Public libraries were early adopters of Internet-based technologies, and the Public Library Funding and Technology Access national surveys began in 1994¹ with the purpose of tracking the growth of public library Internet connectivity and uses as a basis for (1) proposing and promoting public library Internet policies at the federal level; (2) maintaining selected longitudinal data as to the connectivity, services, and deployment of the Internet in public libraries; and (3) providing national estimates regarding public library Internet connectivity. Through 2004, the surveys were conducted roughly every two years. Beginning in 2006, the surveys switched to an annual data collection cycle.

Though the survey remains true to its primary goals, it evolved over time and experienced three clear shifts in data collection, methodology, and approach:

• Prior to 1998, the surveys collected data at the system level (i.e., total number of workstations across all library branches, if applicable).²
• Between 1998 and 2004, the surveys collected data at the building or outlet level (i.e., number of workstations in a particular branch, speed of connectivity at the branch), as well as system-level data (i.e., E-Rate applications).
• Beginning in 2004, the surveys expanded to collect data at the state and national levels and include both building or outlet level and system-level data.
• Beginning in 2002, the survey offered participants a fully online version of the survey as well as a printed version of the survey to complete. Each year, more surveys were completed online, and in 2009, the survey became an online-only survey.

Throughout these shifts, the survey has maintained core longitudinal questions (e.g., numbers of public-access workstations, bandwidth) but consistently explored a range of emerging topics (e.g., jobs assistance, e-government, emergency roles).

Due to the survey's longevity, longitudinal data, and unique data, data from the surveys have appeared over the years in Congressional testimony, filings with the Federal Communications Commission (FCC), filings with the National Telecommunications and Information Administration (particularly regarding the recent Broadband Technology Opportunity Program grant program), the Children's Internet Protection Act US Supreme Court decision, US Senate hearings on the E-Government Act, and many other critical policy venues. State librarians have also used the data in state legislative testimony and in a range of state policy documents and initiatives. In short, the data from the surveys are used by a number of stakeholders in a wide range of ways.

This article seeks to provide an evolutionary perspective on public library Internet connectivity. A full list of all the Public Libraries and the Internet and Public Library Funding and Technology Access studies and their findings is included in Appendix A. Unless otherwise noted, all data discussed in the article are from these studies. All of the reports are also available in electronic format at http://plinternetsurvey.org.

The survey's methodology has evolved over time to meet changing survey data goals. As of this writing, the survey provides both national and state estimates to

• provide branch-level national data regarding public library Internet connectivity and use;
• provide state branch-level data (including the District of Columbia) regarding public library Internet connectivity and use;
• provide system (administrative)–level data (including the District of Columbia) regarding E-Rate use and library operating and technology funding and expenditures; and
• include assessment questions for selected public libraries that are recipients of certain Bill & Melinda Gates Foundation (Foundation) grants.

The last objective is beyond the scope of this article.

The survey has additional objectives of obtaining data to conduct analysis using metropolitan status (e.g., urban, suburban, and rural)³ and poverty (less than 20% [low], 20% to 40% [medium], and greater than 40% [high])⁴ variables. Over the years, the poverty variable has not demonstrated any statistical significance in terms of the survey's findings, and thus the poverty variable was removed beginning with the 2009–2010 survey.

The survey uses a stratified “proportionate to size sample” to ensure a proportionate national sample. The sampling approach taken ensured high-quality and generalizable data within the states analyzed, nationally, and across and within the various strata. The study team uses the Institute of Museum and Library Services (IMLS) public library dataset (formerly maintained by the US National Center for Education Statistics) to draw its sample. Foundation grant recipient data are overlaid on the national library dataset. The survey asks respondents to answer questions about specific library branches and about the library system to which each respondent branch belongs.

Respondents typically answer the survey between September and November of each survey year. Changes in technology have fueled not only the evolution of the Internet in libraries, but the evolution of the survey as well. The development of Web-based surveys and more powerful data analysis tools have facilitated the growth of the survey in terms of reach to libraries, ease of completion, complexity of questions, and depth of statistical analysis.⁵

In each year of the survey except for the 2006–2007 survey, the survey response rate is between 70.0 percent and 86.5 percent, and provides between 5,500 and 8,400 survey responses. The data are weighted for both national- and state-level analysis and have a margin of error or +/– 2 or 3.5 percent, depending on survey year. The high survey response rate and representativeness of responses demonstrate the high quality of the survey data and the ability to generalize to the public library population.

