The Effect of Users' Work Tasks on Librarians' Database Selection

Thumbnail Image
Files
umi-umd-4718.pdf(1.34 MB)
No. of downloads: 3369
Publication or External Link
Date
2007-08-06
Authors
Kim, Soojung
Advisor
White, Marilyn Domas
Citation
DRUM DOI
Abstract
A recent trend in information searching research is task-based information searching, which views a user's task as a central factor for understanding information-seeking behaviors and designing information retrieval systems. To investigate empirically the role of tasks in information searching, particularly in the business domain, this study analyzes the database selection process used by librarians from the perspective of users' tasks. The first part of the study focused on identifying and characterizing business tasks and the associated questions needed to complete the tasks. An inventory of 30 business tasks and 144 associated business questions was developed through content analysis of Harvard Business School cases and other published materials. The second part of the study explored the influence of tasks on database selection by conducting a survey among business librarians in academic institutions. Nine sets of survey questionnaires were created based on the identified business tasks and questions and each questionnaire, containing a total of five business questions for two to three tasks, was disseminated through a Web-based survey tool. Out of 52 sampled librarians, 29 (56 percent response rate) participated in the study. The survey questionnaires focused on participants' use of tasks and related business questions to determine information types likely to answer the questions, to choose databases, and to determine the criteria used to select the databases. The characteristics of business tasks and questions were analyzed and linked to other elements - information types, database selection criteria, and selected databases - to understand the interplay among all elements in the database selection process. The analysis noted the participants' reliance on users' tasks in various aspects of an information searching process. A database selection process was further modeled to describe how five task or context-related criteria - company size, company type, industry sector, geographical setting, and business stage - influence database selection. The inventory of business tasks and questions, along with the patterns among the elements, set the stage for a task-based database selection system.
Notes
Rights