Abstract & Overview
This study investigates whether available replication data leads to increased citation impact for political science publications.The research question is part of a broader movement towards scientific openness that we can trace throughout history.Technological advances bring this movement into the modern times, with increased possibilities of sharing of empirical materials.Whether this data sharing is beneficial for individual scholars is still largely unanswered, and there are shortcomings with the studies investigating the relationship between available replication data and citation impact.This thesis intends to improve on the existing research.The starting point is the article "Posting your data: will you be scooped or will you be famous" by Gleditsch, Metelits, and Strand (2003).Their model attracted serious criticism, especially from Abbott (2007), who objected not only to the solidity and validity of their model, but also the interpretation of the results.Keeping his objections in mind, I attempt to go further and explore not only whether there is a relationship between replication data and citation impact, but also make some observations as to the causes of this relationship.To do this, I anchor the analysis deeper in existing theory and previous research, and thus include more (previously missing) aspects of publishing and citing.Also, because of the criticism of the data used by Gleditsch, Metelits and Strand, I add a second data set available from Dafoe (2014).The new data allows me to investigate the finer points of data sharing, as the information on replication data is more detailed.This study includes more data, better methods, and more aspects of publishing than previous research on the same subject, but confirm the findings of this existing literature (Dorch 2012;Gleditsch, Metelits, and Strand 2003;Ioannidis et al. 2009;Piwowar, Day, and Fridsma 2007;Piwowar and Vision 2013).The analyses indicate that there is a relationship between data sharing and citation impact.