Keyword Classification Scheme: Background

A Keyword Classification Scheme for IS Research Literature: An Update

(MIS Quarterly, June 1993, Vol. 17, No. 2)

Authors: Henri Barki, Suzanne Rivard, Jean Talbot

Background Information:

In June 1988, MIS Quarterly published a classification scheme of IS keywords. The development of this scheme was intended to provide a description of the discipline, introduce a common language, and enable the researching of the field's development. The scheme has been recently updated in order to incorporate the new research topics and methods, hence better reflecting the evolution of the IS discipline.

In order to conduct this revision, all the articles that had been published in MIS Quarterly, Journal of MIS, Information & Management, as well as all the IS articles published in Management Science and Communications of the ACM between 1987 and 1992 inclusive, were consulted. All the articles published in Information Systems Research from its beginning until Fall of 1992 and all the IS articles published in Organization Science since its beginning were also included. The total number of keywords identified from this exercise was approximately 2,000. Each of these keywords was examined by all three authors in relation to the classification scheme, and a decision was made with respect to its inclusion. In all cases the authors agreed on whether or not to include a keyword, and where in the scheme to put it.

The new scheme now includes nearly 1300 keywords, which represents an addition of more than 175 new words (listed in the Appendix) to the previous version. Among these are 32 new keywords added to the RESEARCH section (section AI) in order to better describe the research approaches and methods. Other additions include recent develoments such as hypertext, object-oriented analysis, and electronic meetings. The scheme's categories are basically unchanged, and as can be seen in the Appendix, only eight keywords from the original version were deleted.

A scheme such as this has three important functions. First, it defines the field of IS in some detail. Second, it provides a common vocabulary. Third, it provides a tool with which the evolution of research can be studied. The second function served by the scheme is especially important in a rapidly changing domain such as IS. The lack of a common vocabulary among researchers is a problem that cannot be solved by computerized search systems. Researchers regularly conduct bibliographical searches in different computerized databases that are continuously getting bigger and faster. However, lacking a common research vocabulary, such searches are not as fruitful as they could be. That is, while their recall (the proportion of material actually retrieved) might be high, their precision (the proportion of the retrieved material that is found relevant) is often low. The reason for this is that many researchers use different terms to describe articles dealing with similar subjects. Since most databases index the articles according to the keywords used by the authors, it becomes impossible to have high precision in a database search unless one knows all the different terms used by the different authors. Even with full text searches, one still has to know which keywords to look for. The classification scheme provides a much needed common vocabulary.

This page is maintained by Susan Scanlan who welcomes your feedback at sscanlan@csom.umn.edu. It was last updated on January 22, 1998. Comments about the classification scheme itself should be directed to the authors.