The goal of Reanalysis and Replication is to improve knowledge through building on previous research. The authors focus on a journal article or a book which they reanalyze or replicate. Below, we call it “the reference study.”

A reanalysis manuscript uses the same data as the reference study uses and applies new analysis to it. For instance, the authors may change specification (e.g., adding new control variables and/or interaction terms) or statistical model (using new probabilistic distribution other than normal). Reanalysis may discover a new finding, challenge the original study conclusions, or confirm that the original results are robust to additional specifications. The authors should explain the merit of the reanalysis in terms of its contribution to better understanding an important question of substance and/or methodology.

A replication manuscript creates a new dataset and conducts the same analysis as in the reference study. The authors may apply the same analysis of the reference study to a different empirical domain (e.g., different countries and/or different time periods). Alternatively, the contributors may repeat the same experiment on a different sample. The replication category is intended to provide a forum in which the scholarly community examines the external validity (or lack thereof) of the reference study. As with reanalysis, the authors should explain the merit of the replication in terms of its contribution to better understanding an important question of substance and/or methodology.