Improving Our Field Through Code and Data Sharing

Abstract

Across academia, there is a series of ongoing and intertwined debates about the need for research transparency, greater confidence in the accuracy of empirical findings, and the overall relevance and credibility of our work (for a summary, see Bergh, Sharp, Aguinis, & Li, 2017). Central to these debates is the idea that, as a field, we need greater confidence in the cumulative body of scientific knowledge created by our research. To make progress toward this goal, we need processes that minimize the publication of flawed results emanating from honest errors; insufficient training; and, in the worst of cases, fraud. Many stakeholders are pressing for initiatives related to these challenges. For example, some journals, and those responsible for their oversight, debate the merits of code and data sharing, whereas others are implementing policies that encourage or even require such practices (for a review, see Dosch & Martindale, 2020). Similarly, some grant-awarding organizations require a level of code and/or data transparency in exchange for funding (for a discussion and examples, see Global Innovation Fund, 2021; Metzenbaum, 2021). There is also a growing recognition that replications are integral to the knowledge generation process (e.g., Köhler & Cortina, 2021). Advocacy organizations have also arisen to facilitate the warehousing of code and data related to research efforts (e.g., FIVES Project; Center for Open Science). In this editorial, the Journal of Management has asked us to discuss and describe a related but slightly different approach. As a means of making the research process more efficient,