Problems of scientometrics: Maximum cited publication index as a means of solving some of them

The paper presents a critical analysis of scientometrics as an indicator of scientist’s effectiveness. Six points of criticism of the main scientometric indices clearly show that scientometrics cannot be an indicator of such efficiency. In particular, attention should be paid to the problems of accrual, PR, negative citation, citation of non-scientific publications, publication only in trend areas, co-authorship, and article lifetime. An empirical analysis of the effectiveness of scientometrics is conducted and its ineffectiveness is substantiated. It is explained why, despite all the shortcomings of scientometrics, it continues and is likely to be used in the assessment of scientific activity. It is argued that not many scientists in their lives separately make significant discoveries, most of them work in scientific teams and narrow areas of science. On this basis, the ‘maximum cited publication index’ — the MCP index — is introduced. This index is calculated as the number of citations of a maximally cited work divided by the number of its co-authors. It is proved that, on the one hand, the MCP index really better shows the scientist’s contribution to science, and, on the other hand, it will solve some problems of scientometrics, such as the problem of co-authorship and accrual. An empirical verification of the MCP index is given.