Raymond R. Panko & Richard P. Halverson
Even the earliest writers in end user computing remarked on the potential dangers of end user spreadsheet development. Until recently, there was only anecdotal evidence to support their concerns. Now, there is considerable evidence from experiments, field audits, and surveys of end users and organizations that early concerns were well-founded.
This paper presents a framework for risks in spreadsheeting and organizes selected research findings in terms of this framework.
This paper offers a taxonomy for understanding issues in spreadsheet risks.
The risks are represented in three dimensions:
- Dependent variables. These are the issues to be addressed by research.
- Stages in the systems development life cycle. Error rates vary considerably by stage. Some stages have not been studied at all in experiments and field audits, notably the requirements and design stages.
- Research methodology. Each methodology has strengths and weaknesses. Unfortunately, research in most areas is dominated by a single methodology.
1996, 29th Annual Hawaii International Conference on System Sciences, January, pages 326-335