Using students as subjects in requirements prioritization

Author(s): Berander, P
Venue: International Symposium on Empirical Software Engineering
Date: 2004

Sample Size: 20
Class/Experience Level: Graduate Student


The paper discusses an experimental study on requirements prioritization performed with classroom students as subjects. Prioritization is the aid in selecting and implementing a subset of requirements and still produce a system that meets the most essential needs of the stakeholders and quality for the customers. The study is used in comparison with the results of other similar prioritization made in student projects, other classroom studies, literature, and in industrial case studies. The basis for this study is a requirement prioritization technique called Planning Game - a practice of eXtreme Programming. The goal of the study is to evaluate in which cases students could be successfully used as subjects in experimentation. It also exposes students to an exercise of release planning. The paper concludes that "students in a classroom environment are less suitable than students in projects as representations for professionals in studies of this kind."

Process Outline

  1. Students got the task to come up with 10-20 requirements for mobile telephone domain where each requirement is given a unique number and have it written on post-it notes
  2. Each group is divided into two sub-groups by random where half were the role of customers and other half were the role of developers
  3. Developers were given the task to put relative cost of implementing each requirement while the customers were given instructions to put requirements in three piles (less important, important, very important) and rank them
  4. Developer and customers reunite and plan the three releases of the product together based on their priorities
  5. Lastly, do another negotiation but instead of three releases with the same equal points (300-350 each), the new three releases would have varying points that increases for each sequential release

Requirements prioritization made by students and customers in projects do seem to be similar to the industry. However, projects presented in the paper prioritized differently than other cases that were analyzed. Therefore, no conclusions can be made about the suitability of students in the projects. Nevertheless, the results indicate that students in projects at least are more suitable than students in a classroom environment for studies when commitment is an issue.