This section provides an overview of selected longitudinal data across the survey years. It is important to note, however, that key survey questions, such as broadband connectivity speeds, have changed substantially over the years to reflect the evolving nature of Internet connectivity. For example, the first surveys asked about dial-up connections and their speeds, versus today's questions that explore fiber-optic and other types of Internet connectivity and corresponding higher speeds. Thus some longitudinal comparisons would not make sense. Finally, the section provides selected findings regarding newer services, particularly as they have an impact on future public library Internet-enabled services.

Increasingly, the public library Internet surveys show conflicting results. This section focuses on these somewhat contradictory results, by way of findings reported in the latest (2010–2011) survey. On the one hand, public libraries continue to offer enhanced public-access computing and Internet access to their communities. As presented above, public libraries

• offer wireless (Wi-Fi) access to the Internet;
• have faster public-access broadband speeds;
• offer technology and Internet training; and
• offer a range of other services such as databases, digital reference, e-books, and e-government.

And, as libraries report, almost 65 percent are the only free public-access computing and Internet centers in their communities (see figure 2.6). Thus, public libraries provide critical public-access computing and Internet services that support their communities in a wide range of areas.

On the other hand, however, public libraries indicate that:

• Their broadband speeds are inadequate. 44.9 percent of respondents reported that their connectivity speed is insufficient some or all of the time (see figure 2.7).
• Their numbers of public-access computers are inadequate. 76.2 percent of libraries reported that they have insufficient availability of workstations some or all of the time (see figure 2.8).
• Costs, space, and buildings are real barriers to the public-access environment public libraries can offer. 78.7 percent of libraries reported that cost factors, 77.2 percent reported space limitations, and 54.4 percent reported that the building infrastructure (e.g., cabling, wiring, electrical outlets) were important factors in their decisions to add, or not add, public-access workstations or laptops (see table 2.4).
• They rely on nonprofessional IT staff for technology support. Libraries report that they rely on non-IT public service staff (39.0 percent), library directors (31.7 percent), and sometimes even volunteers (6.7 percent) for support for their technology. The percentage of library directors providing IT support climbs to 47.6 percent for rural libraries and drops to 4.2 percent for urban libraries. 48.1 percent of libraries support their IT with system-level IT staff, but only 32.2 percent of rural libraries have access to such support as compared to 75.5 percent of urban libraries (see table 2.5).

The real significance of these findings is that some libraries continue to face the same challenges in spite of upgrades to their technology infrastructure. And, more significantly, libraries continue to offer a significant number of services to the communities that they serve—licensed databases, technology training, e-government, and more—while often being the only free public-access point within their communities.

Thus, public libraries increasingly report that they are unable to meet patron demands for services due to inadequate technology infrastructure, costs associated with operating and maintaining that infrastructure, and bandwidth quality or availability issues—but not for lack of trying to enhance their services. What is unclear is how libraries will maintain their levels of public-access computer and Internet access services, much less extend and augment them, in the current economic downturn. It is in this mixed and paradoxical context that public libraries provide their public-access services.

The surveys have demonstrated the enthusiastic embrace and incorporation of the Internet and public-access technologies by public libraries—not just from an infrastructure perspective, but also from a service and resource perspective. On average, public libraries increased the number of public-access workstations by several hundred percent in a period of eight years; substantially increased their Internet speeds; expanded service to include Wi-Fi public access; offer a large range of Internet-enabled services and resources such as databases, digital reference, and e-books or audiobooks; and provide technology and Internet resource training services. But the data also show that libraries are stretched, and increasingly challenged to maintain or enhance their levels of services.

This strain has been ongoing for several years and continues to become more pronounced for several reasons:

• As demonstrated as early as the 2006 survey, many libraries began struggling with an “infrastructure plateau” a number of years ago as they no longer had space, connectivity, and other supports to expand the number or capacity of computers available to patrons⁶
• Public libraries are increasingly the only access point in communities for free Internet and assistance using the Internet.⁷
• The Internet both has served to augment existing library services and to establish new social roles, with e-government and emergency response being two of the most prominent new roles.⁸
• Many other types of outlets for information that the public was able to previously rely on—such as government agencies, academic libraries, and law libraries—have moved their information exclusively online or become more exclusive in their service populations, driving people without other access to the public library to get information previously available elsewhere.⁹
• The sustained economic downturn has exacerbated this demand on libraries.

