The KITE Model for Assessment of Academic Software Products
We reflect on the topic of assessing the merit of software products developed by research groups within the academia. To this end, a model is proposed to define the score of an arbitrary software product. The model consists of four determinants, namely new knowledge dissemination effect (K), impact in target population (I), technological innovation (T), and engineering achievement (E). These determinants are integrated into a ”KITE” graphical model. The model admits both geometric and numeric interpretations, enabling decision makers to analyze profiles of software productivity for a particular academic unit from a quantitative or qualitative viewpoint. The ratings, which enable software to be scored regarding each determinant, are also described. Following the model, preliminary test lists are sketched as a proposal of measurement instruments for these scores.