Integrating Public Library Databases onto a Nationwide Digital Base Map

For Enhanced Library Services and Research

NATIONAL IMPACT

Introduction

National digital inventories of accurate locations and critical external information are available for major public agencies such as hospitals and emergency health care, fire and police, and schools.[1] These databases are regularly utilized by policy and decision makers at all levels for administration and management, long and short range planning, regulatory decision making, technology evaluations, and most recently for national security and emergency preparation.[2] At present, there is no comparable and integrated national database of public library statistics and other related information in the United States.

While a nationwide inventory of public library outlets and administrative entities has existed for several years through the Federal-State Cooperative System (FSCS) and National Center for Education Statistics, U.S. Department of Education (NCES/U.S. DOE), the inventory is not utilized regularly for serious decision making at the local, regional, or national levels. This can be attributed to several reasons – including a significant lag time between the collection and standardization of the data (especially at the national level) and its availability to decision makers. In addition, there has been no accurate base map of public library locations developed from the inventory.[3] Due to lack of a base map, detailed mapping of additional new information and data for specific public library outlets or administrative entities could not occur.

In May of 2001, there was a first step to rectify this lack of an accurate inventory and base map of public library outlets. The National Commission of Library and Information Science (NCLIS) and NCES/U.S. DOE contracted with the GeoLib Program of the Florida Resources and Environmental Analysis Center (FREAC) at the Florida State University (FSU) to develop a geographically accurate digital base map for all (an estimated 16,000 outlets, not counting bookmobiles) 1999 FSCS public library outlets. This digital public library base map will be finished in February 2002.

The value of this national inventory and ‘first ever’ base map for U.S. public library outlets cannot be underestimated.[4] Before it can be used for serious decision making, however, the inventory must be linked to other data, as well as to update and maintain the inventory. This demonstration project will integrate key public library data files onto the national digital base map of public libraries to facilitate decision and policy making regarding public libraries.

Need for Integrating Public Library Databases onto a Digital Base Map

Today, public libraries and policy makers must address an ever-growing number of decisions and issues at the local, regional, and national levels. Daily questions face library managers and policy makers such as:

• Is the library providing the right balance of library services in light of changing populations?
• How should funding levels of local library service adjust to these changes?
• How can public libraries bridge the digital divide through provision of equitable access to the Internet?
• What is the relationship between distance (i.e. to a facility, user travel time), and library use?
• What are the effects of poverty and receipt of E-rate discounts on library services?

The answers to these and many other public library-based questions can often be best addressed (or even only answerable) by nationwide databases linked to a geographic information system (GIS) base map. A key ingredient to such a database system is the incorporation of a broad range of consistently acquired and developed data that are of relevance to public library decision and policy making. Such a nationwide public library GIS database system does not exist today.[5] This project is the critical first step towards creating one.

Overview and Applications of the Initial Databases and Digital Base Map

The term “nationwide public library database system” or simply “database system” is used below to refer to the product of this project, a collection of databases of relevance to public libraries that are each tied to the nationwide digital base map. The term “database” by itself is used below to refer to the individual data sets that will comprise the nationwide public library database system. There may be some integration of the data from several different existing databases into a single, more user-friendly database for public librarians and researchers if such an integration is warranted.

There are many possible applications of the initial public library database system for researchers and policymakers as well as local public librarians. This project’s databases will be built upon the geographically-corrected and documented digital base map of public library administrative entities and associated outlets as described in Appendix B.

Data sets that the researchers will review and evaluate for inclusion and integration into the nationwide public library database system to be available at the end of the second year of this project, and the applications envisioned, include the following:

1) The U.S. Census Bureau of the Census 2000 Data

Census data are collected at varying levels of geography, ranging from national to state, regional, county, city, and census tracts down to block groups and blocks. Local geographies (demographics) extracted from census blocks and block groups are extremely valuable for public libraries, as an estimated 80% of a public library’s funding comes from the local sources. Whenever possible, the database system will maintain US Census data at the most detailed geographic level, the census block.[6] However, some of the census-type data in the national public library database system will cover larger geographic areas such as census tracts[7] because an ever-growing number of governmental data sets use larger boundaries than census blocks.

