A Study on the Importance of Order in Requirements Prioritisation

A key principle when performing research studies is that of randomisation, in order to counter any effects that the ordering of tasks, elements, subjects, etc. may have on the dependent variables. When performing requirements prioritisation, it is not always possible (e.g., because of how prioritisation methods are constructed) or even desirable to randomise all requirements before prioritising them. It is thus important to know the effect that the initial order of the requirements will have on their final priorities, and this is studied in this article. The results indicate that the initial order of elements does not significantly influence the resulting priorities of the most and least important requirements, but that it does indeed influence the results when looking at all of the requirements.