Individual Submission Summary
Share...

Direct link:

Lessons from Project TIER: Teaching Integrity in Empirical Research

Fri, August 31, 8:00 to 9:30am, Marriott, Salon H

Abstract

Although much work remains to be done, the last decade or so has witnessed substantial improvements in the standards for documenting empirical research achieved by the economics profession. Much of this progress can be attributed to the many initiatives that have been undertaken for the purpose of enhancing the reproducibility of data processing and analysis. In this paper, we argue that, thanks to these improvements, the printed (or pdf) article is becoming obsolete as the primary medium for communicating empirical research, and should be replaced by the “research compendium,” which includes the article in its traditional form, but also includes documentation containing all the information necessary to reproduce the results reported in the article. The two components of the research compendium—the article and the documentation—should be of equal status: authors should take as much care in preparing transparent and persuasive documentation as they do composing the article; and for someone interested in learning about the research, opening up the documentation and running the code should be considered just as essential as reading the paper itself. The idea that research compendia, containing the research article as well as complete replication documentation, should replace the article by itself as the medium for communicating research has been proposed in the computational and experimental natural sciences; we argue that the idea is equally applicable for empirical social science research.

Author