Life Sciences Research in India:
A Profile based on BIOSIS 1998

Subbiah Arunachalam*, S I Rino

M S Swaminathan Research Foundation
Third Cross Street, Taramani Institutional Area, Chennai-600113
* Address for correspondence: arun@mssrf.res.in; Fax: 91 44 2541319

ABSTRACT

Maps life sciences research in India basing 8,352 Indian papers of 1998 as recorded in the BIOSIS Biological Abstracts 1998-2000 September (CD-ROM version). Of these papers, Indian Veterinary Journal and Indian Journal of Animal Sciences published more than 300 papers each, and Indian Journal of Experimental Biology and Current Science published more than 200 papers each. Eight journals published between 100 to 200 papers each, and 25 journals published between 50 to 100 papers each. The rest of the papers were published in 1,048 journals. Indian contributions on agriculture is found to be maximum with 1,675 papers, followed by biochemistry and molecular biophysics (1,057 ), human medicine - medical sciences (590 ), population studies (357 ), chemical coordination and homeostasis (331 ), environmental sciences (207 ), etc. Indian life scientists published their works in journals emanating from 47 countries. The largest number of papers i.e. 4,630 (55.4%) was published in 75 Indian journals, followed by 1,006 papers in 232 UK journals, 844 papers in 294 US journals, 657 papers in 127 Dutch journals, 311 papers in 86 German journals and the rest of the papers were published in 271 journals from 42 countries. Of the 75 Indian journals, 12 have carried more than 100 papers each, and 20 journals between 50 and 100 papers each. Of the 8,352 Indian papers only 40 Indian papers appeared in journals with an impact factor ³ 5.0, and 135 papers in journals with an impact factor ³ 3.0. The entire amount of literature has come out from around 1,400 institutions. The institutions that has produced more than one hundred papers are: All India Institute of Medical Sciences, New Delhi (192 papers); Indian Agricultural Research Institute, New Delhi (166); Banaras Hindu University, Varanasi (144); CCS Haryana Agricultural University, Hisar(140); Indian Veterinary Research Institute, Izatnagar (139); Punjab Agricultural University, Ludhiana ( 133); Indian Institute of Science, Bangalore (131); and Postgraduate Institute of Medical Education and Research, Chandigarh (107 papers). The papers have come from 428cities/ towns of India of which Delhi leads the list with 802 papers, followed by Bangalore (378), Mumbai (336), Lucknow (336), Chennai (336), Hyderabad (309), and Calcutta (301).

KEYWORDS: India; Life science research; Scientometric study.

INTRODUCTION

Using standard scientometric techniques we have mapped life sciences research in India as reflected by the journal literature of 1998. This is a macroscopic study at the institutional level. Apart from providing an estimate of the volume of work published by Indian researchers, we have identified (i) the contribution of different institutions, cities or towns, and states, (ii) the journals often used by Indian researchers and their standing as reflected by impact factors, and (iii) sub-fields in which Indian researchers have been active in 1998.

METHODOLOGY

The bibliographic data on all papers from India published in the year 1998 were downloaded from BIOSIS Biological Abstracts 1998-2000 September (CD-ROM version). We used the MS-DOS-based Silver Platter retrieving software to retrieve bibliographic data from the database. As BIOSIS gives only the first author's address, we could download only those papers, which had the address of the first author in India. We have missed all papers in which Indian authors were not listed as the first author. The fields downloaded were CS (Corporate source or author affiliation), SO (Source or Journal title), PY (Publication year), DT (Document type), AU (Authors), TI (Title), LN (Language.) and MC (sub-field or major classification).

To download data, we made a search strategy, which included names of more than 350 Indian cities/ towns. This was necessary because, unlike SCI, BIOSIS does not give country names in some of the entries. As the names of some Indian cities are shared by some foreign cities, we had to eliminate papers not belonging to India. We had done this work manually, after converting the data into a relational database using Visual-FoxPro. The names of institutions were standardized. Then names of states were added to all the cities found in the database. The country of publication of the journals and their impact factors were added from Journal Citation Reports 1997 (JCR 1997). We located the country of publication of journals that were not listed in JCR from Serial Sources for the BIOSIS Previews Database 1993 and Publist (on the Internet).

