iPAC Publishes Broadband Quality in Public Libraries: Speed Test Findings and Results
Broadband speeds in U.S. public libraries have improved significantly in recent years yet continues to lag behind national broadband connectivity standards, according to “Broadband Quality in Public Libraries,” a new supplementary summary report released jointly today by the American Library Association (ALA) and the Information Policy & Access Center (iPAC) at the University of Maryland College Park as part of the Digital Inclusion Survey. A more detailed report, Broadband Quality in Public Libraries: Speed Test Findings and Results, is available from iPAC. The study examined the quality of broadband access in more than 2,200 public libraries by collecting data on upload and download speeds in 49 states.
Libraries reported progress in their public Internet speeds—nearly half of all libraries report subscribed Internet download speeds as being greater than 10 Mbps in 2013, compared with only 18 percent of libraries four years earlier. New speed test data collected from July-August 2014 found median download speeds of 30 Mbps for wired and 13 Mbps for Wi-Fi connections in city libraries to rural libraries clocking download speeds of 9 Mbps and 6 Mbps, respectively.
Despite the growth in increased broadband capacity, two-thirds of all libraries indicate that they would like to improve their broadband speeds. According to the 2013 Digital Inclusion Survey, just 2 percent of all libraries meet national benchmarks established by the Federal Communications Commission (FCC) that recommend a minimum of 100 Mbps for serving smaller communities and 1 Gbps for libraries serving populations greater than 50,000 people.
“Considering these speed test findings in context with multiple years of library technology data affirms that investments like those made through the Broadband Technology Opportunities Program are making a difference for libraries and the communities they serve,” said Larra Clark, deputy director of the ALA Office for Information Technology Policy. “As more library services and collections move online, there is urgency to accelerate these gains and address stubborn broadband capacity gaps where we find them. This understanding drives ALA’s ongoing advocacy with the Federal Communications Commission and others to meet 21st century learning needs.”
Lead researcher John Carlo Bertot adds, “We need more and better data related not only to available and subscribed internet speeds, but what is the user experience of this capacity. This new report is a start, but we know that results can be affected by a number of factors, including user device configuration, a library’s local network management, network load, and the design of the speed test tool. I would like to see the FCC expand its Measuring Broadband America initiative to include public libraries and give us more sustained data collection around broadband quality.”
Additional findings in the speed test report:
- City and suburban public libraries provide greater quality of service at the device level compared with town and rural libraries, and there is wide variation across libraries;
- Captured speed delivered to individual users’ devices is significantly less than the subscribed network speed;
- In most cases, quality of service degrades at peak use times, sometimes dramatically – for example, direct connection download speeds in city libraries are 69% lower during heavy usage vs light usage periods ;
- City/ suburban libraries report higher median speeds in their testing – for example, 3.5 Mbps download speed in city libraries compared to a median download speed of 9 Mbps in rural libraries – but also greater degradation during heavy internet usage times than town and rural libraries; and
- Captured (and subscribed) upload speeds lag download speeds considerably, impacting libraries’ ability to support emerging services like digital media labs and other user content creation and dissemination.
The speed test study was administered by the Information Policy and Access Center at the University of Maryland as a supplement to the Digital Inclusion Survey, funded by the Institute of Museum and Library Services. The supplemental speed test research adds to the body of knowledge about the quality of public access in public libraries. The 2014 Digital Inclusion Survey Report, to be published in summer of 2015, will help inform how public libraries are helping to build digitally inclusive communities for all, with interactive mapping tools, issue briefs, infographics and state-level data summaries.
“Digital inclusion resources and services, including high speed Internet, provide people with opportunities to change their lives,” said Kathy Rosa, director of the ALA Office for Research & Statistics. “Released during National Library Week (NLW), the study results provide data that support the NLW theme: "Unlimited possibilities @ your library®”. This is a time to celebrate libraries that foster digital inclusion in the community. “
The speed tests were fielded between July 14 and August 15, 2014. A total of 2,251 public libraries participated, answering questions on the library’s subscribed upload and download speeds, the number of public access computers at the location, and to run the test multiple times – indicating with each test whether the device was directly connected or connected via wi-fi.
Libraries ran the speed test tool on public access devices - including mobile devices, wireless devices, and directly connected computers – using a version of Ookla licensed by the study. For an understanding of how broadband quality changed by time and by load on the system, libraries were asked to run the speed test when the library was closed, during regular hours of operation, and when usage was light, normal, or heavy, by the librarians’ estimation.
Speeds in the test are approximate; true network speeds are affected by numerous factors such as user device, network configuration, arrangements for boosts or “bursts” of broadband speed with the library’s Internet Service Provider, the location of the speed test server, the design of the speed test tool, and other factors.
For more information on the ongoing study: