Project Details
Projekt
TeamDynamics: Interaction and Communication in Distributed Software Teams
Subject Area
Software Engineering and Programming Languages
Social Psychology, Industrial and Organisational Psychology
Social Psychology, Industrial and Organisational Psychology
Term
from 2015 to 2021
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 263807701
Motivation and Background: Today, software is developed in teams. Communication is a key driver for project success, as many studies show. The dynamics of such a socio-technical system is difficult to understand and foresee, due to its high complexity. In the first phase (TeamFLOW) of this project, a number (N=42) of similar student software projects were observed and analyzed in detail. The teams, their communication and behavior were investigated in a longitudinal study. Software engineers and psychologists collaborated closely in the observation of behavior through questionnaires and success criteria. A number of correlations were identified; for example, the concept of FLOW-distance was found to be related to success parameters. In addition, based on pattern recognized, behavior in a later project phase could be predicted within a class of projects that were similar. New Challenges: The original TeamFLOW research focused on similar software projects with simple and comparable communication structures. This homogenous set of projects was well-suited to study correlations between the remaining parameters, such as team mood, media usage, and communication distance (indirection). In the extension, called TeamDynamics, existing limitations will be overcome: - Generalization to other types of projects and communication structures: Student projects were very similar in duration, team size, and experience background of participants. In TeamDynamics, bigger teams, distributed projects, and a variety of agile vs. plan-driven processes will be investigated.- Development and evaluation of prediction models for this extended class of projects, in order to identify symptoms and predict selected aspects of project dynamics. This should happen early enough to support decision making. Prediction models will extend beyond student projects, simple communication structures, and fix process models.- Development of new instruments: The transition from testing hypotheses with a sophisticated analysis technique (in TeamFLOW) to simplified techniques will lay the foundation for applying the insights for operational decision making within a running project.TeamDynamics research vision: The insights and correlations found during TeamFLOW will be generalized to a larger and more diverse set of software projects. The foundations will be laid for operational application of the analysis techniques. Due to the complexity of the analysis, its results were often delivered after project completion (in retrospect). This was acceptable for gaining insights and quantifying correlations. However, results came too late for supporting team leaders in their decisions during essential project phases. The research vision of TeamDynamics is to simplify, adapt or replace analysis techniques and evaluate whether communication and success aspects can still be predicted in real time. This would allow interventions and recommendations during a running project.
DFG Programme
Research Grants
