ABSTRACT
In this chapter, the authors discuss the importance of reproducibility and transparency in computational social science, followed by a discussion of how these ideals are complicated by the realities of doing rigorous collaborative research in their field. Many researchers and organizations have begun to innovate in response to the challenges presented by the need for collaboration and transparency in computational social science. The field of organizational management had cause to re-examine adhocracy; despite its archaic origins, the structure it describes maps well onto the agile development processes frequently employed by contemporary technology firms. By way of a potential solution, the authors introduce Dr. Patrick Ball’s implementation of a ‘principled data processing’ framework – a set of principles and practices to ensure that social scientific projects are transparent, auditable, reproducible, and scalable, regardless of the number of collaborators. They detail the functionality built into pdpp and provide step-by-step instructions for replicating a practical sample workflow.
