National Mapping of Science - India: Life Sciences*

Subbaiah Arunachalam
Project Investigator
MS Swaminathan Research Foundation
3rd Cross Road, Institutional Area, Taramani, Chennai-600113

Abstract

Presents an analysis of the contributions of Indian researchers to life sciences research as indexed in BIOSIS Biological Abstracts 1992-1994. The study aims to map life sciences research in India as reflected by the journal literature, using standard techniques of scientometrics. It is a macroscopic study at the institutional level and does not analyse the data at the level of individual researchers. The report covers the volume of work published in India; journals often used, their standing (as reflected by their impact factors) and country of publication; sub-fields in which Indian researchers are active; and Indian institutions active in publishing their work.

The methodology adopted for the study was that all papers having a first author address in India were downloaded from BIOSIS Biological Abstracts 1992-1994 (CD-ROM, Silver Platter) as addresses of only first authors are given in the database. The years indicate the disc years and not the years of publication of the individual papers. Names of institutions of the authors were standardised. The impact factor values were taken from Journal Citation Reports (JCR) 1992 and 1994 and sub-field classifications for journals were taken mostly from SCI Guide and for some from Ulrich's International Periodicals Directory. The data downloaded and grouped were converted into a database and analysed using FoxPro.

Analysis

In the three years 1992-94 (as covered in BIOSIS), Indian researchers published 20,046 papers in 1,582 journals. They published more than 100 papers (in the three years) in 37 journals and more than 50 papers in 84 journals. At the other extreme, they published only one paper in each of 482 journals and two papers in each of 248 journals. The distribution of papers among journals nearly follows Bradford's law.

Many journals indexed in BIOSIS are not indexed in SCI, and therefore they are not included in the Journal Citation Reports. Their impact factors are shown as 0.0 in the report. Also, for some journals indexed in JCR, the impact factor column is left blank. For these journals also, the report gives the impact factor as 0.0. Some journals are classified into two or more sub-field categories, Bio-resource Technology, a UK journal, is classified into energy, biotechnology and agriculture.

As regards Indian contributions to letters journals, 416 papers (letters) (2 per cent of the total) were published in 26 letters journals and even among these papers, more than 90 papers are not in life sciences proper. The 64 papers published in Tetrahedron Letters are in the area of Organic Chemistry, 18 published in Analytical Letters are in Analytical Chemistry, 8 published in Acta Crystallographica Section C are in Crystallography. In Physics, Indian researchers publish a much higher percentage of papers in letters journals (as per an earlier study of the author).

The number of sub-fields into which the journals were grouped is 110. Of these, Agriculture (2,824 papers), Plant Science (2,300 papers), Biology (1,341 papers) and Veterinary Science (1,024 papers) are the sub-fields in which Indian researchers are most active. A very large part of life science research in India is centred on Agriculture and Classical Biology. Among the sub-fields of New Biology, Biochemistry and Molecular Biology (674 papers) is an area of more than moderate activity.

Classification by Journal Country

In terms of country of publication of the journals in which Indian researchers published papers, more than 55 per cent of the papers from India appeared in 118 Indian journals, more than 12.8 per cent (2,570 papers) in British and 9.8 per cent (1,964 papers) in US journals. These are followed by journals published in the Netherlands, Germany, Switzerland and Japan. In all, Indian researchers used journals from 52 countries to publish their work. Of the 118 Indian journals in which Indian researchers published papers, at least seven in the top ten journals are in the field of agriculture. Of the 50 journals most often used by Indian researchers to publish their work, only five are foreign journals. These are Phytochemistry (U.K.), Biochemistry International (Australia) which has been renamed Biochemistry and Molecular Biology International, Cytologia (Japan), Fitoterapia (Italy), and World Journal of Microbiology and Biotechnology (U.K.).

Similar lists for Physics and Chemistry would have many more foreign journals. There are two reasons for this: one, a large part of biological science research done in India is in the areas of Agriculture (including Veterinary Science and Dairy Science) and Classical Biology. The Classical Biology work done in India, with a few exceptions, is not of great current interest to researchers in the scientifically advanced countries and it may be difficult to get Indian papers in this area published in good foreign journals. Much of what is done in Agriculture is of considerable local interest and publishing such research in Indian journals may serve the purpose better than publishing it in foreign journals.

Two, India's share of the journal literature of New Biology is rather small, and these papers are scattered in a number of journals. The four most often used foreign journals in New Biology are: Biochemistry International (Australia) (renamed Biochemistry and Molecular Biology International), 143 papers; World Journal of Microbiology and Biotechnology (UK), 81 papers; Mutation Research (Netherlands), 62 papers; and Biochemica Biophysica Acta (Netherlands), 56 papers.

