The data structures of information systems can significantly impact the ability of end users to efficiently and effectively retrieve the information they need. Although views can be used to provide virtual data structures, end users typically lack the expertise to change the structure of the data they access. Furthermore, end users are subject to a learning curve whenever they confront a new data structure, i.e., they often must spend substantial effort to become proficient querying a particular data structure. Therefore, this research develops a methodology for evaluating, ex ante, the relative desirability of alternative data structures for end user queries. In particular, this study advocates evaluating competing data structures by calculating Halstead complexity measures for a random sample of typical information requests.
This research theorizes that the data structure that yields the lowest weighted average complexity for a representative sample of information requests is the most desirable data structure. The theory was tested in two laboratory experiments. The first experiment compared two relational databases, one in first normal form and the other in third normal form. The second experiment compared a third normal form relational database with an object-oriented database.
For the first experiment, the sample queries for the first normal form data structure were less complex than the equivalent queries for the third normal form data structure. For the second experiment, the sample queries for the object oriented data structure were less complex than the equivalent queries for the third normal form data structure. As theorized, in the first experiment end users querying the first normal form data structure outperformed those querying third normal form. In the second experiment, as theorized, end users querying the object oriented data structure performed better than those querying the third normal form data structure. Organizations can apply the methodology to help reduce query complexity and enhance end users' ability to efficiently and effectively extract desired information from organizational databases.