NCB demography – democracy?

This journal rejects over 90% of received manuscripts. Our thorough evaluation process is designed to ensure that all decisions are fair and informed (see July editorial DOI:10.1038/ncb0703-583). Nevertheless, some authors of a rejected manuscript might be tempted to look for reasons unrelated to the data presented to explain a rejection, be it hypothetical, geographical, institutional, or indeed individual, biases.

Figure 1

We take pride in looking at the science beyond the authorship, indeed it is clear that for any journal the science ultimately speaks for itself, and it would be a futile exercise to base editorial criteria on anything but the quest to publish the best, most exciting and the most thorough science. Here, we have compiled a few statistics to address some of the more frequent rumblings of biases we hear about.

Regional: fact or fiction?

The scientific community contributes greatly to a journal at two levels: the submission of primary research and peer review. The bar chart displays data for several representative countries from a survey of just under 3,000 manuscripts considered at NCB between January 2002 and October 2003. The data closely reflects that of an independent survey carried out on 500 manuscripts submitted to this journal between December 1999 and May 2000 (third column). It should be noted that both primary research papers and refereed reviews are included (although the latter are a minority not likely to affect the main trends dramatically).

Not surprisingly, the US leads the pack with around half of submissions (first column), closely matched with the overall acceptance rate (second and third columns). The UK contributes close to 10% and excels with a rather high acceptance rate (14.4% of all manuscripts and 15% of UK submissions). Regionalism is not at work here, as Germany (and several other European countries) match this with around 13% of the countries' submissions accepted. The US is only successful 9% of the time, but note that this single country still fills half the pages of NCB. Japan submits around 10% of all manuscripts, with a slightly lower contribution to papers in the journal of 7.4%. Nevertheless, this is significantly up from the 4.4% observed in 2000. Overall, this translates to a 6% acceptance rate for Japan, not dissimilar to Canada (7%), a country with a scientific tradition closer to that of the US.

Fair refereeing

Despite our assurances to the contrary, we occasionally face suggestions of differential treatment on the basis of seniority, fame, or simply personal acquaintance. This is far harder to analyse quantitatively. The simple problem is that undeniably, eminence in a scientific discipline is causally linked to publication in top journals. Equally, there is a statistical likelihood that we will ask a researcher with a well-established broad research programme bridging multiple fields to referee more frequently than the budding principal investigator working on a singular problem.

We take great care to ensure an optimal distribution of referees according to complementary expertise. However, we also aim to match seasoned referees with the more junior faculty. Indeed, we note the absence of a clear correlation between scientific eminence and ability to generate astute and constructive referee reports. Often the lack of experience of refereeing is more than compensated for by the 'hands on' experience of the bench scientist. Finally, we increasingly aim towards an even geographical distribution of referees, but not at the expense of recruiting the most appropriate referees.

Of 2,037 referees assessed at NCB, just over 60% are based in the US, 13% in the UK, and 3–4% each in France and Germany. This distribution is similar to that observed at Nature magazine. Although the UK is slightly over-represented and Germany slightly under-represented, overall this is not dissimilar to the submission statistics presented above, which in turn are likely to be a fair guide to the research activities of these countries. The real outlier is Japan, with under 2%, sevenfold and fivefold less than Japan's submission and acceptance rates at NCB, respectively. This is a complex topic, and part of the low rate may be based historically on language and cultural issues. Indeed, this trend is still echoed in the speaker lists of international conferences and editorial boards. The peer review process is a critical part of participation in the global scientific community and a countries' referee activity ought to be roughly in line with its scientific output. Therefore, we state for the record that we are going to be actively pursuing higher referee participation from Japan and we invite nomination of critical and authoritative reviewers. It is noteworthy that the geographical distribution of our most frequent referees closely reflects that of the overall referee pool.

We conclude that publication rates and refereeing rates at NCB are in line with global research activity and hope to thereby dispel any notions of systematic geographical, or other, biases. In the end, the science speaks for itself.