Less than 1 per cent of researchers publish at least one paper every year, a study has found.

These researchers form an ¡°influential core¡± whose articles make up almost 90 per cent of the most cited work. The papers they produce also account for more than 40 per cent of all those published, according to the research by John Ioannidis, professor of health research and policy at Stanford University, Kevin Boyack, president of metrics analysis firm SciTech Strategies, and Richard Klavans, owner of SciTech.

The trio used Scopus, Elsevier¡¯s abstract and citation database, to examine the number of publishing researchers and the frequency of their output between 1996 and 2011.

They found that more than 15 million researchers had published work held in the database over the 16-year period. But only about 150,600 individuals ¨C less than 1 per cent of the total ¨C had authored at least one paper in each of the years.

³ÉÈËVRÊÓƵ

Overall, these researchers have been involved as authors in about 42 per cent of the 25 million papers listed in Scopus for the period studied, Professor Ioannidis, Mr Boyack and Mr Klavans say in an article published in Plos One on 9 July.

¡°This 1 per cent of scientists defines a very influential core,¡± they write, who are ¡°far more cited than others¡±.

³ÉÈËVRÊÓƵ

Most of the researchers who publish every year ranked highly on a citation metric known as the h-index, which offers a measure of the productivity and impact of a scholar¡¯s work. Almost three-quarters of them had an h-index of 16 or more, which is regarded as the ¡°hallmark of a successful scientist¡±, the authors explain.

Further analysis suggests that 87 per cent of the most highly cited papers in the database came from authors in the 1 per cent of scientists who had published regularly and without breaks.

¡°Only a very small fraction of researchers have an uninterrupted, continuous presence in the scientific literature and these investigators account for the lion¡¯s share of authors who eventually have high citation impact,¡± the authors write.

Perhaps unsurprisingly, researchers working in fields where the accumulation of relatively small bits of new knowledge is valued, such as medical science, accounted for the majority of those in the influential core in a random sample of the data.

³ÉÈËVRÊÓƵ

¡°Given that there are many thousands of universities and research institutions and each has tens and hundreds of teams and departments, the concentration of 150,000 researchers can quickly get rarefied. Many teams, departments or even whole institutions may have none or very few researchers who belong to this core,¡± the authors say.

They add that the widespread interruption of publishing for the remaining 99 per cent of researchers may be a sign of ¡°system inefficiency¡±.

The authors say that this group is likely to be home to ¡°very different¡± categories of researchers, including scholars in the social sciences and humanities whose scholarly communication is predominantly via monographs, those who have taken a break to have and look after children, hospital clinicians and trainees.

holly.else@tsleducation.com