Inflationary times: The changing profile of journal-publishing statistics

Year	Articles per journal	Authors per article	References per article	Citations per cited item
1990	185.10	3.18	22.68	2.09
1995	206.32	3.67	25.45	2.22
2000	207.86	3.98	28.52	2.26
2005	1.73	4.50	30.95	2.44
2009	2.87	4.83	34.20	2.53
% change	+47.42	+51.89	+50.79	+21.05

The table above offers a profile of certain statistics related to journals indexed by Thomson Reuters in its coverage of internationally influential titles. The figures presented indicate that over the past 20 years, journals have grown fatter, co-authorship has increased, references in papers have expanded, and articles are more cited on average – the last point demonstrating a kind of inflation in citation rates.

The average number of articles per journal has increased by more than 47 per cent over the period. If one assumes that the average length of a published article has remained constant, journals have become fatter to cope with increased activity in global research. The actual number of scientific articles indexed by Thomson Reuters was 590,841 in 1990 and 1,015,637 in 2009 – a rise of some 72 per cent in the past two decades.

Collaboration on research publications has increased, too, based on the average number of authors per article. This statistic has grown from 3.18 authors per article in 1990 to 4.83 authors per article in 2009, an increase of nearly 52 per cent. The past 20 years, it should be noted, have witnessed the emergence of more and more research publications that list hundreds of authors, especially those dealing with large-scale clinical trials and high-energy physics experiments.

Meanwhile, the average number of references in research articles has expanded from 22.68 in 1990 to 34.20 in 2009 – an increase of some 51 per cent. Typical practices vary greatly by field: mathematics articles, for example, list a modest number of references on average compared with immunology or molecular biology. The average number of references for a field’s papers has much to do with the average rates for the field.

The final column shows the average citation rate for all fields of science combined and how this has crept up over time – a more modest increase of 21 per cent over the past two decades, from 2.09 in 1990 to 2.53 in 2009. The figure represents citations in one year divided by cited items from any year.

Citation inflation can be observed across all fields of science and in terms of any number of citing and cited windows. Since the absolute rate of citation increases over time, bibliometricians typically use relative rates of citation – such as for a nation in a field compared with the global rate – to gauge actual performance.

In sum, the statistics presented here speak of the growth of science, increased collaboration and an intensification of activity related to the accumulation of knowledge.

For more information, see http://science.thomsonreuters.com/products/esi