Do Negative Replications Affect Citations?

This study examines the effect of negative replications on the citation rates of replicated studies. We study a set of 204 replicated studies and compare their citation performance with an initial sample of 112,000 potential controls taken from Scopus. Approximately half of the replication studies refuted key findings from the original studies, with the remaining half either providing a confirmation or a mixed conclusion. Using matching criteria that accommodate differences in the lengths of time between publication of the original study and its replication, as well as differences in the number of citations, we match each replicated study with multiple controls based on having comparable citation histories. Our main samples consist of 74, 103, and 142 replicated studies and 7,044, 7,552, and 11,202 matched control studies, respectively. We have two main findings. First, studies that are replicated receive somewhat more citations than their matched control studies. Second, there is no evidence that studies that receive negative replications suffer a penalty in the form of fewer citations.