Play 2 – Collect data on broadband availability, affordability, adoption, and quality and use it to drive decision-making

Collecting data about the availability, affordability, adoption, and quality of broadband internet is critical for driving decision-making, understanding the existing barriers to broadband deployment and adoption, identifying the most appropriate long-term solutions, targeting resources, assessing progress, and building public support for broadband initiatives. There are existing federal data sets and mapping efforts that state education leaders may be able to utilize, , and the FCC is currently implementing changes to its data collection processes and systems to gather more granular information on broadband deployment. However, there may be other data that state or local education leaders may need to collect to pinpoint student household needs. Partnerships can be key for helping state education leaders analyze existing data and gather new data to inform planning.

Frequently Asked Questions

1. What type of data should we collect?

States are uniquely positioned to establish common data elements that can be collected across all local educational agencies (LEAs). Establishing common data elements will provide a comprehensive picture of the state of connectivity and allow for better targeting of resources. It will also help ensure that this data can be aggregated, in accordance with applicable privacy laws, at the state and national levels to ensure a coordinated response.

The Council of Chief State School Officers (CCSSO) developed the following list of recommended data elements in collaboration with SEAs, LEAs, and industry experts:¹ CCSSO is engaging with Student Information System (SIS) vendors to incorporate the data elements into their SIS products.

2. What existing data may be available to support this effort?

• Census Bureau American Community Survey (ACS) – Computer and Internet Use Data

The Census ACS data includes questions about the computers and devices that people use, whether people access the internet, and how people access the internet. Every year, over 3.5 million households across the country participate in the ACS. The ACS produces estimates on a wide range of geographies, including low geographic levels such as census tracts and block groups. States can use the 5-year ACS estimates (e.g., most recent covers 2015-2019) to identify which communities across their state may have low levels of computer or internet adoption in order to target resources and investments. The data can also provide insight into the most common types of devices (e.g., desktops or laptops vs. smartphones) and most common types of connectivity (e.g., wired vs. satellite vs. cellular), which can inform leaders about the quality of the access across communities. School districts can look at these same data points at the school-district level to provide a high-level snapshot of connectivity in their community.

• FCC Form 477 – Fixed Broadband Deployment Data

The FCC Form 477 data documents the providers (e.g., Comcast, Verizon), the technology (e.g., DSL, cable), the services (residential or business), and the maximum advertised upload and download speeds available by census block. Form 477 data is self-reported by ISPs twice a year. This data has been criticized for over-estimating broadband coverage because providers may not offer service to every home in every census block in which they report service. However, this data can be a useful tool for schools trying to identify a list of service providers that may serve their communities and could serve as potential partners. The FCC is in the process of implementing a new Broadband Data Collection that will result in more detailed and precise information on the availability of fixed and mobile broadband services.  The Broadband Data Collection program will give the FCC, as well as stakeholders, including schools, the tools they need to determine the availability and quality of broadband deployment on a location-specific basis.

• NTIA Internet Use Survey Data

The NTIA Internet Use Survey covers a range of topics related to digital inclusion and other internet policy issues, including the adoption of different types of devices and internet access technologies, locations of internet use (e.g., home, school, library), types of online activities, and the primary reason households are not online (e.g., cost, relevance). NTIA partners with the U.S. Census Bureau to administer the Internet Use Survey as a periodic supplement (e.g., most recently conducted in November 2019) to the Current Population Survey (CPS). The CPS is a monthly survey that includes approximately 50,000 households across all fifty states and the District of Columbia. States can use the NTIA Data Explorer to track internet use metrics overtime and see state-by-state comparisons or break down metrics by demographics. Pairing this data with the ACS Computer and Internet Use data can provide additional insight into issues of access, affordability, and adoption at the state-level.

• Internet Speed Test Data

Internet speed test data provides a snapshot of the performance, quality, and availability of internet service at a particular moment in time. Because of variations in home Wi-Fi systems and other factors that can affect the accuracy of measurements, such tests are most valuable as a measure of mobile broadband speeds, but they can also provide useful information as to user experience with their fixed broadband service. By measuring latency (or ping), and download and upload speeds, speed tests can confirm the speed of a connection and whether the service is measuring up to an ISPs advertised speeds. States are using openly licensed speed test data from organizations like the Measurement Lab (M-Lab) and Ookla alongside FCC Form 477 data to develop more accurate broadband coverage and performance maps in order to target local, federal, and state broadband investments to underserved communities.