Regardless of the geography covered, many of the decisions and analyses needed for public library decision making will require U.S. Census data. Census data provide the most consistently-collected information for describing local areas--neighborhoods, cities and counties--which are the traditional ‘geographic homes’ of America’s 17,000 public libraries. Koontz (1997) identified selected census data that are of most relevance to library use including population, population density, racial composition, age, education and income levels (high, low and poverty). Select sets of relevant census data will be part of the initial national databases.

Census data are collected at a single point in time. Yet to be most useful, the information should reflect the dynamics of change over time (Myers, 1992). This project’s database system will be developed with this requirement in mind, potentially enabling users to examine not only current public library data but historical and projected data as well.

2) National Center for Education Statistics/Federal State Cooperative System (NCES/FSCS), U.S. Department of Education Public Library Use Data

The NCES/FSCS annually coordinates a public library survey through a network of state data coordinators appointed by the chief officers of state library agencies (represented by COSLA/ALA) in the 50 states, District of Columbia and the outlying areas. These data include: population of legal service area, service outlets, library materials, staff, operating income and expenditures, circulation, reference transactions, library visits, public service hours, circulation of children’s materials and electronic technology information.

In addition to the above two data sets, the principal investigators for this project have already identified other data sets that are likely to be of high interest. The ability to include these data sets into the database system by the end of this IMLS project will be dependent upon an in-depth analysis of the ease and costs of integrating the data sets into the overall database system. These data sets include:

1) The Information Institute (FSU) data sets

The Information Use Management and Policy Institute (Information Institute) at the School of Information Studies, Florida State University, regularly collects a range of sample data describing networked services and activities of public libraries. These national surveys[8] provide important data that describe the connectivity, bandwidth, service capabilities and other factors related to public library services. Currently the databases at the Information Institute are not integrated with any other national databases. A high priority of this project will be to integrate the Information Institute’s data into this project’s national public library databases and digital base map.

2) Relevant Geographic Boundaries

These boundaries may include but are not exclusive to: library legal service areas, county and city geographic boundaries, congressional and local voting districts, school districts, and traffic analysis zones. Identification of public library locations within these, and populations served, is critical to local and state government planning for assessing equity of public fund allocation and impact of those funds; gaps in service related to library facility distribution; distance between library and school facilities for cooperative support; and public transportation availability for equitable access.

3) Important Geographic Features that Affect Library Use

These include roads, railroad tracks, and rivers, large industrial and commercial sites, parks, etc. It is critical to acknowledge boundaries such as these, that limit some potential users from accessing library materials and services. These are often ignored in county or city long range planning and can be effectively displayed in this database.

4) Location of E-rate library applicants and application information[9]

Information regarding E-rate discounts for public libraries is of critical need now, as well as in recent years. Because of lack of relevant library information, the FCC and Congress utilized free-lunch program data from public schools to set the telecommunications discount rate for which a public library was eligible. But this poverty measure ignored poverty among adults with no children as well as the elderly populations served by library outlets. If use of this national database system allowed just five percent of public library outlets to claim a larger telecommunications discount rate than that calculated through use of school lunch program data, millions of dollars worth of E-rate awards may be made available to public libraries in future E-rate years.

5) A map of Library Services and Technology Act funding level by location

The Government Performance and Results Act (GPRA) of 1993 requires every government agency (including IMLS) to establish objectives and measurable performance goals for all programs. All states must report the impact of federal LSTA funds (which all states receive for public library building and technology improvements) every five years. The availability of the national public library database system will greatly improve the ease, usefulness, and impact of such measurements and assessments. Over time the information can create a benchmark of data upon which communities can ask for specific LSTA funds to build upon previous projects and programs, and forecast the local library’s needs more optimally.

