Unintended Impact of Author Impact Factor

The Letters in the March 2005 issue of PHYSICS TODAY (page 12) in response to Mohamed Gad-el-Hak's Opinion piece (March 2004, page 61) on citation rates and impact factors show how important these criteria have become for hiring, tenure, and promotion, and suggest some models that may result in undesirable, unintended consequences. In particular, the suggestion by Loc Vu-Quoc that multiple-author publications be divided in some fashion according to the number of authors might result in having nervous faculty members delete students and important support staff as coauthors and relegate them to acknowledgments.

The notion that all coauthors are equally responsible for content is not valid in many fields; in solid-state physics, crystal growers, with or without PhD degrees, are not technicians but highly skilled collaborators of equal standing, and students may often play a more important role in that field than in theoretical physics. When I was at Bell Labs (1966–72), no one thought that Howard Guggenheim or Joe Remeika should be responsible for the detailed theoretical analyses of data on their superb crystals, but it would have been unethical not to list them as coauthors; they had grown the world's best specimens of new materials.

The law of unintended consequences has many examples in life; one such story, albeit apocryphal, is that of rat extermination in Singapore. According to the anecdote, a bounty of, say, a few cents was offered for each dead rat turned in to the authorities. Within days numerous rats were delivered, and the numbers dropped quickly as the extermination neared completion. Surprisingly, however, after two weeks the numbers suddenly shot up. Young boys were breeding rats! In a similar vein, if the formulas Vu-Quoc proposes were implemented, we might see a sudden explosion in the number of short, single-author publications by untenured faculty members. Probably these would have about the same value as the rats in Singapore.

We must be careful what we recommend.

Vu-Quoc replies: James Scott describes a knee-jerk reaction of short-sighted authors who focus on getting more credit for a single paper but lose sight of the bigger picture.

Ethical guidelines such as those of the American Chemical Society¹ clearly state that "the coauthors of a paper should be all those persons who have made significant scientific contributions to the work." Most authors would follow these guidelines and share the credit—and sometimes the blame—for the work.^2,3

Many journals already require each author of a paper to state his or her contribution.⁴ Coauthors are sometimes listed for ethically questionable reasons.⁵ Inflated authorship, like inflated grades, devalues authentic authorship, does not contribute to good education, and misleads potential employers. The author impact factor (AIF) is a statistical average over a collection of papers. Its unintended consequence is to promote effective and genuine collaboration, good collaborative work, and adherence to the ethical guidelines for authors.¹

Instead of such narrow issues as, for example, trying to get more credit for a paper, the AIF concept, with its robustness against database errors, addresses much more broadly the challenges of ranking the publication impact (reputation) of heterogeneous groups of researchers—for example, for use in the ranking of doctoral programs.

References

1. 1.American Chemical Society, Ethical Guidelines to Publication of Chemical Research, January 2000, available at [LINK]; see also American Physical Society, Supplementary Guidelines on Responsibilities of Coauthors and Collaborators, [LINK].
2. 2.D. Kennedy, Science 301, 733 (2003) [MEDLINE].
3. 3.J. E. Hirsch, Proc. Natl. Acad. Sci. U.S.A. 102, 16569 (2005) [MEDLINE].
4. 4.N. Wade, New York Times, 17 Jan 2006, p. F2.
5. 5.E. Marris, E. Check, Nature 439, 768 (2006) [MEDLINE]; N. Wade, C. Sang-Hun, New York Times, 10 Jan 2006, [LINK].