By TIM WINKLER

Analysis of millions of peer-reviewed papers has shown that it is better to choose a relatively unproductive but highly cited researcher to co-author your next paper than a highly productive one.

Professor Mike Thelwall from the University of Wolverhampton in the UK, has published a paper in Scientometrics analysing the outcomes of millions of articles published in Scopus in 2017 by authors from Australia, Ireland, New Zealand the UK and USA. He found that first author citation impact was more important than author productivity in achieving more highly cited research.

Using the number of citations as a proxy for research quality, Professor Thelwall set about analysing publication data in the hope that information on selecting co-authors would be useful to young researchers wanting to build their career.

The results indicated that a young lead author would typically be better to find a collaborator with low productivity but a high number of citations, rather than a researcher who was more highly productive, if they wanted their article to be likely to be more highly cited.

The paper raises a host of questions in relation to management and productivity of academic staff at all levels, and also a fundamental tension between driving better performance in rankings and researcher productivity.

Universities seeking higher rankings are more likely to have success by building a stable of less productive, highly cited researchers than a workforce of younger, more prolific researchers.

There are disciplinary differences in the results – first author productivity tends to be a disadvantage in the physical science and life sciences, which probably reflected the reliance on PhD research to break new ground in these fields, Professor Thelwall said.

Further research would be required to understand the cause of these phenomena, and also the implications for research income and impact, he said.