RESULTS

Biological Abstracts on CD (Silver Platter version) for the years 1998 _ 2000 September indexed 8,352 papers from Indian addresses for the publication year 1998. Of these 8,056 were journal articles and 188 literature reviews. Except two papers in French and one in Chinese-English, all the rest were in English.

Distribution by Journal

These papers were published in 1,085 journals. Indian Veterinary Journal and Indian Journal of Animal Sciences published more than 300 papers each, and Indian Journal of Experimental Biology and Current Science more than 200 papers each. Eight journals published between 100 to 200 papers each, and 25 journals published between 50 and 100 papers each. At the other extreme, 419 journals published just one paper each, 207 journals two papers each, and 93 journals three papers each.. Of the 1,085 journals used by Indian life scientists to publish their works in 1998, only 23 were letters journals that carried 226 Indian papers. There seems to be no urgency among Indian life scientists to use rapid communication channels. The situation is different in physics and to some extent in chemistry.

Distribution by sub-field

The journals were assigned to 162 sub-fields [Table 1]. Often a journal is classified under more than one sub-field. In such cases we have assigned the journal to the category listed first. Agronomy tops the list with 978 papers, followed by Biochemistry and molecular biophysics (646 papers), Horticulture (336 papers), and Animal husbandry (315 papers). Indian researchers published more than 200 papers but less than 300 in five sub-fields, and more than 100 papers but less than 200 in 15 sub-fields. Biological Abstracts has also indicated an intermediate level categorization for many journals. For example, the sub-fields Animal husbandry and Horticulture, among others, are grouped under Agriculture. At this level, there are 1,675 papers in Agriculture, 1057 papers in Biochemistry and molecular biophysics, 357 papers in Population studies, 331 papers in Chemical coordination and homeostasis, 590 papers in Human medicine - medical sciences, and 207 in Environmental sciences.

Table 1: India's Contribution to the World Literature of Life Sciences categorized by Sub-Fields
[only 23 sub-fields are shown here]

Distribution by Journal Country

Indian life scientists published their works in journals published in 47countries. We could not identify the journal country for 23 papers published in eight journals. The largest number of papers i.e. 4,630 (55.4%) was published in 75 Indian journals. Indian authors have used 232 UK journals to publish 1,006 papers, 294 US journals to publish 844 papers, 127 Dutch journals to publish 657 papers, and 86 German journals to publish 311 papers. Ten or more papers have been published in journals emanating from 24 countries. Of the 75 Indian journals, 12 have carried more than 100 papers each, and 20 journals between 50 and 100 papers. Of these 75, only 15 journals were included in Journal Citation Reports 1997, but none has an impact factor greater than 0.4. Biological Abstracts covers a much larger number of Indian journals than Science Citation Index, as it aims to cover the literature of life sciences comprehensively whereas SCI's coverage is selective and restricted to the `significant' journals.

Distribution by Journal Impact Factor

Of the 1,085 journals in which Indian scientists have published their work in 1998, 241 carrying 3,565 papers were not covered by JCR. A further 3,316 papers had appeared in journals whose impact factors are less than 1.0. As seen from Table 2, only 40 Indian papers had appeared in journals with an impact factor greater than or equal to 5.0, and 135 papers in journals with an impact factor greater than or equal to 3.0.

Table 2: Distribution of Indian Life Sciences Papers by Journal Impact Factor* (JCR1997)

* The impact factor of non-SCI journals have been considered as 0, which may not be true in all cases. - Ed

Distribution by Institution

India's output of life science literature in 1998 generated from more than 1,390 institutions.. We could not find institutional addresses for 479 papers. Institutions publishing 60 or more papers are listed in Table 3. Eight institutions have published more than 100 papers each. These include two university-level medical institutions, three agricultural universities, a veterinary university, a central university and the Indian Institute of Science. Fourteen institutions published between 50 and 100 papers. At the other extreme, 719 institutions published just one paper each, and 199 institutions two papers each. Academic institutions accounted for 4,879 papers; universities and colleges 2,345; and agricultural universities and colleges 1761 papers. Among the central government research organizations, ICAR published 759 and CSIR 638 papers. The distribution of papers over institutional type is shown in Figure 1.

