Author:
Correia Luiz Gabriel,Pascual Mena-Chalco Jesús
Abstract
The manipulation of the Journal Impact Factor (JIF) through editorial decisions that increase self-citations has been empirically observed in recent studies, sparking discussions on how to avoid such practices. This study aims to verify whether a policy of removing self-citations from the JIF calculation is enough to eliminate incentives for manipulation. We introduce an agent-based model of the academic social network formed by journals to test this hypothesis. We model journals as rational agents seeking to gain positions in the JIF ranking. The model was implemented in NetLogo and calibrated using data from the SCIMago Journal Ranking. Simulation results demonstrate that the model is capable of reproducing patterns observed in empirical networks and that the removal of self-citations can significantly reduce the adoption of manipulation strategies (e.g., reduction from 90% to 30%). This study contributes to understanding the role of self-citations in the JIF and provides tools for developing policies that promote scientific integrity.