Reader Comments

Post a new comment on this article

Doesn't work for economics

Posted by jbrueckner on 02 Sep 2019 at 14:14 GMT

The article and its findings are impressive, but the method doesn't really work for a field like mine (economics), where alphabetical ordering of author names is the norm. Some of my distinguished colleagues rank a lot lower than me simply because their last name is closer to the end of the alphabet. It would be interesting to see the results with the name-order-based citation counts dropped in the computation.

No competing interests declared.

RE: Doesn't work for economics

John_Ioannidis replied to jbrueckner on 02 Sep 2019 at 15:39 GMT

As we show in the analysis, social sciences, humanities and mathematics have the highest proportion of alphabetically ordered papers, but even in these disciplines alphabetical order remains a minority. Most economics research is clustered within the social sciences group. If a sub-field uses alphabetical order routinely, it is possible to exclude the author-order indicators from the calculations. In our recent application of the composite metric across Scopus (PLoS Biology 2019), we provide detailed data on top-ranked 100,000 scientists with values on each of the 6 indicators, so one can explore different combinations. However, in principle, these post hoc should be seen as exploratory and may not necessarily do better overall than the composite even in these subfields where some alphabetic ordering is not uncommon. Single cases of a few scientists may not suffice for getting a good sense of validation performance.

No competing interests declared.