DHB: Investigating socialization of new members into self-organizing technology-supported distributed teams


Increasingly, organizational work is performed by distributed teams of interdependent knowledge workers. Such teams have many benefits, but geographic, organizational and social distance between members makes it difficult for team members to create the shared understandings and social structures necessary to be effective. These distances are particularly problematic for new members seeking to join the teams. But as yet, research and practitioner communities know little about the dynamics of socialization in distributed teams, which our literature review suggests are likely to be substantially different from those in others kinds of organizations. These dynamics are particularly challenging when teams have the autonomy or responsibility to self-organize (e.g., in teams that span multiple formal organizations). The goal of our study is to better understand the cognitive and social structures that underlie changes in individual and team behaviors in these teams as this mode of work is becoming increasingly more common. Our study addresses the general research question: What are the dynamics by which new members are socialized into self-organizing technology-supported distributed teams?
To study the dynamics of self-organizing distributed teams, specifically new member socialization, we propose a multi-disciplinary and inter-disciplinary study that integrates the analysis of multiple sources of data using multiple research methods. We first review the literature on socialization in conventional organizations and in FLOSS projects to develop a theoretical framework to guide our study. We will use a combination of human coding, natural language processing (NLP) and social network analysis (SNA) to analyze large quantities of developer email and chat logs. We will correlate these findings with analysis of the software structure of the code produced by the teams to understand the effects of the team dynamics on the teams’ output. FLOSS teams provide a perfect setting for such a study because large quantities of interaction data and program source code are readily available for study.

PDF icon 2008_NSF_HSD.pdf1.04 MB