|Authors||M. Jørgensen and K. Teigen|
|Title||Uncertainty Intervals Versus Interval Uncertainty: an Alternative Method for Eliciting Effort Prediction Intervals in Software Development Projects|
|Publication Type||Proceedings, refereed|
|Year of Publication||2002|
|Conference Name||Proceedings of International Conference on Project Management (ProMAC)|
Frequently, there is a poor correspondence between the judged and the actual uncertainty of effort usage in software development projects. This may to some extent be a consequence of the uncertainty elicitation process. Traditionally, software developers are asked to provide the minimum and maximum effort of development work for a given confidence level, e.g., minimum and maximum effort that includes the actual effort usage with a 90% probability. An alternative uncertainty elicitation process is to instruct the software developers to provide the uncertainty of a given effort interval, e.g., the probability that the actual effort is between 50% and 200% of the estimated most likely effort. In an empirical investigation, this alternative process led to significant improvement of prediction interval accuracy. The observed improvement using this alternative elicitation process can, we believe, be explained through a simplified interpretation of historical prediction accuracy data, less “conflicting estimation goal”, and less influence from the “anchoring effect”.