On the current obsession with publication statistics

Abstract

Crude publication statistics such as publication counts and impact factors are routinely being employed to assess individuals and institutions. Although they can play a role in an approximate preliminary assessment, using them for anything more is inappropriate due to their over-simplicity and ease of manipulation. Furthermore, it is argued that rewarding scientists for achieving high scores in such number-based evaluations ultimately leads to a slowing of scientific progress. Suggestions are given on how reliance on statistics can be reduced and their manipulation discouraged.