Using i* to Support a Summative Evaluation

Summative evaluation of a software tool requires the assessment of the defined target outcomes, or high-level goals, of the product. This poses the challenge of how to carry out the assessment in practice. We report our research into addressing this problem by using i* modelling for a summative evaluation of a work-based learning tool. We describe our use of i* to identify a set of detailed goals suitable for qualitative assessment. In particular, we report the development and characteristics of the large-scale SR model used in the process, and the utility it provided to contribute towards a successful evaluation. We believe this to be a novel application of i*, and we present our research outcomes and lessons learned in this area.