Influence of different journal publishing models in the presence and detection of scientific errors and misconduct

Abstract

This study attempts to test how different journal publishing models can favor or reduce the presence of errors and misconduct articles, as well as to measure the response of journals to problematic articles according to these publishing models. For this, a new approach for the study of scientific misconduct in publications is proposed.comments expressed in PubPeer about 17,244 troublesome articles were compared with the editorial response of journals (i.e. editorial notices). Journals of these publications were classified according to several publishing criteria: publisher type, access type, publication fee model and peer review type. The results show that in spite of scholar-published journals suffer more from problematic papers, they release the same editorial notices than commercial journals; open access journals react better to problematic articles than paywall journals; open access journals without APC has a special presence of Publishing fraud; and journals that use open review suffer less from misconduct, slightly releasing more editorial notices.