The impact factor as a measuring tool of the prestige of the journals in research assessment in mathematics

The (2-year) Impact Factor of Thomson-Reuters (IF) has become the fundamental tool for analysing the scientific production of academic researchers in a lot of countries. In this article we show that this index and the ordering criterion obtained by using it are highly unstable in the case of mathematics, to the extent that sometimes no reliability can be assigned to its use. We explain the reasons of this behaviour by the specific properties of the mathematical journals and publications, attending mainly the point of view of the researchers in pure mathematics. Using the Journal Citation Report list of journals as a source of information, we analyse the stability in the position of the mathematical journals—the so-called rank-normalized impact factor—compared with journals in applied physics and microbiology during the period 2002–12. Due to the lack of stability of the position of the journals of mathematics in these lists, we propose a ‘cumulative index’ that fits better the characteristics of mathematical journals. The computation of this index uses the values of the IF of the journals in previous years, providing in this way a more stable indicator.This work was supported by the Ministerio de Economía y Competitividad (Spain) [CS02012-39632-C02 to A.F.S, F.P., R.A.B.] and [MTM2012-36740-C02-02 to E.A.S.P.].Peer Reviewe