Handover of managerial responsibilities in global software development: a case study of source code evolution and quality

Abstract

Context: Studies report on the negative effect on quality in global software development (GSD) due to communication and coordination related challenges. However, empirical studies reporting on the magnitude of the effect are scarce. Research Method: This paper presents findings from an embedded explanatory case study on the change in quality over time, across multiple releases, for products that were developed in a GSD setting. The GSD setting involved periods of distributed development between geographically dispersed sites as well as a handover of project management responsibilities between the involved sites. Investigations were performed on two medium-sized products from a company that is part of a large multinational corporation. Quality is investigated quantitatively using defect data, and measures that quantify two source code properties, size and complexity. Observations were triangulated with subjective views from company representatives. Results and Findings: There were no observable indications that the distribution of work or handover of project management responsibilities had an impact on quality on both products. Among the product-, process-, and people-related success factors, we identified welldesigned product architectures, early handover planning, support from the sending site to the receiving site after the handover, and skilled employees at the involved sites. Overall, these results can be useful input for decision-makers who are considering distributing development work between globally dispersed sites or handing over project management responsibilities from one site to another. Moreover, our study shows that analyzing the evolution of size and complexity properties of a product's source code can provide valuable information to support decision-making during similar projects. Finally, the strategy used by the company to relocate responsibilities can also be considered as an alternative for software transfers, which have been linked with a decline in efficiency, productivity and quality.

Distributed software development in an offshore outsourcing project: A case study of source code evolution and quality

“…Apart from a change in number of companies involved, a handover of project management responsibilities to an offshore location is another change that occurred in the present studied case. In a previous study Jabangwe et al (2014) investigated the implications of a handover to an offshore location belonging to the same company after a long period of distributed development, and observed a negligible link of the change with quality. The study piloted comprehensive data analysis and focused on collecting quality measures, defects and internal source code measures (e.g., the complexity measure, (McCabe, 1976)).…”

A method for investigating the quality of evolving object‐oriented software using defects in global software development projects

“…This modification included using input from a systematic literature review that was performed to identify empirically evaluated source code measures for object-oriented software systems. Overall, this modification resulted in an updated method, which was then applied in another empirical study that is presented in [16]. Further refinements were then performed based on insights and lessons learned from the application of the method in that study.…”

“…It also details data collection and aggregation at a level that matches the development setting in the GSD context. The method focuses on object-oriented programming languages and is based on previous empirical research by the authors (e.g., [3,16]). A method engineering approach [17] was adopted in the incremental development of the method.…”

“…The method is based on lessons from empirical studies presented in [3,16,18] and [19], and the data collection methods for each study are presented in detail in each study. Included in the present paper, apart from the complete method, are details of how the method was developed and how it was refined, as well as how the lessons learned from each individual empirical study helped in the development of the method.…”

“…We, therefore, gained insights through the individual studies, and this enabled us to construct a systematic and coherent method by following a method engineering approach. Therefore, the current version of the method presented in this paper is a synthesis and aggregation of lessons learned from GSD empirical studies presented in [3,16,18] and [19]. Although we do not invent the method fragments and the fundamental properties of the method (e.g., the importance of understanding context and the goal-oriented approach), the novelty of the method is in how it combines them into a coherent structure and the insights that are obtained in following the method.…”