Kellam / Peter Numeric Data Services and Sources for the General Reference Librarian

1. Auflage 2011
ISBN: 978-1-78063-259-9
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 248 Seiten

Reihe: Chandos Information Professional Series

The proliferation of online access to social science statistical and numeric data sources, such as the U.S. Census Bureau's American Fact Finder, has lead to an increased interest in supporting these sources in academic libraries. Many large libraries have been able to devote staff to data services for years, and recently smaller academic libraries have recognized the need to provide numeric data services and support. This guidebook serves as a primer to developing and supporting social science statistical and numerical data sources in the academic library. It provides strategies for the establishment of data services and offers short descriptions of the essential sources of free and commercial social science statistical and numeric data. Finally, it discusses the future of numeric data services, including the integration of statistics and data into library instruction and the use of Web 2.0 tools to visualize data. - Written for a general reference audience with little knowledge of data services and sources who would like to incorporate support into their general reference practice - Combines information on establishing data services with an introduction to available statistical and numeric data sources - Provides insight into the integration of statistics and data into library instruction and the social science research process

Lynda M. Kellam is the Data Services and Government Information Librarian at the University of North Carolina at Greensboro's University Libraries. She is UNCG's first data librarian with the mandate to create and develop data services for the Reference and Instructional Services Department. In addition to providing research assistance and instruction on data and government sources, she is the library instruction liaison to the Political Science Department, the Environmental Studies program, and the pre-Law program. She received her M.A. in Political Science from the University of Wisconsin, Madison and her MLIS from the University of North Carolina at Greensboro. She serves on the conference planning committee for the International Association of Social Science Information Services and Technology, the primary data librarianship association, and works closely with the American Library Association's Government Documents Round Table. She was named an American Library Association Emerging Leader in 2010 and received the Association of College and Research Libraries Librarian Scholarship in 2009. She is also a member of the American Political Science Association.

Autoren/Hrsg.