Classification by Journal Impact Factor

The study classifies the journals publishing Indian papers in life sciences, into different ranges of impact factors as seen from Journal Citation Reports 1992 and 1994. A very large number of papers appeared in very low impact journals. About 84 per cent of papers appeared in journals whose impact factor is either zero (meaning these journals are not included in SCI) or less than 1.0.

About two-thirds of the journals used are in this lowest impact category. At the other extreme, 54 papers were published in 10 journals whose impact factor (IF 92) is higher than 6.0, and 103 papers in 24 journals with an impact factor of over 5.0. There is a slight increase in the impact factor of many journals between 1992 and 1994. For example, there are 15 journals with a 1994 IF higher than 6.0 (as against 10 in 1992) and these journals published 72 Indian papers in the three years under study (as against 54 in the 10 journals whose IF was above 6.0).

Classification by Institution

Among the Indian institutions whose work is indexed in BIOSIS 1992-1994, there are three agricultural universities and two medical education and research institutes in the top seven institutions. Banaras Hindu University (BHU), University of Delhi, University of Calcutta and Indian Institute of Science (IISc) are among the most prolific publishers of life sciences research in India. Fourteen institutions published more than 200 papers, and 35 published between 100 and 200 papers in the three years period covered. Leaving out home addresses, the study finds that life sciences research is published from over 1400 institutions in India. The distribution of papers over institutions follows a typical Bradford curve.

The institutions are classified into different categories. Academic institutions comprising general, agricultural and medical universities and colleges and engineering colleges account for 64.5 per cent of all papers from India. Scientific agencies of the central government such as the Indian Council of Agricultural Research (ICAR) (1,708 papers), Council of Scientific & Industrial Research (CSIR) (1,545 papers), and Indian Council of Medical Research (ICMR) (416 papers) published 4,179 papers (about 21%). Organisations under the central ministries account for 1,092 papers. The reason for the greater volume of work at the higher educational institutions, compared to be better-endowed national laboratories, as per the author, is the presence of a very large number of doctoral students.

Classification by City and State

The study shows that Uttar Pradesh, Maharashtra and Delhi are the leading states. Delhi leads (1,935 papers) the rest of the cities by a large margin. This is largely because of the concentration of research institutions of different kinds in the Capital. Four other cities, viz., Calcutta, Lucknow, Mumbai and Bangalore published more than 800 papers each. Ludhiana and Hissar (largely by the Panjab and CSS Haryana Agricultural Universities) and Hyderabad published more than 700 papers each. Papers came from more than 450 Indian cities and towns, of which 47 published at least 100 papers each in the three years and 82 more than 50 papers.

Use of High Impact Journals

One has to be very cautious in interpreting the data based on the impact factors of journals. Publication of a paper in a high IF journal does not necessarily mean the same IF for the paper also. Ideally, one should count the number of times a paper is cited and see in which journals these citations occur, rather than merely look at the impact factor of the journal in which a paper is published. It is seen that in certain cases, Indian papers often bring down the impact factor of journals.

As the impact factors of agricultural and classical biology journals are low, one does not find many highly cited papers coming out of agricultural universities. Also, results of agricultural research are best published in local/national journals. Institutions carrying out work in New Biology and Medicine have a better chance of placing some of their papers in high impact journals. IISc, for example, published 61 out of 284 papers in journals with IF higher than 2.5. The Centre for Cellular and Molecular Biology (CCMB) published 44 papers, about half of its entire output, in journals with IF higher than 2.5. The National Institute of Immunology (NII) has 29 papers, Indian Institute of Chemical Biology (IICB) 22 papers, and University of Delhi 37 papers in journals with IFs higher than 2.5. The Post Graduate Institute of Medical Education and Research (PGI), Chandigarh, had 18 papers, All India Institute of Medical Sciences (AIIMS) 29 papers, and National Chemical Laboratory (NCL) 18 papers in this list. Scientists from some of these institutions also publish their work in non-SCI journals, mostly Indian journals (such as Indian Journal of Experimental Biology) that could, with a little improvement, get into SCI.

