Stylized Facts as an Instrument for Literature Review and Cumulative Information Systems Research

Constantin Houy, Peter Fettke, Peter Loos

In: Communications of the AIS (CAIS) 37 1 (Art. 10) Pages 225-256 Association for Information Systems (AIS) 8/2015.


The accumulation of scientific knowledge is an important objective of Information Systems (IS) research. Although different review approaches exist in the continuum between narrative reviews and meta-analyses, most reviews in IS are narrative or descriptive – with all related drawbacks concerning objectivity and reliability – as available underlying sources in IS do typically not fulfil the requirements of formal approaches such as meta-analyses. In order to contribute to the solution of this problem and the discussion on how cumulative IS research can be effectively advanced using a more formalized approach fitting the current situation in IS research, this article points out the potential of Stylized Facts (SF). SF are interesting, sometimes counterintuitive patterns in empirical data which focus on the most relevant aspects of observable phenomena by abstracting from details (stylization). SF originate from the field of economics and have been successfully used in different fields of research for years. In this article, we discuss their potential and challenges for literature reviews in IS. We supplement our argumentation with an application example reporting our experience with SF. As SF show considerable potential for cumulative research, they seem to be a promising instrument for literature reviews and especially for theory development in IS.


Weitere Links

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz