Several studies have looked at the RG score, for which details about how it is calculated are not published. These studies concluded that the RG score was "intransparent and irreproducible", criticized the way it incorporates the journal impact factor into the user score, and suggested that it should "not be considered in the evaluation of academics". The results were confirmed in a second "response" study, which also found the score to depend mostly on journal impact factors. The RG score was found to be negatively correlated with network centrality, i. e. , that users that are the most active (and thus central to the network) on ResearchGate usually do not have high RG scores. It was also found to be strongly positively correlated with Quacquarelli Symonds university rankings at the institutional level, but only weakly with Elsevier SciVal rankings of individual authors. While it was found to be correlated with different university rankings, the correlation in between these rankings themselves was higher.