Table 3: Indian Institutions Publishing 60 or more Papers

Figure 1: Contribution made by different institutions as seen from BIOSIS

Distribution by City and State

The papers being analysed originated from 428 Indian cities/towns. Delhi leads the list with 802 papers, followed by Bangalore (378), Mumbai (336), Lucknow (336), Chennai (336), Hyderabad (309), and Calcutta (301). Sixteen cities have published more than 100 papers, 14 cities between 50 and 100 papers, and 33 cities between 25 and 50 papers. At the other extreme, 173 cities have contributed just one paper each and 43 cities two papers each. No city could be assigned to 487 papers.

Indian Institute of Science and All India Institute of Medical Sciences have published a much larger number of papers in high impact journals than other institutions. Jawaharlal Nehru University, Postgraduate Institute of Medical Education and Research, Christian Medical College, University of Delhi, National Chemical Laboratory, and Bose Institute have also published a few papers in high impact journals.

A matrix of sub-fields vs. impact factor ranges of journals reveals that Biochemistry and molecular biophysics, Cell biology, and Human medicine- Medical sciences are the fields that contribute most of the papers in high impact factor journals. This is to be expected as these are areas of higher impact than most classical biology and agricultural research fields.

We have also identified journals used by leading institutions and sub-fields in which these institutions are active. The detailed tables are available in our report submitted to NISSAT.

CONCLUSION

BIOSIS is a good source of information for such a study. It covers all areas of life sciences, including classical and new biology, agriculture and some areas of medicine. From a bibliometrician's point of view also, it is a good database as it gives journal titles in full and journal country information is provided in the print version of the Serial Sources. One minor problem is that for publication year 1998 it had indexed only 75 journals. For the three years 1992-1994, it had covered 120 Indian journals. This is a problem with almost all databases. None will cover all Indian journals in its area of coverage. Especially, the citation index databases of ISI cover much smaller percentage of Indian journals. The reason for the poor coverage of Indian journals in international databases has to do, at least in part, with the poor quality of the papers published in Indian journals.

About 55% of Indian papers have been published in Indian journals. There has not been much change from 1992-1994, when Indian journals accounted for 54% of all Indian papers indexed in BIOSIS. Agriculture is the only field where a much larger proportion of Indian papers are published in home country journals (about 77% in 1990-1994 and greater than 78% in 1998). In mathematics, as seen from 1988-1998, 38.5% of Indian papers were published in home country journals.

A high proportion of Indian papers have been published in low impact factor journals: more than 82% in 1998 and 84% in 1992-1994. Less than 1% papers have appeared in journals with impact factors of 4.0 or higher: 0.87% in 1998 and 0.69% in 1992-1994. Thus, over the years there has been a shift, even if it is only marginal, towards publishing in higher impact journals. A point of concern is that even better known institutions publish a substantial proportion of their papers in low impact factor journals or even journals not indexed in SCI . For example, Indian Institute of Science, where the entire thrust in new biology (as distinct from classical biology) has published 48 of its 133 papers in journals of impact factor less than 1.0. In general, new biology journals have a higher impact factor than classical biology journals. All India Institute of Medical Sciences has published 123 of its 192 papers in low impact (IF < 1.0) journals.

Another point that stands out is that the share of academic institutions has come down from 64.5% of all Indian papers in 1992-1994 to 58.4% in 1998. Is it because, recruitment to higher level teaching and research positions have been frozen in recent years? Or, are our universities unable to allocate or attract funds for research in life sciences?

Literature-based mapping can be used in framing policies for higher education and science. It can even be used, with abundant caution of course, in evaluation and monitoring institutional performance.

ACKNOWLEDGEMENT

We thank NISSAT-DSIR for financial assistance and NCSI-IISc for allowing us to download data from their collection of BIOSIS.

-----
The paper is a summary of the Report prepared under the National Mapping of Science Project, sponsored by NISSAT and does not include all the tables and figures. The literature covered in the study is selective inasmuch as BIOSIS Biological Abstracts 1998-2000 covered only 75 Indian journals when we have more than 250 research journals on life sciences The study definitely provides an indication as to how much of our life science literature is being covered by the international abstracting service like BIOSIS, the extent of our literature being published abroad, etc. - Ed.

-------
Information Today & Tomorrow, Vol. 20, No. 4, December 2001, p.18-p.22
http://itt.nissat.tripod.com/itt0104/biorep.htm