803–01
Notice and Request for Comments: National Broadband Research Agenda


The National Digital Inclusion Alliance (NDIA) is grateful for the opportunity to contribute to NTIA’s and NSF’s deliberations regarding a National Broadband Research Agenda.

NDIA is a national nonprofit organization which seeks to provide a united voice for local technology training, home broadband access and public broadband access programs. NDIA’s 200+ affiliated organizations currently include 33 national nonprofits and 154 local public and nonprofit organizations in 34 states and the District of Columbia. Our local affiliates include 18 municipal government bodies, 25 local public libraries and regional library councils, and 88 local nonprofits, along with college and university programs, state broadband agencies, school districts and public housing authorities. A full, updated list of NDIA’s affiliates is at

We are sure that this proceeding will elicit a long list of important research topics and proposals, many of which NDIA would endorse. We are limiting our comments to three points that we believe have great practical importance for all our affiliates.

  1. We need reliable, actionable local data about household broadband access that we can use now (point 1, addressing question 13 of the RFC).
  2. We need data from large-scale, local community digital inclusion efforts that received Federal funding in the past to be preserved and made available for researchers (point 2, addressing question 13 of the RFC).
  3. We need research on the consequences of continued “digitalization” of access to Federal services and programs without accompanying investments in digital inclusion to preserve low income households’ access to those services and programs (point 3, addressing questions 4, 7 and 10 of the RFC).

1) The Department of Commerce should find a way to make existing Census tract-level data on household Internet access available to researchers and communities now. (Addressing RFC question 13)

Since 2013, the U.S. Census American Community Survey has included a series of questions on household computer ownership and Internet access. The resulting data has been published in the ACS One Year Estimates for 2013, 2014, and 2015 as Tables B28001 through B28009. It includes household Internet subscription totals, technology shares within those totals, and demographics of connected and unconnected households including householder age, income, educational attainment, labor force status and race.

Unfortunately, this invaluable data has been published only at the “place” level, and only for places above 65,000 population (and in more limited form for places above 20,000). This is an unintended consequence of the Census Bureau’s 2015 decision to eliminate the ACS Three Year Estimates for budgetary reasons. A Three Year Estimate for 2015 would have provided the full range of Table 28001-28009 data for all Census tracts — and thus for rural communities, tribal areas, urban neighborhoods, counties, Congressional Districts, metropolitan areas and states. The elimination of Three Year Estimates has delayed its availability for any and all of these geographies until Fall 2018 at the earliest. [NOTE:  The italicized passage is incorrect. The ACS Three Year Estimates did not provide data for Census tracts — only for “places” above 20,000 population.]

Releasing the ACS computer and Internet series tract-level data now, rather than holding it back for two more years, is the single most important step the Government could take to facilitate local communities’ understanding of their home broadband access and adoption situations — and thus their ability to plan and take action to improve those situations.  Immediate release of this data would provide a key data resource for two important initiatives arising from Broadband Opportunity Council recommendations – NTIA’s own Community Connectivity Initiative, as well as the Department of Housing and Urban Development’s proposal to add low/moderate income residential broadband access to the topics covered by its Consolidated Plan process. And of course it would it constitute a major new resource for academic research in the field.

NDIA urges NTIA and the Census Bureau to find a way to make ACS Table 28001-28009 tract-level data available for the use of communities and researchers now.

2) NTIA should take immediate steps to collect and preserve the records of BTOP SBA and PCC projects.  (Addressing RFC question 13)

Between 2009 and 2013, NTIA’s Broadband Technology Opportunities Program invested more than $450 million in the nation’s largest-ever digital inclusion initiative – 110 Sustainable Broadband Adoption and Public Computer Center projects in cities, rural communities, tribal areas and institutional settings throughout the U.S. In just over three years, BTOP SBA and PCC projects provided digital literacy training to between 800,000 and 900,000 individuals, and helped more than half a million households connect to the Internet.

All SBA and PCC projects were required to report detailed results on a quarterly basis, and to preserve the records supporting their reports for several years after project completion. NDIA affiliates operated a number of SBA and PCC projects, and are intimately familiar with the large volume of program and client data in these project records.