Further, the IMLS manages LSTA funds. Assuming a $10,000 savings per state (a conservative estimate) in data acquisition and analysis cost for each five-year LSTA report, there is an estimated savings of $500,000 per five-year cycle as a result of having access to and using this project’s nationwide database system.[10] State library agencies would accrue further savings during annual update cycles. The IMLS could disperse to libraries the savings this project offers for other high-priority library needs.

Other data sets will be included as these become available and are assessed by the study team and proposed advisory committee, as appropriate, for inclusion in the national database system, e.g., the Public Library Association’s The Public Library Data Service (PLDS) which is collected annually on a sampling basis. (See discussion in the Adaptability section of this proposal for greater detail on this topic).

ADAPTABILITY

The most basic tenet for this database system is that it will be flexible and adaptable enough to be of use to a wide range of potential stakeholders. The establishment of an advisory committee during this project is the first step to identify types of data needed. Over the long-term, the database development will involve consultation and collaboration by GeoLib/FREAC database development professionals, and Information Institute personnel with library and information science professionals from throughout the U.S. in a variety of private and public institutions. The database system will be dynamic, allowing for incorporation of other data sets identified and needed by others, and that meet technical requirements for inclusion. This dynamism will ensure that the system can be adapted to suit the unique needs of many others. The study team will consider a number of user communities and uses as it develops the national public library database system:

• Local public library administrative entities and outlets, to conduct community analyses and facility master plans;

• Federal and state policy makers, to assess library funding;
• State libraries, to facilitate LSTA funding and meeting IMLS criteria for state five-year plans;
• State data coordinators for FSCS, to better track changes in population attribute variables for annual reports to the NCES;
• The library research community, library schools and public library consultants;
• The Institute of Museum and Library Services (IMLS) itself, to 1) identify national library needs based upon changes in the public library’s environment or to 2) evaluate the effects of its funding programs on libraries within specified geographical areas of the country; and
• The larger academic community, such as political scientists, urban and regional planners, sociologists, and others who will finally have a nationwide database upon which to include libraries in said research discipline.

As the study progresses, the study team will assess the feasibility of meeting the needs of these various user communities and uses given the resources provided. The organizational structure and technical methodology used in developing this database system will be documented and downloadable from the GeoLib website (http://www.geolib.org) as part of this project and will be made available to others who may wish to utilize and adapt the information into their own customized projects.

PROJECT DESIGN

The goal of this project is to begin development of a dynamic nationwide public library database system with a digital base map to spatially describe and facilitate the analysis of public library services and programs at all governmental levels. The objectives of this project are:

• Integrate currently developed and relevant national data sets into a nationwide public library database system;
• Develop and maintain an up-to-date national database of all public library outlets that is geographically accurate with respect to location;
• Review the feasibility of developing a protocol and methodology for incorporating other public library data sets and information into the national database system as identified and prioritized by the research community, policy makers, public librarians, library managers, and public library users;
• Longer-term, as data sets continue to be integrated into the database system, be a “one-stop” Internet access point for public library researchers and library managers with an integrated data source for library projects and planning.

Together, these objectives will drive the development of the database system, as well as the system’s design, availability, and use.

Project Schedule and Tasking

The study will begin October 2002 and be completed September 2004. The Schedule of Completion provided as part of this proposal provides an overview of the schedule and the timing of key tasks for the project.

The tasks for completing this project can be divided into eight categories. One category of tasks is the establishment and holding of advisory committee meetings. The role of the advisory committee will be to provide feedback and guide the project. The researchers anticipate having a small committee of five to six outside members with representatives from public library administrators, public library researchers, and public library policymakers. Two meetings of the advisory committee are planned, during this project, all coinciding with American Library Association conferences.

The second category of tasks is the development and completion of needs assessments and studies. This includes the identification of other significant data sets that may be incorporated into the initial release of the national database system and the need for those data sets among potential stakeholders. Having identified the data sets that will comprise the initial database system, the second study will identify potential database applications that would be of interest to stakeholders. These studies will involve a survey of a wide range of potential stakeholders throughout the entire U.S. and will occur during the first year of this project.

