How do we assess learning in Software Engineering?

Our IUSE project, Exploring Brownfield Programming Assignments in Undergraduate Computing Education, aimed at improving software engineering education by creating courses in which students contributed to existing code bases in ways that mirrored authentic industry software engineering practice. However, assessing the extent to which this intervention was successful proved more challenging that we had originally anticipated. In this talk I describe the challenge of assessing learning in software engineering courses, which includes the problem of assessing the quality and quantity of software engineering efforts—a problem that has proven difficult for decades.