Summary

The importance of information systems to organizations continues to grow. Large sums of money are invested in information system (IS) projects. However, many IS projects fail to meet their targets or even fail completely. In the effort to successfully control such projects, there appears to be relatively little focus on the people who are actually making the decisions for the project. Prior research has shown that these decision makers aren’t always perfectly rational and that they may (consciously or subconsciously) use heuristics (mental shortcuts) which can lead to systematic biases and irrational decision making. Such biases can result in project escalation where resources continue to be devoted to a failing project despite the presence of information which indicates that the project is in trouble.

In this dissertation, several biases are studied which thus far have remained (mostly) unexplored as potential causal factors of project escalation. In two experiments it was tested whether, ceteris paribus, a difference in the name of a project could influence decision making preferences. The results of these experiments indicate that project names can lead to an overall more positive or negative view of the project (Chapter 2) and that they are capable of drawing decision makers’ attention towards specific project attributes (Chapter 3). The outcomes of our study on the effects of differences in construal levels show that decision makers with a higher construal level are more likely to escalate commitment to a project (Chapter 5). Because the construal level of decision makers can be influenced by factors that are completely unrelated to the project, this implies that even these types of factors can indirectly influence the likelihood of project escalation. In addition, this dissertation also includes a study on a previously known causal factor of project escalation, framing, but applies it in a novel manner (Chapter 4). The results of the text analysis process, based on interviews with managers, suggest that there is a strong link between the natural framing usage of managers and the view that they have of a project. This implies that the framing usage of managers could provide valuable additional information about their underlying project view.

The studies in this dissertation demonstrate that factors which at first sight may seem innocent or unimportant, are quite capable of influencing the likelihood of project escalation. In fact, even factors that are completely unrelated to the project appear to be capable of indirectly influencing project decision making. These findings underline the importance of not only focusing on charecteristics of the project and the organizational context but also on the potential biases of the decision maker in order to obtain a more complete view when evaluating projects.