Kovscek

During the review of a recent paper, an Associate Editor pointed out that an author had almost exclusively cited proceedings manuscripts rather than the final peer-reviewed, published versions of papers. This happens with some regularity among the manuscripts that we receive for review. Clearly, this is a result of the SPE no-paper, no-podium policy; the large volume of manuscripts produced for proceedings; and the resulting fact that the first version of a manuscript that many authors read is the proceedings version. Also recently, I attended the 2011 SPE Western Regional Meeting in Anchorage, Alaska. There, an author asked me if the 2010 impact factor for SPE Journal had been made available yet. The 2009 impact factor and 5-year impact factor for SPE Journal are 1.089 and 1.455, respectively; the ISI (Institute for Scientific Information) Journal Citations Report to which Stanford University subscribes has not yet obtained the 2010 information.

These two events reminded me that it is important that we as authors, reviewers, and editors make sure to refer to and cite the journal-published versions of manuscripts. Firstly, much author and reviewer effort went into consideration and review of the published version of a paper. Citing the published version acknowledges these efforts and we expect that the final version is actually more polished and complete in comparison to the proceedings version.

Secondly, many metrics that gauge the influence and relative importance of journals make use of citations in their various formulae. The impact factor is the metric most often referred to regarding a journal, even if it is an imperfect measure of a journal's quality.* Generally, the impact factor is the number of citations to articles from the previous 2 years within a journal divided by the number of citable items for those years. Five-year and lifetime impact factors follow a similar formula. Accordingly, when a sizeable fraction of authors refers to the proceedings version instead of the journal-specific published version of a manuscript, our impact factor is less than it might be. As stated previously,* we would like to know ideally how many people actually read any given article, because a true measure of impact is the number of times an article is read. Thus, we are at somewhat of a disadvantage because of the dilution of data caused by the production of both proceedings papers and peer-reviewed manuscripts of the same material. As an example, for the prior two years used to compute the impact factor of SPE Journal for 2009, we published 90 papers, and there were 98 citations to papers appearing in SPE Journal those previous 2 years. If each of the 90 papers had one citation that was updated from a proceedings manuscript to a publication appearing in the pages of SPE Journal, the impact factor of SPE Journal would increase from 1.089 to 2.089.

Similarly, the impact of individual researchers is now being gauged using citation data. For instance, the h-index of a researcher signifies that s/he has published h papers, each of which has been cited by others at least h times.** The h-index does not give significant weight to highly cited papers or to papers that have not yet been cited, and thereby it avoids some of the problems associated with the impact factor. Thus, it may be more difficult to influence an h-index through a few key publications. The h-index tends to be smaller for more junior researchers, because it is not time-limited in comparison to the impact factor. Wikipedia has a succinct entry on the h-index if you are interested in learning more.^† Again, dilution of the peer-reviewed citation record affects the h-index.

What I have tried to convey is that the publication and citation records of engineering and scientific journals and individual authors are scrutinized closely nowadays. The data gleaned is analyzed and metrics are computed. While we may agree or disagree that a given or group of metrics conveys useful unbiased information, these metrics are in some cases used to differentiate the contributions of journals and of various researchers relative to their peers. Some institutions use such publication information when making promotion and other personnel decisions as well as decisions about library subscriptions to journals. Clearly, some journals and researchers find high rankings to be to their competitive advantage. This trend only seems likely to continue.

I hope that you are convinced for the need of accurate and up-to-date citations. If you are not, I hope that you have the patience to bear with the editorial team when we ask you to double-check your citations before final approval of your manuscript. In any event, OnePetro makes it easy to get your reference list correct, as pointed out by the Associate Editor to the authors of the paper mentioned in the opening paragraph of this Summary.

Thank you for reading. I hope that you enjoy this issue and its 20 manuscripts.

* Kovscek, A.R. 2010. Executive Summary. SPE J.15 (3): 586. http://www.spe.org/ejournals/jsp/journalapp.jsp?jid=ESJ&webcurl=%2Fejournals%2Fspe%2FESJ%2F2010%2F09%2FJournal_Article_000313.jsp#.

**Hirsch, J.E. 2005. An index to quantify an individual's scientific research output. Proc. Natl. Acad. Sci. U.S.A.102 (46): 16569–16572. doi: 10.1073/pnas.0507655102.

^† Wikipedia. 2011. h-index, http://en.wikipedia.org/wiki/H-index (accessed 16 May 2011).