The h-index and all that

The UoH has benefited greatly in the past few years by the Department of Science and Technology’s PURSE (PROMOTION OF UNIVERSITY RESEARCH AND SCIENTIFIC EXCELLENCE) programme.

What was also unusual about how the grant was given to the 14 Universities first chosen was that it was based on the h-indexa recently introduced scientometric tool that has now gone viral. For those not familiar with it, the h-index of an academic is a number such that of her or his scholarly papers has been cited or quoted (formally, i.e. in a peer reviewed journal) at least h times. First introduced by Jorge Hirsch, the h-index (which has inspired any number of similar indices) captures both the longevity as well as the contribution of contemporary academic lives. On average, that is. A person who has written only 1 scientific paper that was cited once would have an h-index of 1. So would a person who wrote only 1 paper but had 1000 citations, say. Although the index cannot decrease with time, the largest indices are typically associated with the famous and influential- Nobel laureates typically have indices between 60 and 100 (with the all time high being around 120 or so).

A related measure in the scholarly publishing arena is the impact factor of a journal which measures the average numbers of citations to recent articles in it. Discussing that could take up another post in itself, but the connection with the h-index is clear. Publishing in high impact factor journals is generally good for your h-index. Of course, doing great work is even better.

The DST decided to use the productivity and citability of the scholarly output of a university by calculating the h-index of the institution, namely aggregating papers by university (rather than by author). The hope is that this would reflect the carrying capacity of the institution to support research in a range of disciplines, and would therefore reflect the extent of the scholarly base.

By that token we did very well.  Three Universities were classified as A class, Delhi, Panjab, and UoH, with indices above 50. Give the huge disparity is sizes its pretty clear that in general we punch significantly above our weight.

Anyhow, the PURSE grant (parenthetically one should note that the DST has a way with acronyms, and PURSE is just one in a list that includes FIST, BOYSCAST & INSPIRE…) was used to support research in the University, and somewhat narrowly, just science research. One can make the argument that it is the entire academic climate of an institution that matters. Perhaps when we get the second phase of the grant, PURSE-2, we can have a wider discussion on it and enlarge the beneficiaries as well.

Which brings me to the real point of this post, that many colleagues are doing particularly well on the matter of research, both in terms of its volume as well as its significance, and the PURSE award is just one recognition of that. So many have been elected to professional bodies, others are invited time and again to advise on issues related to research and academics. And several have got the India Citation Award organised by Thomson-Reuters in this as well as previous years. Some of the highest cited published work in the country- and thus some of the most influential ideas- have come out of UoH. By any yardstick, this is very impressive for so young an institution. Our collective thanks to all who contributed to this- You know who you are!

One way to continue that, to keep up the good work so to speak, would be to keep our standards high! Publishing in high impact journals, or in high quality journals that set a significant bar is one way.  Doing work that is important is another. But staying active, not settling for less, and having academic longevity…  That would probably be the best way of all…


11 thoughts on “The h-index and all that

  1. h-index is deceptive. Witten’s is 132 and Dirac’s is 16, while Isaac Newton’s is 2. So it should be taken with a pinch of salt. Andrew Wiles who solved Fermat’s last theorem has an h-index of 1.

    • Of course the h-index can be deceptive, and I already said that in the post. But the DST has found it useful to distinguish between universities based on that measure and has given us some money to accelerate our research, which was the point of the post in the first place. Although your numbers are incorrect- Wiles’ h-index is not 1, and neither is Dirac’s 16 or Newton’s 2, it is useful to make two other related points: 1) The h-index does not capture the importance of the contribution, just the number of citations. So Pythagoras, Archimedes or Euclid, who get used all the time without citation don’t have an h-index at all, and I’ve inserted the word contemporary in the post to alert a reader to that fact. And 2) one’s h-index cannot exceed the number of papers that one writes. So when it is used in the modern context and for people who work today, it is not totally irrelevant so long as you keep the caveats in mind…

  2. It would be better if we can put the list of papers constituting the h-index of our university on the university website and keep on updating it periodically.

  3. Bindu has a point. Great work is done by individuals (e.g., Newton and Einstein and Wiles) but routine work is done by a war team with a war budget. Perhaps the index is most suitable for physicists and biologists in their mid career. Others need not apply.

    Many of us has good ideas but we are poor in implementing. We also need good students. Great results comes like a “revelation” or, if you please call that “enlightenment”, and we need not spend either time or money for that: just some opportunity is enough. Good work demands the sweat of the brow (for both the student and the teacher).

    I have been blocked from taking a student for several years. What a pity!

    • I don’t think you are trying to understand what the h-index captures. Perhaps being self-referential has its drawbacks. Dirac wrote less than 50 papers. If you compute his h-index, regardless of anything it cannot exceed 50 and that does not say that his contributions are not great. My post was to appreciate those who contributed to the research profile of the university, and to make the additional point that those who continue to contribute, those who remain scientifically alive and active long into their careers are deserving of our admiration and gratitude.

      • I obviously did not get your point. If you do not publish much, your h-index cannot be high and that means the quality, as measured by citation, is not getting into the h-index. So you must be prolific writer to start with. To get high number of citations, you need to cite your papers and your friends (co-authors) papers in all your publication.

        If one works in two different areas, the papers in each are to be counted separately. The h-index will be automatically low. The broad shape of the distribution curve is similar for most people (as I checked with scopus for some of colleagues) and therefore my point is essentially valid.

        The time factor is also important. Papers take time to gather citations but very old papers, unless it is a classic one, are rarely cited. So the h-index will continute to improve for some time ever after you are academically dead. This distribution is also interesting (once reviewer recently has asked me to include a 50 year old paper in the references).

        I cannot embed graphs or figures in this blog (at least I do not know how), but I have done some studies on this and I have some results (unpublished).

        In spite of repeatedly asking for a student, you have failed to keep your promise. Without good students, we cannot be expected to do good work. I really do not believe in the h-index or the impact factor, but they are your choices.

  4. I welcome the idea to list all the (peer-reviewed) publications of current faculties / students of UH in the website and have the publications of retired or faculties who have left UH in a separate link.

  5. Good post. Both Alok and Nirmal made important point. As beneficiaries of PURSE grant from DST, let those details be known to everyone through the “official” website of the University. What are those publications that brought this distinction to this “otherwise young” University? If the University website is made more user-friendly and academically more elegant and informative (sadly very poorly organized at present) we can certainly attract better students (may not be by giving fellowships to all!!). Let all the University publications find place as pdf files on the website and show distinctly the ones that earned PURSE grant for the University. The enabling atmosphere in the institution referred in the post is arguably a debatable point. The University has to follow the prescribed rules of financial management by judiciously deploying the grants for the purposes for which they are granted. Do we have a rationale in changing the budget heads after the release of grants? For example, the PI of a project in our University and elsewhere in the country will not be allowed to buy chemicals (which are absolutely essential for that project work) from the grants left over under other budget heads of the same project. University burnt fingers (continue to be so) whenever we have deviated from such practices. We must learn from our past experiences.

    Your numbers on the h-indices of the Nobel laureates could be incorrect from what I have been reading of the h-index. Quite often the number has been greater than 100 while we may have a few exceptions!!.

  6. Pingback: Digital Colonizations | And then…

  7. Pingback: Digital Colonizations | Indian Academy of Sciences - Confluence

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s