Investigating a Conceptual Construct for Software Context

TL;DR

This paper proposes a theoretical framework with six dimensions to define software context, tests its applicability through literature categorization, and highlights the need for clearer conceptualization in software engineering research.

Contribution

It introduces and evaluates a minimal, six-dimensional construct for software context, advancing theoretical understanding in empirical software engineering.

Findings

The 'Why' dimension relates more to objectives than context.

Some factors do not fit into the proposed framework and need further study.

The framework helps clarify the meaning of context in software engineering.

Abstract

A growing number of empirical software engineering researchers suggest that a complementary focus on theory is required if the discipline is to mature. A first step in theory-building involves the establishment of suitable theoretical constructs. For researchers studying software projects, the lack of a theoretical construct for context is problematic for both experimentation and effort estimation. For experiments, insufficiently understood contextual factors confound results, and for estimation, unstated contextual factors affect estimation reliability. We have earlier proposed a framework that we suggest may be suitable as a construct for context i.e. represents a minimal, spanning set for the space of software contexts. The framework has six dimensions, described as Who, Where, What, When, How and Why. In this paper, we report the outcomes of a pilot study to test its suitability by categorising contextual factors from the software engineering literature into the framework. We found that one of the dimensions, Why, does not represent context, but rather is associated with objectives. We also identified some factors that do not clearly fit into the framework and require further investigation. Our contributions are the pursuing of a theoretical approach to understanding software context, the initial establishment and evaluation of a construct for context and the exposure of a lack of clarity of meaning in many 'contexts' currently applied as factors for estimating project outcomes.