The third category of tasks will be to establish a project website. The website will be updated and maintained throughout the duration of the project. Work on these first three task categories can begin almost immediately in October of 2002. The advisory committee meetings and the web site maintenance, of course, will continue to occur throughout the duration of this project.

The fourth category of tasks is the development of the initial version of the national public library database system. Some preliminary work can begin fairly quickly but the bulk of the work cannot begin until after the first study on data sets is completed. Intensive development of a first-release database system can begin approximately six months after this project is initiated, with the first public access being available approximately one year after this project is begun. Based on usability studies conducted at the Information Institute, data refinements, and end-user access statistics, the first-release database system will be revised and a final version of the initial national public library database system will be accessible over the Internet about 22 months into the project.

The fifth category of tasks is associated with testing the usability of the database system. These tasks cannot begin until almost one year after project initiation (i.e., after the first release of the database has been made). The findings from these usability tasks will be used to improve the final release of the initial database system.

The sixth category of tasks is working with COSLA to develop a timely method in which to receive public library administrative entity and outlet changes directly from the states themselves rather than through the NCES. These tasks can begin after the first advisory committee meeting.

The seventh category of tasks will begin near the end of the first year, with the most intensive effort still being concentrated towards the end of the project. All of the tasks in this category are related to ensuring the sustainability and value of the developed national public library database system after IMLS funding ends. Protocols and methodologies for incorporating new databases will be developed as well as the development of a cost sharing mechanisms among all stakeholders to continue the development and maintenance of the national database system.

The eighth and final category of tasks is related to dissemination of the database and the evaluation of the progress and success of the project.

MANAGEMENT PLAN

The project will be managed jointly by FSU’s GeoLib Program within the Florida Resources and Environmental Analysis Center (FREAC) and the Information Use Management and Policy Institute within the School of Information Studies, all within FSU. The Co-Principal Investigators (Koontz, Jue, McClure, and Bertot) will comprise the study team and will jointly management the project. Dr. Koontz will serve as overall project director.

Although most project tasks will be the shared responsibility of both groups (see Appendix X), the management approach will take advantage of the strengths within each program. The GeoLib Program within FREAC, with its experience in public libraries, GIS, database applications development, and census data, will take primary charge of the development of the national public library database system and identification of potential database applications of the database system. The Information Institute, with its experience in library data needs and library assessments and evaluations, will take primary charge of the needs assessment to assist in developing the content of the database. The Institute will also engage in usability testing of the database, fine tuning the selection of data elements to best meet the decision making needs of the library community and others, and evaluation of the project itself.

PERSONNEL

As can be seen in the proposal section containing the resumes for all key project personnel, all permanent staff for this project are senior personnel within their respective program, center, or institute, usually with over a decade of experience in their respective area of expertise.

Dr. Christie Koontz is director of GeoLib, a research program within FREAC, utilizing geographic and marketing research, for the past decade, to solve public library problems. Dr. Koontz's areas of expertise include using GIS software for facility location and siting, and market determination and demographic profiling. She has an emphasis on use of public libraries in majority-minority and rural and low income market areas. Mr. Dean Jue, has been active in the public library research arena since 1995, contributing expertise in public policy and computer science. Mr. Jue has co-directed several national library projects with Dr. Koontz. He is currently the project manager and co-director with Dr. Koontz, of the NCLIS/NCES national public library geo-mapping project developing the digital base map of all the 1999 FSCS public library outlets and administrative entities upon which this project’s database system will be referenced.

Dr. Charles R. McClure is the Francis Eppes Professor of Information Studies and the Director of the Information Institute. He has written extensively on topics related to the planning and evaluation of public libraries. He is a co-principal investigator with Dr. John Bertot on a research study developing new national models for public library data collection, analysis, and reporting. Dr. McClure has also served as co-principal investigator with Dr. Bertot on a study assessing the impact of LSTA, E-rate, and Gates Awards on libraries. Dr. John Carlo Bertot is an associate professor of Information Studies at FSU and also Associate Director of the Information Institute. He is co-author, with Dr. McClure, of the recently published book Statistics and Performance Measures for Public Library Networked Services (ALA, 2000). Dr. Bertot was also the principal investigator in obtaining and analyzing E-rate data from the School Library Division.

