J. Describe Information-Seeking Behavior

Introduction to Core Competency J. Describe the fundamental concepts of information-seeking behaviors.

Nearly ever issue of America Libraries, the magazine of the American Library Association, has an article on information seeking behavior. “Search Fatigue” (Beall, 2007, p. 47) discusses the difference between key word and metadata-enabled searches. Reliance on keyword or full text searching is the chief cause of search fatigue. Metadata-enabled search engines search metadata fields and their controlled vocabularies—such as subject headings and descriptors—to generate results. Beall divides searching into casual information seeking and serious information seeking. A casual seeker will be satisfied with the first three hits on Google and ignores the remaining millions, while a serious seeker conducting exhaustive research must aim for relevance and precision, which only metadata-enabled search engines can provide. Beall claims that unless librarians persistently explain to searchers the great value of metadata and metadata-enabled search engines, keyword search engines will dominate and eventually replace metadata-enabled engines.

I discovered that this phenomenon is already happening when researching a community college library. In LIBR 251, I analyze the usability of De Anza College library's online catalog based on accepted standards and guidelines. I found that the librarians encourage and instruct keyword over subject searches and that the catalog does not even have a browse mode on their basic search, just keyword. The librarian in charge of that catalog, whom I interviewed, said that students don't understand what they are browsing because they don't understand Library of Congress Subject Headings and the browse page just confuses them.

With the low number of librarians, at that time only four, serving this community college's population of nearly 23,000, I think the decision to fall back on keyword searching was because students were familiar with it and the time was not available to provide true bibliographic instruction. In my experience, many people who use public library catalogs have no idea how to make use of subject headings for expanded searches or that the subject headings in their local catalogs are standardized across all libraries and all catalogs. Nor do most people understand that to search successfully, you have to match your search terms with those already present in the online catalog.

Until I was starting in the Library Technology program, I didn't understand where I was when a library catalog threw me into a browse mode. I couldn't figure out why the search engine wasn't searching by keyword because I didn't know what metadata was or that my search term had to match that metadata. I had no idea that the list of subjects on the browse page were Library of Congress Subject Heading and that the subject headings in one catalog were the same as those in all other catalogs. In the Cataloging Final Exam, I discuss this further and demonstrate my understanding of how descriptive cataloging and subject analysis impacts user access to library resources.

When creating a database for a collection of articles relating to information storage and retrieval, my team developed a procedure for database testing and evaluating our subject headings. We decided that in order to verify the accuracy of our controlled vocabulary, our tests needed to meet a threshold of 20% recall and 40% precision. We felt that given the small size of the database, 14 items, and the limited size of the vocabulary this was an acceptable level. We had to place ourselves in the shoes of potential searchers. That meant having something to search for in the database, which meant constructing questions that might relate to a student researching the subject of information retrieval. Our questions and results start on page 21 of the Subject Headings for a Database document.

In the Midterm Exam for 202 Information Retrieval, I discuss the concepts of discriminating and aggregating in information seeking, information surrogates, inverted files, how pre- and postcoordinate vocabularies affect searches, interface navigation, and briefly touch on MARC records and Dublin core metadata schemas. In the Final Exam for 202, I discuss the concepts of precision and recall, the pearl growing technique in analytical searches, the difference between a classification system and a controlled vocabulary, summarize an article on searching by Marcia Bates, and briefly discuss the process of requirements elicitation.

In LIBR 247 Vocabulary Design, I created a back of the book, single text index for Chapter 2 of The Author and the Index that represents my understanding of information seeking behavior. Creator of book indexes must analyze the text for relevant terms and anticipate how users will search the index. I also created a thesaurus in a team project in LIBR 247. The purpose of our thesaurus is to provide terms used to search a database of 22 articles on outdoor, non-organized sports. A thesaurus has a flat structure compared with the layered, hierarchical structure of a classification system such as subject headings, or a subject directory where users drill down to reach a specific concept. Items in a thesaurus are listed in alphabetical order and are layered only one level up or down, broader term (BT), narrower term (NT) and related term (NT), while classification systems are alphabetical only at the top-most categories or bushes.

Beall, J. (2007). Search fatigue. America Libraries, 38(3), 47-50.