3. What maps or tools already exist?

• ACS-ED Maps

The ACS-ED Maps tool identifies conditions of school-age children in school districts based on data from the U.S. Census Bureau’s American Community Survey Education Tabulation (ACS-ED). The map displays, by school district, the percent of households with a computer and the percent of households with a broadband internet subscription.

• NTIA Indicators of Broadband Need map

Layers in this map, which uses different data sources to show information on broadband availability and areas of need, were created using data sourced from the American Community Survey collected by the U.S. Census, Measurement Lab (M-Lab), Ookla, Microsoft and the Federal Communications Commission (FCC).  In addition, there is a layer in the map that displays the locations of higher education institutions eligible as Minority Serving Institutions (MSIs) under NTIA’s Connecting Minority Communities grant program. There is also a layer that shows areas designated as American Indian, Alaska Native and Native Hawaiian Areas by the U.S. Census in 2020. Reference the FAQ document or User Guide for more information.

• FCC Map of Fixed Broadband Deployment

The FCC’s interactive map of fixed broadband deployment depicts the census-block based fixed deployment data that providers report on the FCC Form 477. Users can search the map for information by provider and reported speed in an area. The Broadband Data Collection that the FCC is currently developing will provide more detailed and granular fixed and mobile deployment data than is currently collected through Form 477.

• Digital Divide Index

The Digital Divide Index (DDI) calculates a DDI score of 0 to 100 for each county or census tract. A score of 100 indicates the highest digital divide. The DDI is comprised of two scores: (1) the infrastructure/adoption (INFA) score and (2) the socioeconomic (SE) score. The INFA score is developed by examining five variables tied to infrastructure and adoption (e.g., percentage of homes without a computer or internet access; median maximum advertised download and upload speeds). The SE score is developed by examining five variables known to impact internet adoption. The DDI uses ACS 5-year estimate data and FCC Form 477 data. A community with a high INFA score may need to focus efforts on improving broadband infrastructure, while a community with a high SE score may need to focus on digital literacy training and internet adoption efforts. States can quickly see which communities have the highest DDI score and where additional investments and resources may be needed.

Play in Practice

New Mexico collects granular data to close the Homework Gap

In response to the COVID-19 pandemic, New Mexico formed the Homework Gap Team (NM HGT), bringing together leaders from across the state to take a holistic approach to addressing the Homework Gap.⁴ Starting March 2020, the state administered a series of surveys to better understand and address students’ home access issues.

1. The first survey asked school technology directors to identify how many student devices, such as hotspots and computers, might be needed. For example, the survey asked, “If funding became available for end-user devices (e.g., Chromebook, tablets), and you had some students that could benefit from that, how many would you ask for?” The results revealed that approximately 44,000 students needed broadband connectivity and another 55,000 students required learning devices.
2. A second survey was issued to students, families, and teachers to understand the quality of connectivity during distance learning. It asked questions such as, “Who is your Internet service provider?” and “My experience has been…” Respondents also completed a M-Lab Speed Test that measured upload and download speeds as well as their latency and retransmission. 

The survey data was used to target the rollout of several strategies to provide home connectivity to students including the purchase of 700 residential hotspots, 101 facility-based mobile hotspots, and over 6,000 student devices. The State also created the New Mexico Broadband Map⁵ to identify unserved and underserved locations across the state by pairing the NM HGT survey data with ISP coverage data by technology type. The interactive map allows users to input an address and see information about the available internet service options. It also maps public Wi-Fi parking lot hotspots throughout public, Tribal, and ISP locations. The New Mexico Broadband Map and the underlying survey data informed the development of the State of New Mexico Broadband Strategic Plan.⁶

