The effect of data model, system and task characteristics on user query performance

Abstract

With the proliferation of end-user computing and empowerment of end users in the organizations, the user-database interface is increasingly being recognized as an important research area. A user-database interface consists of two key components: the data model and the query language. Database management system designers can manipulate these two components to produce an effective database interface to maximize user task performance. However, in order to do this, the designers first need to know the effect of the two components on end users and the interaction effect of the two components. Despite the acknowledgment of user-database interface as a cornerstone in database utilization, there is still a paucity of research in this area.In this paper, we report on an empirical study that investigates the effect of entity-relationship versus relational models, and textual versus visual query languages for user-database interfaces. A good understanding of how the data models and query languages affect the user performance will enable the database developer to choose and design interfaces that can provide effective and efficient support for end users.