The last point is critical, and likely will have lasting effects on the public library service environment. As the study shows ongoing reductions to public library budgets and funding, the economic conditions have caused more people need to rely on the library computers out of necessity for looking for assistance and seeking employment, as well as entertainment and communication, as many people can no longer afford their own access, all while most libraries were suffering significant budget cuts.¹⁰ Public libraries have served as “America's first responders to the economic crisis.”¹¹ However, as the usage rates of libraries and the demands on their computers have increased dramatically—in some cases as much as 500 percent in a period of three years—the library staff, computers, and even the most basic infrastructure like carpets have struggled to hold up under the usage and needs.¹²

In light of the ever-increasing demands being placed on the computers, connectivity, and staff of public libraries, public libraries, policy makers, and others need to consider several key issues as public libraries continue to fulfill their role as community-based providers of cost-free public access to the Internet and computing:

• How much is enough? It's never enough. A question that the surveys have never been able to adequately address is “How much access (workstations, broadband) is enough?” In fact, as the bar gets higher in terms of service provision, so too does the assessment of how much is needed. Libraries—even the smallest in the most rural areas—are rapidly approaching the need for fiber-optic connections. Nearly half of all libraries report a T1 (1.5 Mbps) connection—something that only a few short years ago was considered robust bandwidth. And yet, a vast majority of libraries report that this increase in bandwidth is inadequate.
• Better understanding of the relationship between infrastructure and services. The initial Public Libraries and the Internet surveys showed that libraries viewed Internet connectivity as an experimental service—one that had substantial potential, but it was unclear at the time (the Mosaic browser was introduced in 1993, the same year that the first survey went into the field) just how revolutionary the Internet would be to public library service. As Internet-enabled services are a mainstay of the public library, there is a substantial need to better understand which services require what amount of bandwidth. Increasingly, for example, streaming video content is in high-definition format, which consumes substantially more bandwidth than Web browsing.
• The need for comprehensive capacity planning. Public-access services and resources require libraries to look across their Internet-enabled services and resources comprehensively. Public-access workstations, broadband, and Wi-Fi are part of a collective public-access technology environment that directly impact the ability of libraries to offer patrons high-quality Internet services and resources—and moreover, high-quality user experiences. A library that has seven public-access workstations and offers Wi-Fi but has a DSL connection ultimately provides a dial-up experience to its users. Capacity planning needs to include not just the last mile, but also internal library infrastructure, including routers, switches, and up-to-date workstations, at the least, to provide quality public-access services.
• Continual upgrades to technology and staff. As the surveys demonstrate, public-access computing and Internet infrastructure and services are not a one-time investment. There is a continual need to upgrade computing technology, Internet connectivity, and buildings. Also, the surveys show that, given the demands placed on libraries for training, e-government, education, employment, and other critical service areas, there is a need to continually train library staff on a range of technologies as well as services (such as how to help with applying for government benefits, seeking employment, taking certification exams, and more).
• Setting service quality benchmarks. Given increasing demands, and libraries that report the inability to keep up with demand, libraries may need to consider setting levels-of-service-quality benchmarks. Libraries will need to decide whether they will offer as many services at the highest level of quality as possible, or set levels of service quality, realizing that a library may not be able to meet all requests and demands. In some cases, service quality levels may be dictated by, for example, the broadband that is available to a library due to cost and geography. These levels of service may place libraries at odds with their user communities, which may request or desire more services at levels of quality that the library is unable to afford or provide.
• Public-access technology versus other services. Internet-enabled services have clearly become essential public library services. It is important to also note that libraries continue to offer and provide significant services such as programs and access to printed materials. Public libraries reside, and have for some time, in both print and electronic environments and this requires libraries to strike a balance between these two contexts, particularly in light of resource constraints.
• Library physical capacity limitations. Libraries have consistently identified the limitations of their buildings through the surveys since 2006. Indeed, public library space will remain a critical factor as libraries balance the use of their physical space to house materials, provide public-access technologies, and facilitate public use. The IMLS public library data show that 25 percent of public libraries are 2,400 square feet or less, with 50 percent of public libraries having 5,700 square feet or less¹³—and one should note that the usable space for collections and services is different from overall space. And though offering wireless access increases public-access technology capacity, the need for differing workspaces to accommodate mobile devices and the power needs of such devices presents an issue for libraries.
• The library divide. One factor across all the surveys conducted in the last ten years remains: rural libraries in general have fewer resources, less connectivity, fewer workstations, less access to technology support, and other factors. This does not mean that urban libraries are infinitely better off—in fact, urban libraries often report similar issues in terms of keeping up with demand. But the survey data are clear: rural libraries face substantial challenges in supporting their public-access technology environment, and there is no indication of abatement in this circumstance. Additionally, long-term funding decisions have created notable divides in overall workstations and connectivity between regions of the country.¹⁴

The above are a selection of issues that will need consideration and resolution if public libraries are to continue their role as critical providers of community public Internet and computing access.

