12.040

Of course we're pleased. We want to shout it from the rooftops. Nature Physics has its first impact factor, and it's a good one.

Nature Physics' first impact factor — counting the citations through 2006 of those papers published in our first three issues, from October to December 2005 — has now been announced. We had hoped that we might make double figures; we didn't guess that the number would be quite as high as 12.040 (ref. 1). Among physics journals, that puts us second only to the mighty Reviews of Modern Physics; but first among physics journals publishing primary research, and second to Nature Materials in the wider category of primary titles in physical science.

The publication of the impact factor puts Nature Physics on the map. From our launch in October 2005, we set out to publish (as the blurb says²) “papers of the highest quality and significance in all areas of physics”. That was the plan, and it's been quite a wait to see its execution vindicated in the impact factor. We've often fielded enquiries from physicists, considering submitting their work to Nature Physics, about the value of our impact factor. It is a number that matters.

We do, however, recognize that the impact factor is a very imperfect measure of a journal — not least in the case of Nature Physics because we publish more than primary research and review articles; the commissioned content of our many 'front-half' sections, including Commentary and News & Views, is a vital part of the journal's identity. Citations may also be a misleading guide to the merit of a scientific paper, as disputed or erroneous results can be highly cited in the community's effort to set the record straight.

There are now many other measures on the market, although these are still based in some way on citations. For example, there is the recursive algorithm, as implemented in Google's PageRank facility³ and more recently in Eigenfactor⁴ from the University of Washington, in which citations made from journals with high impact are given greater weight than those from lower-impact titles. Or there is the h-index — most commonly applied to individual scientists, although it works for journals too — which is the number of publications having at least that same number of citations (that is, n papers having at least n citations gives an h-index of n). Each measure has its own merits and shortcomings, perhaps favouring authors or journals who publish a little, or those who publish a lot; or suffering the skew of self-citations. None is perfect, and a solid assessment would take several factors into account. For general bean-counting purposes, however, the impact factors calculated in the ISI Journal Citation Reports¹ (from the US publishing corporation Thomson) are hard to beat. These widely touted numbers still have precedence; the responsibility for maintaining correct information in the burgeoning databases of ISI is huge.

At Nature Physics — where a team of four professional editors oversee the peer-review process, with no editorial board — we try to keep the citation-potential of manuscripts out of the decision-making process. Anyway, it can be difficult to second-guess at the submission stage how highly cited a paper might be, once published. Although long-awaited results in citation-heavy fields are obvious winners, for most papers it's hard to tell. Of our clutch of publications in 2005 (as this editorial goes to press), two papers are currently tied with the highest number of citations (51 each): Spatial imaging of the spin Hall effect and current-induced polarization in two-dimensional electron gases⁵ and Matter-wave interferometry in a double well on an atom chip⁶. In retrospect, it is probably unsurprising that the paper on the spin Hall effect has done well — it's a 'hot' area of physics — but that the atom-chip paper could match it for citations was perhaps less obvious.

Indeed, there's particular pleasure (for editors, at least!) in the 'sleeping beauties' — those papers whose value is only realized further down the line, and whose citations only accumulate much later, beyond the two-year window following the year of publication in which a paper contributes to an impact factor. It's still early days for Nature Physics, but we watch the evolution of citations in the ISI database with interest.

Interest is, after all, our central criterion for publication in Nature Physics — that the results reported in a paper, supported by peer review, will be of interest to the wide community of physicists, across many disciplines. That might not always translate into high impact, but the content of the journal will always be, we hope, 'interesting'. We don't know how our impact factor will evolve in future years (although we hope to follow the example of our sister title Nature Materials and move steadily upwards). Right now, we're glad to have made an impact.