Retraction: The "other face" of research collaboration

There is an increasing amount of research investigating retractions. Yet little attention has been paid to the relationship between retractions and collaboration. In this paper, we test the hypotheses regarding the relationship between retractions and collaboration on a unique publication dataset of retractions and its control group constructed by the nearest-neighbor-matching approach. Our analysis finds no significant evidence indicating that collaboration suffers from producing flawed research, at least in the form of retraction. We also find that ceteris paribus publications with authors from elite universities are less likely but more quickly to be retracted. There also is no significant impact of collaboration size on the speed of retraction of Chinese articles, although China stands out with the fastest retracting speed. Our findings have policy implications for the governance of global science, especially that involves collaboration.