PROJECT EVALUATION AND OUTCOMES

The evaluation of this project will be multi-faceted due to the broad impact of the database and will consider the outcomes of this project on individual library outlets and/or administrative entities, the library research community, the state libraries, and the federal policy-making community. The evaluation will provide both a formative evaluation as well as a summative assessment of project activities and outcomes.

While the study team will assess, develop, and implement a number of evaluation techniques as part of the study, below are possible evaluation activities in which the study team can engage to determine the impact of the database system:

• Web log analysis. The primary means of access to the initial nationwide public library database system will be through the web. As a result, it will be possible to track uses of the databases via log analysis (Rubin, 2001). In and of itself, web log analysis can provide a variety of evaluative data regarding database use.
• Survey. The researchers would take the results of the web log analysis to indicate which databases are used most frequently along with the more prevalent actions in which users engage in those databases. Based on these actions, the study team will identify a sample of database users who will be asked to fill out a short questionnaire describing what they have accessed and what difference the database system has made in their particular situation. Those sampled users will also be asked about desired improvements to the database system as well. The questionnaire would be online and accessible through the database website.
• Focus groups. The study team envisions conducting a limited set of focus groups with key stakeholder groups, researchers, and policy makers regarding the impact of the database system on their decision-making abilities. The study team would coordinate and integrate these focus groups with the surveys to maximize the data collection activity findings.

Together, these evaluation activities enable the study team to explore broadly a number of possible project outcomes such as:

· Public library market-based facility location decisions;

· Library program development/modification activities based on trend analysis with data contained in the database system (e.g., the need for rural telecommunications support should the data demonstrate a deficit of bandwidth in rural areas);

· Research development, enhancement, and continuation based on the availability of the library datasets to researchers, doctoral students, and others; and

• Program evaluation support for state library agencies, public libraries, and others due to data contained within the database system (e.g., LSTA evaluation activities).

These are illustrative of the types of outcome evaluation topics that the study team will consider exploring through the project’s evaluation activities. The study team will refine these topics and develop specific research questions throughout the project. The final summative evaluation will also suggest specific next steps in the development, deployment, and management of the national database system.

PROJECT DISSEMINATION

This project will be disseminated through a wide variety of means, both electronically and in more traditional formats. By virtue of the web-based nature of the national public library database system, the project will have an established web site and can be searchable and accessed through any of the web search engines. The researchers will contact the web masters of other web pages commonly accessed by the public library community (e.g., the NCES web page) to develop linkages to the project’s web site.

This project will also be disseminated through the project’s advisory committee and by the researchers’ continued active involvement in a broad range of organizations and their conferences that would benefit from knowledge about this database. Examples include the American Library Association, the Public Library Association, the Library and Information Technology Association, and the International Federation of Library Associations. The researchers will also offer presentations and promote the use of the database at selected conferences (e.g., PLA, ALA, IFLA, COSLA, etc.) to help ensure that the use of the database system approaches its potential.

SUSTAINABILITY

The study team will insure the sustainability of the database system by conducting a number of efforts in year two to obtain additional and ongoing support to continue to develop, update, and maintain the database system and its associated databases. Efforts to obtain additional ongoing resources will come by soliciting assistance from both within and outside the library community. Dr. McClure, as well as other members of the study team, are well-known by a number of organizations, foundations, and state/federal agencies and expect to obtain adequate resources to continue the database system past year two.

Regardless, GeoLib/FREAC and the Information Institute are committed to incorporating the annual FSCS public library use data into the appropriate databases within the system after funding for this IMLS project ends. In addition, GeoLib/FREAC will update the digital public library base map on an annual basis as well, using data provided by FSCS/NCES. These commitments will ensure that the opportunities for constituency development on an up-to-date base map and database will continue after the end of IMLS funding.

