Scientific Ethics 2(2): 44-45, 2007	MISCONDUCT EXPOSED
http://im1.biz
© Truthfinding Cyberpress	THE IMPACT FACTOR SCAM

Hwang’s Retracted Publication Still Contributes to Science’s Impact Factor

Shi V. Liu

Eagle Institute of Molecular Medicine

Apex, NC 27502, USA

Corresponding with SVL@logibio.com

(Received 2007-04-26; accepted 2007-05-01; published 2007-05-02*)

HIGHLIGHT

To a significant extent the retracted publications earn “top” journals more citations than other equivalent publications and thus contribute to their unusually high “impact factors”. However, knowing this cheating component in their impact factors, “top” journals have not done anything to correct it.

ABSTRACT

Impact factors are based on the total citations collected for a given period on those so-called “citable” publications. Many “citable” publications are those later retracted publications. Ironically those retracted publications often earn more citations than the other equivalent publications. However, while the publications are retracted, the citations were not subtracted from the impact factors. Thus, the unusually high impact factors of the “top” journals to some degree represent an outright cheating. But such scam is not only remained as uncorrected but becomes even forceful in citation-based evaluation of scientific publishing.

KEY WORDS

Citation, Impact factor, Retraction, Cheating, Scam, Misconduct, Top journal, Hwang scandal

In some earlier publications I concluded that the unusually high impact factors for some “top” journals are based on some useless or even misleading citations (Sci. Ethics 1: 72-73 and 83-84, 2006). This is because not only the true meanings of these citations are not reflected by the numbers summarizing them but also the retracted publications often earn more citations than non-retracted similar publications (Logical Biol. 5: 125-127 and 128-130, 2005).

Even with some high-profile and thus well know retraction notices, the retracted papers are usually still big contributors to the impact factors of their journals.

I recently (on April 10, 2007) re-downloaded the PDF file of the famous or infamous Hwang stem cell discovery or fakery paper in Science. Thanks to the new citation revelation feature offered by Science I easily saw the citations for this retracted paper. The citations collected by ISI Web of Science and HighWire Press for the retracted Hwang paper were 148 and 35, respectively which is much higher than the average citations of all “citable” publications in science.

A more objective comparison would be a comparison between the citation numbers for the Hwang’s paper and the citation numbers for the other reports published in the same issue. There are a total of 13 Reports published in that issue and only 10 of them contained citation information (I do not know why the other 3 Reports did not contain citation information and thus have excluded them in my following analysis).

The citations collected by ISI Web of Science and HighWire Press for the other 9 papers were in the range of 5-48 and 1-14, respectively, which are much lower than those for the Hwang paper. Although some scientific discipline usually have low citation rates the citations collected by ISI Web of Science and HighWire Press for other 4 biology Reports were still much lower (in the ranges of 25-30 and 5-9, respectively) than the Hwang paper. There were a total of 377 and 89 citations collected by ISI Web of Science and HighWire Press for all the 10 Reports and the contributions of Hwang’s paper are 39.26% and 39.33%, respectively.

So Hwang’s retracted paper is still earning Science some big impact. When Science receives its new impact factor covering the period of the Hwang scandal it should ask ISI weather the contributions by Hwang (and other publications it retracted) were removed from the citation counting and impact factor calculation. If not, Science has a right to demand a new truthful impact factor. If Science does not do so, then it is cheating public in a very big way.

* This letter was submitted to Science on 2007-04-26 and was rejected on 2007-05-01. The publication here is the same as submitted to Science except for the added highlight, abstract and keywords.