Checklist & Key Questions

• Establish the objectives for the data collection effort.
  • What questions are you trying to answer about student home internet access and device access? Whether students have internet access? What type and speed of access? The costs of home access? The data caps associated with home access? Whether students have a dedicated learning device? Type of device? Can students access that device year-round or only during the school year?
  • How will your data collection align with and support your long-term goals for internet and device access and the use of technology to support learning?
  • How do you plan to use the data you collect?
  • How long will the data be retained before being destroyed?
  • How will the data be shared, if at all? Will it be used internally or displayed publicly via a map? Does this data sharing comply with applicable privacy laws?
• Identify existing data sets and data gaps.
  • What information is currently collected and what new information is needed to understand the scope of the challenge?
  • Are there other entities (e.g., state agencies, non-profit organizations, ISPs) collecting the needed data? Is it possible to establish data-sharing agreements to get access to this data? Do these data-sharing agreements comply with applicable privacy laws?
  • Is your state a partner in the NTIA National Broadband Availability Map? Can any data be accessed through this partnership?
• Identify what existing data collection tools already exist.
  • Are there existing state or district data collection efforts that already exist (e.g., surveys)?
  • Is it possible to add additional questions or data elements to these existing tools to collect the needed information?
• Identify what skills and expertise are needed and available to support the data collection effort (e.g., survey methodologists, data analysts, GIS experts).
  • Is the expertise available in-house or is a partnership with another state agency, institution of higher education, or other entity necessary to support the data collection?
• Design the data collection.
  • Is it possible to establish a set of consistent, standard questions or data elements?
  • Who needs to be engaged (e.g., school or district leaders, families, representation from diverse communities, Tribal and state agencies) to gain buy-in?
  • Have the questions been written in plain language and piloted with the anticipated respondents (e.g., school leaders, families) to ensure they are clear? Consider reusing question language from the NTIA Internet Use Survey or the ACS that have been tested by Census Bureau experts to help ensure clear and understandable questions.
  • Has the survey been translated into the most common household languages?
  • Is the survey accessible under Web Content Accessibility Guidelines (WCAG) 2.0 Guidelines?
  • Have a range of data collection methods been considered to ensure students and families from a range of circumstances can be reached (e.g., paper surveys, email, text, phone calls)?
  • Are there trusted partners that can support the data collection (e.g., school principals, community-based organizations)?
• Determine how data will be stored and accessed.
  • Is there a central online survey tool and data repository that the state can utilize to collate LEA data in a central location?
  • Where will the data be stored? What security features will be implemented?
  • Who will have access to the data? Will it be shared publicly? Does this data sharing comply with applicable privacy laws?
• Continuously update the data to monitor progress.
  • What are the key timelines that may impact how frequently the data must be updated (e.g., new student addresses; annual FCC Form 477 data)?
  • How often will the new data be collected? By whom?
• Visualize the data and share the visualizations publicly to help build public support.
  • Is there a state-level broadband map that currently exists?
  • Is there data from the education community that can be included on these maps (e.g., school locations; connectivity data for student households)?

1. Council of Chief State Officers, 2020. Restart & Recover: Home Digital Access Data Collection: Blueprint for State Education Leaders. [online] Retrieved from: https://ccsso.org/sites/default/files/2020-07/7.22.20_CCSSO%20Home%20Digital%20Access%20Data%20Collection%20Blueprint%20for%20State%20Leaders.pdf.

2. Chandra, S., Chang, A., Day, L., Fazlullah, A., Liu, J., McBride, L., Mudalige, T. & Weiss, D. (2020). Closing the K–12 Digital Divide in the Age of Distance Learning. San Francisco, CA: Common Sense Media. Boston, Massachusetts, Boston Consulting Group, available at https://www2.ed.gov/documents/coronavirus/reopening-2.pdf.

3. Hampton, K. N., Fernandez, L., Robertson, C. T., & Bauer, J. M., Broadband and Student Performance Gaps, James H. and Mary B. Quello Center, Michigan State University (March 2020), available at https://quello.msu.edu/broadbandgap/.

4. The team included leaders from the New Mexico Public Education Department, New Mexico Indian Affairs Department, Department of Information Technology, Department of Cultural Affairs, State Library, Santa Fe Indian School, Navajo Nation, Community Advocacy.

5. New Mexico Broadband Map: https://nmbbmapping.org/mapping/

6. State of New Mexico Broadband Strategic Plan and Rural Broadband Assessment: https://www.doit.state.nm.us/broadband/reports/nmbbp_strategic20200616Rev2Final.pdf