Assessing divergence and convergence in the design process: Results from two experiments

It is often assumed that the design process comprises divergent and convergent thinking cycles. Although many authors studied the design process in depth, a quantification of divergence and convergence is yet to be established. This paper attempts to quantify divergence and convergence in the design process for teams of design students. Student submissions from two independent design-related courses from two Universities are analysed for divergent and convergent thinking. Such an approach uses Natural Language Processing to harvest existing student submissions in written documents and converts them into a 'real-time' dashboard for feedback that can be used in multiple ways: searching for a signature process for good design or as a feedback tool for both students and instructors. A specific universal pattern for "good design thinking"was not found in the experiment. However, several parameters often obscured by the lack of quantitative data about levels of divergence and convergence were uncovered. Future work must explore other patterns or factors relevant to instructors and students.