Meinung / Dialog Empirical Research Strategies in Conceptual Modeling – Silver Bullet or Academic Toys ?

In Information Systems research conceptual modeling is an important field. Wand and Weber for example distinguish four elements: conceptual modeling grammars, conceptual modeling methods, conceptual modeling scripts, and conceptual modeling contexts [WaWe02]. With these elements they discuss the research opportunities to gain and evaluate results like artifacts. A common way to ensure a statistical, sampling-based generalizability is empirical testing, particularly used in the Anglo American region [cp. in detail LeBa03]. However, methodical problems can occur: To evaluate the quality of a new modeling method, the test person has to be familiar with the method before he/she can use it appropriately, which may affect the results. Thus the evaluation results are difficult to analyze and interpret. Extensive empirical tests for modeling methods – being very labor intensive and time consuming – are also rather counterproductive to the practical science process. Furthermore, economic or social objectives of conceptual modeling projects may affect the results too. So what are the guidelines in information systems research [HMPR04] and especially for empirical research strategies in conceptual modeling? Are empirical evaluations of different conceptual modeling elements, as described by Wand and Weber acceptable or not? What type of empirical evaluation should be used in each case? Which challenges and problems have to be faced, and more important which quality factors of a modeling artifact cannot be measured appropriately by empirical research strategies? The following seven contributions (in alphabetical order of authors’ names) will address the state of the art as well as specify some future research opportunities regarding empirical testing of conceptual modeling artifacts. The distinguished views demand further discussion of information system research yet. If you would like to present your point of view on this matter, please submit your article (max. 2 pages) to the chief publisher: Prof. Dr. Wolfgang König, Universität Frankfurt am Main, email: [email protected]. References