Empirical Studies into UML in Practice: Pitfalls and Prospects

A key open question in the area of software modeling is which costs and benefits it brings to software development and maintenance. For answering this question, better empirical studies into software modeling are needed. In this paper I focus on what I believe are the pitfalls in- and prospects for such types of studies. This paper is an abstract for an invited keynote at the Modeling in Software Engineering (MiSE) workshop at ICSE 2017.