iasese  2015

13th International Advanced School of Empirical Software Engineering

Theories in Empirical Software Engineering

Empirical Software Engineering has produced an extensive base of empirical data about different aspects of software engineering products and processes. Yet, there is relatively little focus on theories that could explain or predict software engineering phenomena. But without theories, we do not have a science of software engineering. And without empirically validated theories, it is hard for practitioners to see whether or how they could apply the results of academic research to their own situation. In software engineering, as in other design disciplines, theories are needed to assess which knowledge is applicable when, and to estimate what the effect of alternative treatments of the problem might be.

However, it is possible (but not desirable) to set up impeccable statistical research designs that neither make use of, nor contribute to scientific theories; likewise for qualitative research designs.

Learning Objectives

In this edition of the International Advanced School of Empirical Software Engineering, we will discuss and illustrate in dialog form how theories can be used and developed in empirical software engineering research. We will discuss the nature of scientific theories in general and of design theories in particular, and discuss examples of software engineering design theories with the audience. We will discuss the role of theories in motivating a research design, both for statistical, sample-based research and for case-based research. We will also discuss the three major steps that lead from data to theories: description, explanation and generalization.

The School will take the form of a dialog among the teachers in which a general presentation about theories in design science will be critiqued from a software engineering point of view. The audience is encouraged to participate in this discussion, using their own research as examples or counterexamples of points made in the discussion.

Who Should Attend?

Anyone interested in using or developing theories in empirical software engineering both in academic research and in industrial research. Anyone interested in research that is effective.

What Will Participants Take Away?

Participants will take away a greater appreciation of the use of theories in software engineering, whether they are academic researchers or practitioners.


  • Daniel Méndez Fernández (Technische Universität München, Germany)
  • Lutz Prechelt (Freie Universität Berlin, Germany)
  • Roel Wieringa (University of Twente, The Netherlands)

co-located events


supported by