In all, India as a whole published 671 papers (about 3.35 per cent of all papers) in journals with an IF (94) higher than 2.5 and 270 papers in journals with an IF higher than 4.0. Among the papers indexed in BIOSIS 1992-94, only two papers in Nature and six in Lancet are from India. Indian researchers however, manage to publish a large number of papers in good impact biochemistry journals. In the three years considered, they published 30 papers in Journal of Biological Chemistry, 17 in Biochemistry, 56 in Biochimica et Biophysica Acta, 13 in Journal of Molecular Biology, 48 in Biochemical and Biophysical Research Communications, 21 in Biochemical Journal, and 19 in European Journal of Biochemistry. Apart from Biochemistry and Molecular Biology (238 of the 674 papers), Indian researchers published in medium-high impact journals (IF>2.5 in JCR 1994) in Neuroscience (46 papers), Biophysics (41 papers), Microbiology (41 papers), Immunology (39 papers) and Genetics (33 papers). A small number of institutions account for a large percentage of papers in high impact journals. Scientists at the IISc published eight of the 30 Indian papers in Journal of Biological Chemistry, seven of the 13 Indian papers in Journal of Molecular Biology, one paper in Journal of Immunology, and one paper in Nature. Scientists of IICB published seven papers in Journal of Biological Chemistry. AIIMS had one paper in Journal of Molecular Biology, two papers in American Review of Respiratory Diseases, two in Hepatology, two in Lancet, and one in Arthritis. The International Centre for Genetic Engineering and Biotechnology (ICGEBT) had four papers in Journal of Immunology, and one in Journal of Biological Chemistry. All the four Indian papers in Journal of Biomolecular NMR came from Tata Institute of Fundamental Research (TIFR). The only Indian paper in FASEB Journal came from Christian Medical College (CMC) in Vellore, and the only Indian paper in Journal of Cell Biology came from the University of Hyderabad. Indian Statistical Institute (ISI) provided the only Indian paper in American Journal of Human Genetics.

Considering the large number of people engaged in new biology research in India, one cannot be happy with this low level of use of high impact journals. Throughout the three year period studied, as seen from BIOSIS, no Indian researcher published any paper in Science, the weekly journal of the American Association for the Advancement of Science. One can also note that more than 46.4 per cent of India's work indexed in BIOSIS 1992-94 appeared in journals not indexed in SCI and another 37.4 per cent of papers appeared in journals with IF below 1.0. A silver lining to the cloud is provided by a few less-known institutions publishing papers in some better-known journals. There was a paper in Lancet from Dharbhanga Medical Hospital, and another in Nature from Utkal University.

A few well-meaning scientists are worried that such studies as this might encourage Indian scientists to publish their work in foreign journals to the great detriment of Indian initiatives to publish quality journals. However, without a large number of Indian scientists and institutions performing first-rate work, Indian journals cannot sustain their quality for long.

Clustering of institutions

The study also covers the contributions of institutions to different sub-fields and it is seen that a cluster of institutions working on classical biology and agriculture concentrate on certain sub-fields and publish often in journals of low impact and another small cluster of institutions working on new biology publish part of their papers in high impact journals. For example, IISc, University of Delhi, Aligarh Muslim University, NCL, IICB, CCMB, University of Madras and NII are reasonably strong in publishing in Biochemistry and Molecular Biology, whereas agricultural universities concentrate on Agriculture, Plant Science, Zoology, Entomology, etc. The author feels that it would be in the nation's interest if the same set of practical problems are tackled both by classical biologists in the larger cluster and the new biologists in the smaller cluster. There seems to be very little overlap between the interests of these two clusters and hardly any joint efforts involving institutions across the cluster boundary.

Conclusion

BIOSIS is a good source of information on India's contribution to life science research in terms of extensive coverage and details of the publications and the authors, etc. A main limitation however is, it indexes less than 120 Indian journals although several hundreds of journals are published in India in the fields of Biology, Life Sciences, Medicine and Agriculture. Poor quality of content and delays in publication could be the reasons for their small coverage, apart from the fact that even an international indexing periodical cannot cover all the publications of the world in the relevant subject fields.

Life science research in India is much larger than physics research and there are more biologists than physicists in the country. A very large proportion of the biologists are working on classical biology.

This study shows that more than 54 per cent of Indian papers indexed in BIOSIS are published in Indian journals. One reason for this is that a substantial percentage of Indian papers in life sciences is in the area of Agriculture (including Veterinary Science, Horticulture and Forestry) and most of it is published in Indian journals. According to an earlier study of the author this output (in agriculture) is about 77 per cent.

A very large proportion of Indian life science papers are published either in low-impact journals (IF<1.0) or in journals not indexed in SCI. Less than one per cent of Indian papers appeared in journals with an impact factor higher than 4.0 (as per Journal Citation Reports 1992 & 1994). It is also seen that even scientists from better-known institutions publish a large number of papers in journals not indexed in SCI.

Academica accounts for a very large percentage of India's publications, although one is painfully aware of the sorry state of neglect of most of India's higher educational institutions. This output is largely due to the large number of doctoral students, who are usually keen to finish their work and publish it in good journals.

Literature-based maping can provide valuable insights for policy makers. It can reveal a nation's (or any large aggregate's) strengths in terms of journals used for publishing, active institutions and sub-fields.

* The paper is based on the report prepared under National Mapping of Science project, sponsored by NISSAT. The databases used for the study does not cover Indian literature comprehensively. Hence, the scenario of Indian output drawn on the basis of data may not reflect the real Indian situation. - Editor

-------
Information Today & Tomorrow, Vol. 19, No. 1, March 2000, p.22-p.25
http://vkv.tripod.com/itt20001/nmslife.htm