OVERVIEW.

This paper will present findings from the Primarily History project, concerning how historians are locating primary sources in the early years of the digital age. Primarily History, an international study housed at the University of North Carolina at Chapel Hill and the University of Glasgow, also explores how historians teach their students to find such research materials and how archivists and librarians can facilitate this education and resource discovery and use. Data provided here reflect information-seeking behaviors of historians working in the United States.

NEW PATHWAYS TO PRIMARY RESOURCES.

Historical research has long been a detective game. While ascertaining the veracity and authenticity of primary resources and the critical interpretation of documents within the context of historical and cultural understanding lie at the heart of historical scholarship, the not so trivial task of locating these materials that serve as the grist of history, must precede any high level analysis. Throughout the 20th century, historians sought the materials they needed to shed light on the past in a number of time-tested ways. Studies conducted in the 1970s and 1980s found that historians most often located relevant primary source materials by following references in published histories, talking with colleagues, and searching likely repositories. In a print paradigm these were all appropriate and efficient methodologies. Recent technologies, however, present the historian with many new possibilities for locating research materials that may prove more efficient and even more effective The road to online finding aids and other digitized resources was not an easy one for archival repositories. For the past twenty years archivists have expended a good deal of time, money, intellectual effort, and angst to produce electronic access tools for the collections in their repositories. The first efforts focused on creating MARC/AMC (Machine Readable Cataloging/Archives and Manuscripts Collections) records for online catalogs and bibliographic utilities such as OCLC and RLIN. Many archivists doubted the efficacy of MARC, a library-based standard, for archival description. In the early 1990’s, some pioneering archivists mounted machine-readable text finding aids on gopher sites on the Internet. By the mid-1990s, a small group of archivists were developing the Encoded Archival Description SGML DTD. By the turn of the millennium, most special collection repositories, at least those in larger units such as academic libraries, had some sort of website, many of which contained HTML encoded finding aids. Today, a small but steadily growing number of repositories have EAD finding aids at their websites. These networked tools not only facilitate information discovery, they can also prepare a scholar for a very productive in-person visit to a repository. Today’s historians can study digitized collections of materials online as well as search bibliographic databases with descriptions of both secondary and primary materials. Perhaps most significantly, historians can read, download, or print entire finding aids for collections in a large number of archival and manuscript repositories world wide. Now that there is a significant corpus of finding aids online it is time to explore how researchers such as historians use them and how archivists might make them more useful. While spending a good deal to create electronic access, to date, archivists have conducted few user studies to judge its effectiveness or ascertain the need for enhanced user education.

PRIMARILY HISTORY PROJECT.

Funded by the Gladys Kriebel Delmas Foundation, The Primarily History project, a collaboration of the School of Information and Library Science at the University of North Carolina at Chapel Hill (UNC-CH) and the Humanities Advanced Technology and Information Institute (HATII) at the University of Glasgow, Scotland, is the first international, comparative project to explore historians’ information-seeking behaviors in today’s web-based, networked environments. Perhaps most importantly, we are examining how historians are preparing the next generation of scholars, specifically, what they are teaching their graduate students about information seeking in the digital library environment and how the students are learning to use retrieval tools. This project is also surveying how special collections libraries and archives provide access to these materials and is seeking enhanced models for outreach and user education that will facilitate historians and their students in locating and using primary resources. Through surveys and interviews we are exploring how historians are employing these new tools and techniques. Dr. Tibbo from UNC-CH has surveyed 700 historians located at leading U.S. history programs; Dr. Ian Anderson from Glasgow surveyed close to 800 historians working at universities in the United Kingdom. Both investigators followed the surveys with in-depth interviews with a subset of these populations. Questions in the initial stage of this research focused on scholarly information-seeking in archives, manuscript repositories, and special collections libraries. We asked historians how they found the materials upon which they based their research. Questions included: Do they use library catalogs or the OCLC or RLIN databases that contain MARC records of finding aids? Do they use electronic indexes such as Archives USA? Do they search the Web for finding aids using keywords? Do they go directly to the websites of likely repositories and search for materials? The interviews asked historians to describe how they had located primary resource materials for a recent project and what they taught their graduate students regarding information seeking methodologies.

A SAMPLING OF FINDINGS.

For many historians working in the United States, the traditional methodologies for locating primary materials remain the most utilized. Ninety-eight percent indicated that they found materials by following leads and citations in printed sources; 77% searched printed bibliographies; 57% consulted printed documentary editions; 73% searched printed finding aids; 73% searched printed repository guides; 51% used newspaper files to find other materials; 50% used government documents in this way; and 38% used the now out-of-date printed NUCMC volumes. The most noticeable difference in behaviors across ranks is that only 24% of assistant professors used the printed NUCMC volumes with 40% of all full professors, including distinguished faculty and deans, using this now out-of-date tool. Interestingly, however, only 18% of those individuals who had taught history for 40 years or more searched the printed NUCMC. Use of traditional resources, however, does not preclude the use of digital technologies developed within the last decade and populated with research materials (e.g., a critical mass of electronic finding aids and some digitized documents) only within the last three to four years. Sixty-nine percent of the history scholars used their own institution’s online (library) public access catalog (OPAC) [while digital, and now probably a web-based technology, not really a new approach]; 64% searched other institutions’ OPACs via the Internet; 59% used bibliographic utilities such as RLIN and OCLC; 58% said they looked for information directly on repository websites; 40% indicated that they searched the Web for primary materials using a search engine such as Alta Vista or Google. Twelve percent of respondents indicated they used the Archives USA database, while 14% searched NUCMC online from the Library of Congress website with only 8 individuals (3% of respondents) exhibiting both behaviors. At the most superficial level, we can see that historians are using traditional finding tools more heavily than newer approaches but we need qualitative interview data to understand why. It may well be that many individuals are most comfortable employing methodologies they have found useful throughout their careers and that these tools remain sufficient. It may also be that they have not found searching the web or bibliographic utilities very useful. After all, most archival websites have mounted electronic finding aids only since 1996 with many institutions still working on making a significant portion of their inventories web accessible. When we look at the 220 projects with end dates of 1998 or later, 135 or 61% of these researchers visited repository websites within the course of their work and 98 (45%) used web search engines to find relevant collections. Significantly, the historians in this study averaged over twenty years of teaching experience each. This is significant because the initial round of interviews indicated that many of the veteran historians had completed the largest part of their information seeking years ago, having located large stores of records and papers to analyze for many years. These individuals appear to invest little time in learning new information-seeking behaviors. Other individuals who more dramatically shifted topics across projects, might well seek out primary resources during their entire careers.

PRELIMINARY CONCLUSIONS.

While much work remains in this study, it is clear that archivists and special collection curators must both maintain traditional print descriptive tools and create an increasing number of electronic finding aids and digitized documents as historians are employing a wide range of information seeking behaviors. Evidence to date indicates that younger scholars are turning to digital resources more often than their older counterparts but such behaviors may also linked be linked to research topic. Additional interviews, especially with current Ph.D. students will clarify emerging historical research trends in the early years of the Digital Age. Increasing use of the Web for information seeking and virtual trips to repositories demands that archivists create highly accessible websites, study user behavior so as to improve discovery tools, and provide increased user instruction for those who will only visit the collection from afar.