Future Public Libraries and the Internet surveys will need to continue exploring the evolving service context of public-access Internet and computing services within public libraries. Public libraries have clearly moved beyond issues of “getting connected” and into critical services provision via the Internet. Indeed, one cannot imagine a public library that is not connected to the Internet, not providing public-access computers, or not offering users a number of online resources such as licensed databases, assistance with using technology or the Internet, or, increasingly, Wi-Fi access. One can no longer separate the public library from public Internet access. The issue is not one of measuring connectivity, but rather of better understanding the nature and roles of public libraries as providers of community-based public access.

American Library Association. Libraries Connect Communities: Public Library Funding and Technology Access Study: 2006–2007. Chicago, IL: American Library Association, 2007.

———. Libraries Connect Communities: Public Library Funding and Technology Access Study: 2007–2008. Chicago, IL: American Library Association, 2008.

———. Libraries Connect Communities 3: Public Library Funding and Technology Access Study: 2008–2009. Chicago, IL: American Library Association, 2009.

———. Libraries Connect Communities: Public Library Funding and Technology Access Study: 2009–2010.Chicago, IL: American Library Association, 2010.

———. Libraries Connect Communities: Public Library Funding and Technology Access Study: 2010–2011. Chicago, IL: American Library Association, 2011.

Bertot, John Carlo, and Charles R. McClure. Moving toward More Effective Public Internet Access: The 1998 National Survey of Public Library Outlet Internet Connectivity. Washington, DC: National Commission on Libraries and Information Science, 1998. Available at http://plinternetsurvey.org.

———. The 1998 National Survey of US Public Library Outlet Internet Connectivity: Final Report. Washington, DC: National Commission on Libraries and Information Science, 1998. Available at http://plinternetsurvey.org.

———. Public Libraries and the Internet 2000: Summary Findings and Tables. Washington, DC: National Commission on Libraries and Information Science, 2000. Available at http://plinternetsurvey.org.

Bertot, John Carlo, Charles R. McClure, Kristin M. Barton, Susan Thomas, and Jessica McGilvray. Public Libraries and the Internet 2007: Survey Results and Findings. Tallahassee, FL: Information Use Management and Policy Institute, 2007. Available at http://plinternetsurvey.org/

Bertot, John Carlo, Charles R. McClure, and Patricia Diamond Fletcher. The 1997 National Survey of US Public Libraries and the Internet: Final Report. Washington, DC: American Library Association, Office for Information Technology Policy, 1997. Available at http://plinternetsurvey.org.

Bertot, John Carlo, Charles R. McClure, and Paul T. Jaeger. Public Libraries and the Internet 2004: Survey Results and Findings. Tallahassee, FL: Information Use Management and Policy Institute, 2004. Available at http://plinternetsurvey.org.

———. Public Libraries and the Internet 2006: Survey Results and Findings. Tallahassee, FL: Information Use Management and Policy Institute, 2006. Available at http://plinternetsurvey.org.

Bertot, John Carlo, and Charles R. McClure with Kim M. Thompson. Public Libraries and the Internet 2002: Internet Connectivity and Networked Services. Tallahassee, FL: Information Use Management and Policy Institute, 2002. Available at http://plinternetsurvey.org.

Bertot, John Carlo, Charles R. McClure, Carla B. Wright, Elise Jensen, and Susan Thomas. Public Libraries and the Internet 2008: Survey Results and Findings. Tallahassee, FL: Information Use Management and Policy Institute, 2008. Available at http://plinternetsurvey.org.

———. Public Libraries and the Internet 2009: Survey Results and Findings. College Park, MD: Center for Library and Information Innovation, 2009. Available at http://plinternetsurvey.org.

Bertot, John Carlo, Charles R. McClure, and Douglas L. Zweizig. The 1996 National Survey of Public Libraries and the Internet: Progress and Issues. Washington, DC: National Commission on Libraries and Information Science, 1996. Available at http://plinternetsurvey.org.

McClure, Charles R., John Carlo Bertot, and Douglas L. Zweizig. Public Libraries and the Internet: Study Results, Policy Issues, and Recommendations. Washington, D.C.: National Commission on Libraries and Information Science, 1994. Available at http://plinternetsurvey.org.