Kellam, Lynda

Peter, Katharin

Leseproben

1 Introduction to data services and sources
Abstract:
The purpose of this chapter is to provide an introduction to data services within the academic library by describing the history and development of data services in the United States and other countries. The chapter also introduces the primary concepts necessary for understanding numeric data, datasets, and secondary data analysis, including the difference between aggregate statistical products and microdata, the purpose of coding data, and the use of data documentation. Keywords history of data services data services numeric data secondary data analysis description of aggregate statistics description of microdata Recent years have seen a proliferation of free and subscription-based numeric data resources on the Web. From the World Bank’s Open Data Initiative to the US Census Bureau’s American FactFinder, facts and figures are readily available on an array of topics. This increase in access is the result of several converging technological advances, which include wider Internet availability and use, better and faster Internet connections, and the ability to create compressed files. All of these changes have decreased download times and increased portability of large amounts of information. In addition, as it has become easier to access numeric facts and figures, librarians have seen an increasing interest in quantitative information among users with a range of skill sets and knowledge levels. Quantitative information is no longer only the purview of statisticians; more people are interested in the possibility of expressing ideas through numbers. The increasing ease of use of spreadsheet tools like Microsoft Excel, and the explosion of Web 2.0-based data visualization websites such as Many Eyes (http://manyeyes.alphaworks.ibm.com), have allowed more people to see the benefits of using and visually representing data. With these changes in the access to numeric data, librarians have become central participants in assisting users. Our traditional focus may have been on the written word, but the rise in digital formats and files has carved out a new role for the library, one that supports information in all its forms – from the written word, to the digital image, the streaming media sample, and the numeric data file. Moreover, our promotion of information literacy and emphasis on information-literate users means we need to pay attention to all types of information sources, even the non-textual. Users may not immediately associate numeric data sources with the library, but increasingly, libraries and librarians are being called upon to purchase, support, and archive these sources. Because libraries have acquired more numeric data sources, patrons have increasing expectations that librarians will provide support for these sources. As faculty and graduate students move from institution to institution, they may have expectations that the services offered at one library will be available at another. The difficulty facing many libraries, especially smaller ones, is that the use of numeric data sources, from the basic statistical database to a large data set, requires basic statistical literacy and potentially more advanced skills. Although some librarians may have had a statistics course as an undergraduate or a research methods course in graduate school (or both), most Library and Information Studies programs do not teach the skills necessary to support numeric data sources. Few new librarians have had the same exposure to data sources and quantitative analysis as they have had to more traditional library principles, such as cataloging standards. Even though users may have increasing expectations about the role of libraries in supporting numeric data sources, our professional education, especially in the United States, has not quite caught up with those expectations. No matter the size of an institution, librarians can expect an increasing interest in numeric data sources. As user expectations for support increase, the need for advanced skills to support data sources will also rise. In small libraries, the social sciences librarian or the business librarian may serve as a de facto data specialist and assume responsibility for data questions. This model can be problematic if the designated librarian does not have the skills to support numeric data sources and has only limited possibilities for training. Gerhan (1999) notes that librarians without a basic understanding of the use of quantitative analysis will have difficulty asking the appropriate questions when conducting reference interviews. History of support for numeric data
How did libraries begin to support numeric information? Statistical publications in print or microtext formats have always been a part of library holdings, especially those American libraries involved in the Federal Depository Library Program (http://www.fdlp.gov). From Statistical Abstract of the United States to the print Census volumes, librarians have had experience with these publications for a long time. A shift began with the emergence of Machine Readable Data Files (MRDFs). These MRDFs encoded data about all types of information into a variety of machine-readable formats – from punch cards to tapes to CD-ROMs. As these formats became more accessible and commonplace, the need to support numeric data (in addition to other digital files) increased. In the 1960s and 1970s many libraries did not have the infrastructure to support numeric data files. Several authors note the lack of infrastructure and expertise in libraries combined with a perceived unwillingness to support these emerging technologies (Heim, 1982; Rowe, 1984; Chiang et al., 1993). This void led some universities, especially large research institutions, to create and house data support centers or archives in academic departments or computing centers, rather than within the libraries. Rowe describes the establishment of the Roper Center for Public Opinion Research and the Inter-university Consortium for Political and Social Research (ICPSR) as ‘the first organized efforts to formalize the distribution of MRDF’ (1984: 327). Created in 1962, the Inter-university Consortium for Political and Social Research housed primarily the ‘American National Election Study’. Over the next few decades its mission grew to the archiving of data from a wide variety of studies and responsibility for instructional support. In the early 1990s the United States Census Bureau released the 1990 Decennial Census on CD-ROM. As these new formats began arriving in the federal depository units of many libraries, librarians and administrators started to consider support for these data files in addition to the typical statistical sources in print (Geraci et al., 2008; Treadwell and Cogswell, 1994). Similarly, Statistics Canada gave Canadian universities more access to data sources through the Data Liberation Initiative (Geraci et al., 2008). Through the 1990s libraries began creating data centers housed within the library, such as the Machine Readable Data Center at the University of Minnesota, and data retrieval systems, such as INFeRS at Cornell University’s Mann Library (Treadwell and Cogswell, 1994; Chiang et al., 1993). Although data support and data librarian positions began to proliferate, they were generally concentrated in the larger research institutions.1 With the increase in access to the Internet and the decrease in download times over the past decade, we have seen yet another shift in the approach to data services, at least in the United States. While patrons may have quicker and more immediate access, they may not understand how to use specific sources or have the training necessary to comprehend data. For these reasons, smaller universities and colleges have expressed a need for a librarian with a data background – one who is dedicated to supporting and educating users about numeric information. In supporting numeric data services, libraries have several possible organizational approaches. They could have a de facto model, with data responsibilities supported by the current social sciences or business librarian. A coalition approach is also possible, in which a library’s reference staff supports basic statistical and some numeric data sources while another unit supports more advanced data tasks. The third model, the creation of a designated data services librarian position, is becoming common at smaller universities and colleges. There are various permutations of this approach – sometimes the entire school has only one data support person, while other institutions may have a designated data librarian in combination with units outside of the library supporting quantitative research. Every approach has strengths and challenges and each institution will need to gauge the most appropriate one for their staff, patrons and environment. Our goal is to provide general reference librarians with a framework for understanding numeric data services and sources, and to increase the level of awareness of and comfort with this specialization. We will focus primarily on social science data sources throughout this book because of the current shape of data librarianship. Many data services positions have been combined with government information or other social science liaison...