Time heterogeneity in stochastic actor-oriented models: an empirical test on problem-solving dynamics in an open source software project
Conaldi, Guido and Tonellato, Marco (2011) Time heterogeneity in stochastic actor-oriented models: an empirical test on problem-solving dynamics in an open source software project. In: XXXI Sunbelt Social Networks Conference, 8-13 Feb 2011, St. Pete Beach, FL, USA. (Unpublished)Full text not available from this repository.
How do organizations induce and coordinate individual action toward collective goals? The interest of organizational and sociological scholars in this question has been renewed by the emergence of communities of production composed by volunteers who can choose which task to take on, why and when. Relatively little is known about the social micro-mechanisms that drive individual action and ensure coordination between interdependent tasks when traditional market-based and hierarchy-based mechanisms of coordination are unavailable. In previous research we applied stochastic actor-oriented models to analyze one of such communities. We showed that decisions of individuals to take on specific tasks are influenced by their local network neighborhood through feedback mechanisms, which shape the different levels of engagement across individuals, and balancing feedback mechanisms, which affect the different popularity of tasks.
In this study we investigate whether the effect of these local network structures on the association of individuals and tasks varies according to when it unfolds in the life-cycle of the project. In order to address this question we analyze the same Free/Open Source Software (F/OSS) community of our previous study. We reconstruct the two-mode network generated by problem-solving activities undertaken by 135 software developers on 719 software bugs during an entire release cycle of the software. We estimate new stochastic actor-oriented models by allowing the effects of local networks structures to be heterogeneous over time. We discuss the differences with previous restricted models and their implications for the understanding of social micro-mechanisms underpinning the emergence of endogenous coordination in communities of production.
|Item Type:||Conference or Workshop Item (Speech)|
|Additional Information:|| The International Sunbelt Social Network Conference is the official conference of the International Network for Social Network Analysis (INSNA) ) Guido Conaldi also chaired a session entitled Network Dynamics on the Wednesday afternoon of this conference.|
|Uncontrolled Keywords:||Stochastic Actor-Oriented Models (SAOM), time heterogeneity, free/open source software|
|Subjects:||Q Science > QA Mathematics > QA76 Computer software|
|School / Department / Research Groups:||School of Business|
School of Business > Centre for Business Network Analysis
|Last Modified:||12 Jul 2012 15:45|
Actions (login required)