Six strategies for generalizing software engineering theories

Abstract

General theories of software engineering must balance between providing full understanding of a single case and providing partial understanding of many cases. In this paper we argue that for theories to be useful in practice, they should give sufficient understanding of a sufficiently large class of cases, without having to be universal or complete. We provide six strategies for developing such theories of the middle range.In lab-to-lab strategies, theories of laboratory phenomena are developed and generalized to other laboratory phenomena. This is a characteristic strategy for basic science. In lab-to-field strategies, theories are developed of artifacts that first operate under idealized laboratory conditions, which are then scaled up until they can operate under uncontrolled field conditions. This is the characteristic strategy for the engineering sciences.In case-based strategies, we generalize about components of real-world cases, that are supposed to exhibit less variation than the cases as a whole. In sample-based strategies, we generalize about the aggregate behavior of samples of cases, which can exhibit patterns not visible at the case level. We discuss three examples of sample-based strategies.Throughout the paper, we use examples of theories and generalization strategies from software engineering to illustrate our analysis. The paper concludes with a discussion of related work and implications for empirical software engineering research.

The Strategy Blueprint - A Strategy Process Computer-Aided Design Tool

“…Design Science is the design and investigation of artifacts in context (Wieringa and Daneva, 2015). As a research method, it is solution-oriented and is focused on the interaction of a proposed solution and the context in which the solution is used.…”

“…Third, following Wieringa and Daneva (2015), we acknowledge the need for more evaluation to improve the generalisability of the results. A central question in this respect is evaluating the extent to which our current results could be observable in other similar but different organizations (e.g.…”

Human factors in software development: On its underlying theories and the value of learning from related disciplines. A guest editorial introduction to the special issue

“…For example, Endres and Rombach [24] catalogued theories in the various sub-fields of SE, while searching ''for the fundamental rules in SE, why they work and why they are useful''. Also, empirical SE researchers have made calls for borrowing theories from other disciplines [25] and applying them to explain SE phenomena [26] and for improving our reasoning about generalization of empirical claims in SE studies [26]. The most recent indicator for this increased interest in SE theories is the SEMAT initiative (www.semat.org) [27] that aims to increase the focus on general theories in SE.…”

“…Wieringa and Daneva [26] consider that theories need to have a (i) conceptual framework, (ii) generalizations and/ or models, (iii) scope conditions need to be defined, and optionally (iv) causal explanations.…”

“…Human Factors in SD can be considered a subfield of Empirical Software Engineering, and hence the theory structure adopted by Sjøberg et al [33] (and related to Wieringa and Daneva [26]) is applicable. We hence adopt Sjøberg et al's [33] definition of theory, in order to briefly analyse the essential theories related to Human Factors in SD later on in this section.…”