IMPORTANCE AND VISION OF THE PROJECT

The design and development of a national public library database system linked to a digital base map will benefit a large number of stakeholders in the library community as well as a range of policy makers from the local, state, national, and international level. The database system will be a valuable tool for those library analyses and assessments that utilize the group of datasets that will comprise the initial database system’s release.

Potential analyses will range from those at the local library level (e.g., estimated impacts of library outlet openings and closures on certain segments of the population) to the regional level (e.g., equity of state level library funding) to the national level (e.g., role of public libraries in bridging the digital divide).

The true value and benefits from this project’s databases and map, however, will only be realized over time. As new databases are developed and added into the national public library database system, it will become an ever more important and essential tool in helping researchers, policy makers, and public librarians better meet the needs of the people libraries serve. The map, the databases and web site will become a “one-stop” data source for public library data and databases. It will be of critical importance to settling public library policy, at local, regional and national levels. The long term vision of this first ever integrated public library database system is for it to serve as the premiere data repository for library decision makers, just as the public library serves as the premiere data repository for library users.

References

Bertot, J.C., McClure, C.R., Ryan, J. (2000). Statistics and Performance Measures for

Public Library Network Services. Chicago, IL., American Library Association.

Huxhold, William E. and Levinson, Allan G. 1995. Managing Geographic Information

Systems. New York and Oxford: Oxford University Press.

Jue, Dean K. Christine M. Koontz, J. Andrew Magpantay, Keith Curry Lance and Ann

M.. Seidl, “Using Public Libraries to Provide Technology Access for Individuals in Poverty: A Nationwide Analysis of Library Market Areas Using a Geographic Information System,” Library and Information Science Research Vol. 21 (3): 299-325

Koontz, Christine M. 1997. Library Facility Siting and Location Handbook.

Greenwood, CT., Greenwood Press.

Myers, Dowell. 1992. Analysis with Local Census Data: Portrait of Change. San

Diego, CA., Academic Press, Inc.

Rubin, Jeffrey H. 2001. “Introduction to Log analysis Techniques: Methods for

Evaluating Networked Services,” chapter in, Evaluating Networked Information Services: Techniques, Policies and Issues, edited by McClure, Charles R. and John Carlo Bertot, Medford, NJ., American Society for Information and Technology.

APPENDIX A

LETTERS OF SUPPORT

APPENDIX B

THE NATIONAL GEOCODED PUBLIC LIBRARY MAP

---

[1] These data sets are developed and maintained by federal, state and local governmental agencies of emergency management (E-911), law enforcement, and educational entities (e.g., U.S. Department of Education.).

[2] These digital data sets are often maintained in a geographic information system (GIS) environment. GIS is a collection of information technology, data, and procedures for collecting, storing, manipulating, analyzing and presenting maps and descriptive information about features that can be represented on maps. GIS enables data that is geographically referenced to be graphically displayed (Huxhold and Levinsohn 1995).

[3] A base map is a map that has a known acceptable level of accuracy that is accurate enough for the geometric control of other positional information, ( i.e., one can accurately “overlay” other geographically-referenced data to this base map).

[4]See Letters of Support Appendix A

[5] Lack of such a database led the Federal Communications Commission (FCC) to use student eligibility for free-lunch programs within schools as the measure for setting the telecommunications discount rate for public libraries (Jue, et al 1998).

[6] Census blocks are the smallest of census geographic designations. These are roughly equivalent to a city block. Census blocks do not cross boundaries of countries, tracts, or block numbering areas.

[7] Census tracts are small relatively permanent areas, designed to be homogeneous with respect to population characteristics, economic status and living conditions, averaging 4,000 people.

[8] These national surveys have been conducted since 1994 and are available at http://www.nclis.gov.

[9] The inclusion of this will require data from the Schools and Libraries Division of USAC.

[10] This estimate is based upon past GeoLib contracts with public libraries involving data acquisition and analysis.