Read the articles; don't count them.

or many years journal impact factor has been used for the evaluation of researchers and research centers. After proposal of h-index by Hirsch, this metric has soon found its place and currently many authorities use it as a substitute for the journal impact factor for assessing the researchers and research centers or even journals. The Iranian Ministry of Health has recently decided to evaluate the faculties of Iranian Universities of Medical Sciences by comparing their h-index. However, the appropriateness of this index is questioned by some researchers. But, before continuing our discussion, let us examine what h-index is. In 2005, Jorge E. Hirsch, a physicist at University of California, San Diego (USCD), proposed a metric for determining theoretical physicists’ relative quality. This metric was then referred to as “h-index.” The h-index of a scientist is h if from the total of N articles he/she published, h articles have received at least h citations each; the remaining (N – h) articles have received at most h citations each. As an example, according to the Scopus, out of the 36 articles which one of the authors (FH) has published, 3 have received at least 3 citations and therefore, his h-index is 3. Although two of his remaining articles have also received 3 citations, they could not increase his hindex for their fourth and fifth positions when the articles are sorted on the number of citations they received. Obviously, in the calculation of the h-index we not only consider the number of published articles but also take into account the number of citations each of those published articles have received. However, like many other scientometrics, it carries an inculcated belief that “the more citation an article receives, the higher the quality it has” which is not a concept accepted by all scholars. Furthermore, h-index has other limitations. The very first step in the calculation of h-index is to recognize the articles published by a certain researcher. For several reasons, like common names and use of different spelling out of a certain name, this is not always a simple straight forward task. Calculation of h-index depends seriously on the database in use. As an example, the ISI Web of Science reports FH’s total published articles as 22 and his h-index as 2; and Harzing’s Publish or Perish, a software for the calculation of h-index which uses Google Scholar as its database reports his total published articles as 84 and his h-index as 5. Even the number of citations each of the articles received is reported differently—the number of citations to one of FH’s articles on “weighted impact factor” was reported as 3 by Scopus, one by ISI Web of Science, and zero by Harzing’s Publish or Perish. Another trouble with h-index is that if a researcher has contributed to science by publishing only one or two influential articles, his/her h-index does not really reflect his/her contributions proportionately. Some facts are so well-accepted that they are rarely cited; the Korotkoff sounds heard during the measurement of blood pressure is so familiar to health care personnel that no one generally cites the original article of Nikolai Korotkoff. Consequently, the h-index of Korotkoff whose contribution to medicine is undeniable, is not more than 4. For the very same reason, the h-index of Wilhelm Conrad Röntgen who won Nobel Prize in 1901 for his discovery and seminal work on X-ray is no more than 6. Another limitation of h-index is that it is subjected to manipulation. In the calculation of hindex, the number of citations is counted, no matter where is your place in the author list. In this way, being a member of a large prolific research group is a favorable factor in boosting up your h-index. Opinion