Determinants of Collaboration level on Github

Many developers contribute to open-source projects without any monetary pay-offs, which seems to be a counterexample of the “rational agent” assumption in economics. Therefore, The success of the open-source movement has attracted scholars' attention for a rather long time. However, due to data limitations, only a few empirical investigations on open-source project developers’ motivation have been completed in the current literature. By applying advanced machine learning techniques on the GitHub data, This project aims at examining the dynamic evolution of open-source projects. In particular, we intend to address two questions:

• What are the incentives for unpaid programmers to continue working on open-source projects?
• How do the developers’ internal and external social networks affect their levels of contribution?

This research is important both academically and practically. Academically, this project will provide insights and empirical evidence on how developers’ social networks impact their motivations for contributing to open-source projects. Practically, it will help open-source project managers to better organise and motivate developers and foster the open-source community as a whole.

The proposed project is based on the well-documented Application Programming Interface (API) of GitHub (https://docs.github.com/en/rest), the world’s largest source code hosting service provider. It intends to address the above questions and to construct a sampling database of the GitHub network for further exploration of two-fold network analysis among developers.