Unfortunately, NTIA has never arranged to collect, preserve or curate these project records of digital inclusion services delivered to a million or more participants, which could provide a unique – indeed an irreplaceable – trove of data for researchers to investigate program models, operations and outcomes.

We note that the obligations of BTOP project sponsors to preserve their project records are expiring. For many sponsors, three years have already passed since their contract closing dates.  With no expression of interest by NTIA or other parties, it’s quite likely that sponsors will soon begin disposing of records they are no longer obliged to preserve.

NDIA urges NTIA to take steps as soon as possible to ensure that BTOP SBA and PCC project records — by far the biggest potential source of data on the only major Federal investment in broadband access, adoption and training for disadvantaged Americans — are collected, safely stored and made available to researchers.

3) NTIA and NSF should give priority in the Broadband Research Agenda to investigating the impacts of “digitalization without inclusion”, especially through Federal actions and policies, on vulnerable populations and communities .  (Addressing RFC questions 4, 7 and 10)

One of NDIA’s recommendations to the Broadband Opportunities Council in June 2015 was to institute a process for Federal agencies to assess the “digital exclusion impact” of any new online service before its launch. We want to reiterate this recommendation here, but extend its scope to all Federal policies and programs which effectively require (or significantly privilege) the use of broadband access and skills; and urge NTIA and NSF to adopt it as a priority of the National Broadband Research Agenda.

According to the American Community Survey[1],  51% of all U.S. households with incomes below $20,000, and 44% of households with incomes below $35,000, did not have home broadband subscriptions (including mobile device access) in 2015. The numbers behind these percentages are large – about ten million unconnected households below $20,000, and almost seventeen million unconnected households below $35,000.  Other than

the Federal Communications Commission’s modest Lifeline broadband initiative – whose impact remains to be seen – there is no comprehensive Federal strategy to ensure that even a significant fraction of these households will gain affordable access to mainstream Internet service any time soon, let alone training or support to become competent Internet users.

So when Federal agencies, or their state or local intermediaries, consider steps to move their public services to the Internet – especially services that are important to low income households – both prudence and fairness should lead those agencies to investigate (in order to effectively mitigate) the potential negative impacts of their actions on their unconnected constituents.

Here are just two examples of Federally supported “digitalization without inclusion” which raise important questions for research of this kind:

As part of its ongoing effort to promote electronic health records (EHRs), the Department of Health and Human Services ties a portion of healthcare providers’ Medicaid and Medicare reimbursements to their success in getting patients in these programs to use online personal health record applications. But of course, these older and poorer patients are precisely those most likely to lack Internet access. One leading healthcare sociologist has warned that “emerging inequality” in the uptake of personal health records “could increase or exacerbate health disparities.” [2] As medical practices are reorganized to take advantage of electronic tools for scheduling appointments, renewing prescriptions, sharing test results and patient-doctor communication –  due, at least in part, to HHS incentives – will patients without broadband find it harder to get access to information or care?

Some state agencies managing Federal Medicaid, SNAP, Unemployment Compensation and Workforce Investment Act dollars have aggressively migrated their client services – application, redetermination, compliance reporting, job searching and other workforce support — to Internet sites, and are either requiring or pressuring clients to use these sites (often by closing local offices and otherwise reducing access to face-to-face alternatives.)  The people most likely to need these Federally funded services include those least likely to have home Internet connections, including mobile connections. What impact is state and local digitalization of income and employment services having on unconnected households’ access to income support and job assistance?

These examples illustrate the broader point:  Each migration of services and resources to the Internet is a potential loss of access, another increment of digital exclusion, for millions who can’t go there.

NDIA believes that this perspective, and its consequences, deserve far more attention that they have received from Federal broadband policymakers and researchers, and we strongly recommend it as a focus of the National Broadband Research Agenda. Thank you for the opportunity to submit these comments.

Angela Siefer
National Digital Inclusion Alliance

[1]    American Community Survey 2015 1 Year Estimates, Table B28004