Evaluation of a Taxonomy for Communities of Practice in Large-Scale Agile Software Development by Practitioners

This thesis will be published here end of this year (2025) after publication.

Abstract

Agile Methods such as SCRUM or the Scaled Agile Framework (SAFe) are becoming more popular due to their ability to quickly adapt to new or changing requirements and environments. Transitioning from a traditional approach of software development to agile methods, such as SAFe, presents many challenges. Especially larger organizations face many challenges with incorporating agile methods into their organization’s structures and handling the resulting organizational changes. To support the transition in larger organizations, Communities of Practice (CoP) are recommended. As CoPs are complex constructs, a structured analysis and classification of CoPs with their characteristics and dimensions provides structure and organization and combines research results in one place. A taxonomy gives researchers a clearer understanding of CoPs and their relationships and provides practitioners with a clear overview of CoPs. Tobisch et al. (2025) have created such a taxonomy about CoPs in the context of large-scale agile software development (LSAD). Their goal was to offer a comprehensive and practical framework that captures the key factors relevant to the establishment and cultivation of CoPs in complex agile environments. Even though the taxonomy is based on insights from literature and expert interviews, a thorough evaluation of the final taxonomy hasn’t been done yet. The aim of this thesis is to evaluate how well the taxonomy captures the essential dimensions and characteristics of CoPs in LSAD, and how effectively it supports practitioners in applying it within real organizational contexts. To achieve this, we conducted a combination of expert interviews and a survey with practitioners. This mixed-method approach allows us to gather in-depth feedback despite time and resource constraints. Within the interviews and survey, we collected feedback on the completeness and conciseness of the dimensions and characteristics, as well as insights regarding the taxonomy’s understandability, applicability, extensibility, and usefulness. The outcomes of the expert interviews and surveys provided insight into the strengths and weaknesses and identified areas for improvement of the taxonomy. Overall, the taxonomy was found useful by the practitioners. Also, the taxonomy was perceived as easily applicable, extensible, and overall understandable. Though for the understandability, there were diverse answers. The overall completeness and conciseness were perceived positively as well, though leaving some room for improvement, for example, within the dimensions Purpose, Size, and Location.

Research Questions

RQ1: How can the taxonomy of CoPs in LSAD be evaluated in practice?

RQ2: How well does the existing taxonomy fulfill the quality criteria?

RQ3: How can the existing taxonomy of CoPs in LSAD be improved?