Jointly published by Akadémiai Kiadó, Budapest and Springer, Dordrecht
Scientometrics, Vol. 65, No. 2 (2005) 161–171
Cuban scientific articles in ISI Citation Indexes and CubaCiencias databases (1988–2003) JUAN A. ARAUJO RUIZ,a GUIDO VAN HOOYDONKb, RAUL G. TORRICELLA MORALES,c RICARDO ARENCIBIA JORGEa a
National Center for Scientific Research (CNIC). Department of Scientific Information. Havana City (Cuba) b University of Gent, Department of Library Sciences (Belgium) c Ministry of Higher Education (MES), Havana City (Cuba) This comparative study covers the period 1988–2003 of the Institute for Scientific Information Databases (ISI-DBs), CD-ROM edition: Science Citation Index (SCI), Social Sciences Citation Index (SSCI) and Arts& Humanities Citation Index (A&HCI) as international databases and from the CubaCiencias (CubaCiencias) as an internal database. The number of articles published in Cuban journals, ISI-DBs, the author associativeness trend, the most important institutions and other indicators are collected. However, it is observed that CubaCiencias and ISI-DBs are not perfectly suitable for a study of the productivity of Cuban authors. It is necessary to properly standardize the author fields. For bibliometric studies, Cuba needs a database not only for the published papers in Cuban journals, but also for all the papers published by Cuban authors.
Introduction The use of bibliometric indicators to measure the scientific output of a country or an organization should also consider other economic, social and demographic indicators. According to R OUSSEAU (1997), the national gross product, the expenses for education, the number of professionals in Research and Development, the scientific publications per number of inhabitants, etc., are suitable indicators. This paper does not intend to cover such indicators, so its results are only relative and they should be taken only as indicative. Cuba is a developing country, but the Cuban government has always encouraged the improvement of Science and Technology Studies (STS), in order to allocate resources for developing the country. The Ministry of Higher Education of the Republic of Cuba (MES) and its Universities or Research Centers are very interested in the measurement of research productivity and the visibility characteristics of Cuban scientists. The measurement of productivity in scientific and technological research is important;
Received April 28, 2005 Address for correspondence: J UAN A. ARAUJO R UIZ P. O. Box 4145, CP 10 400, Ciudad de La Habana, CUBA E-mail:
[email protected] 0138–9130/US $ 20.00 Copyright © 2005 Akadémiai Kiadó, Budapest All rights reserved
J. A. ARAUJO R UIZ et al.: Cuban scientific articles
not only to define the best funding policy, but also to identify which institutions and scientists obtained the best results in their STS in order to promote them (ARAUJO RUIZ et al., 2002) In general, the executive personnel of the MES advises publishing in foreign scientific journals rather than in the national journals, because, in theory, foreign journals have a higher impact. (DIRECCIÓN INFORMATIZACIÓN, MES (2004)) However, we do not use the citation data to assess the scientific work because it is very difficult for the Cuban scientists to obtain such data. On the other hand, the citation rate of Cuban scientists is very low and it is questionable if it accurately represents the current quality of the research activities in Cuba (SHRUM, 1997). LICEA DE ARENAS (1994, 1995) has studied the productivity of Cuban scientists by using the bibliometric measurement in international databases, but that kind of study is not frequent in Cuba. However, the Science Citation Indexes produced by the Institute for Scientific Information (ISI) are widely used for the measurement of the productivity of countries and scientists. SHRUM (1997) and LANCASTER et al. (1986), state that “the view from afar” based exclusively on information drawn from international databases does not accurately represent the population of researchers or domestic productivity in less developed countries. (LANCASTER et al., 1986) The Scientific and Technical National Library (STNL), a division of the Institute of Scientific and Technological Information of the Republic of Cuba (ISTI) produces a bibliographic database called CubaCiencias (BNCT, 1998). It contains the bibliographic description of all the articles published in Cuban scientific journals. The aim is to verify which of the productivity indicators derived from the international bibliometric sources correspond to those obtained from a local Cuban DB. Methodology a) Visibility Two kinds of information are needed. The first is the information on the international visibility of scientists, as indicated by their appearance in international databases. The second is the information on the domestic scientific production encountered in the national database. This study covered the 1988-2003 period from Science Citation Index (SCI), Social Sciences Citation Index (SSCI) and Arts & Humanities Citation Index (A&HCI) as international databases and CubaCiencias as an internal database. The databases from ISI: SCI, SSCI, and A&HCI from 1988 to 2003, and CubaCiencias supported on CD-ROM were used. The search was carried out by using the word ‘CUBA’ within the ‘Address’ field. All the records were manually reviewed to
162
Scientometrics 65 (2005)
J. A. ARAUJO RUIZ et al.: Cuban scientific articles
remove the records that could have the word chosen and would not belong to Cuban authors. A Pentium IV microcomputer with Windows XP and the Procite software package, ver. 5.0, the UNESCO developed CDS/ISIS for Windows and Microsoft Access database management system were used. b) Quality In addition, ISI-DBs were also used to retrieve citations to access the “quality” of the production. The total number of citations was found according to the following formula: Tc(x) =
∑ Ca(i, x) nx
i =1
where Tc (x): Total citations in the year (x) Ca (i, x): Number of citations to the article (i) in the year (x) x: years (1988–2003) nx: total articles in the year (x)
c) Associativeness The associativeness (As) is the index that measures the average of authors per paper within the set of papers researched (SPINAK, 1996). This set of papers can represent an indicator of the research, bibliographic yield from a scientific institution, a country, etc. As(x) =
∑ Aa(i, x)/n nx
i =1
where As (x)= Associativeness in the year (x) Aa (i,x)= Authors of the paper (i) in the year (x) nx = Total papers in the year (x)
Results and discussion Figure 1 shows the number of articles per year in the ISI data bases.
Scientometrics 65 (2005)
163
J. A. ARAUJO R UIZ et al.: Cuban scientific articles
Figure 1. Number of Cuban articles per year of publication in the Web of Science 1988–2004
The articles published in Cuban journals decreased from 3620 in 1989 to 1453 in 1994. This effect is produced by the lack of possibilities for publication in the Cuban journals. In 1995, a moderate resurgence of the most part of Cuban journals began: the production of articles increased until 1999. In 2000, a decrease was caused by the delay in updating the CubaCiencias database, not by the lack of publications. On the other hand, the publication of articles by Cuban authors significantly increases in journals covered by the ISI-DBs. In 1988, 236 articles and in 2003, 734 articles were published more than three times a higher number. The possible cause of this effect may be the necessity for Cuban authors to publish anyway, and the delays of publication for most of the Cuban journals. Table 1 shows the relative percentage of articles appearing in CubaCiencias DB compared to the articles appearing in ISI-DBs per year of publication. The percentage of articles appearing in ISI-DBs, increases from 3.3% to 10.3%, more than three times a higher number. Table 2 shows the authors with 45 or more articles in the ISI-DB in the period of 1988–2003. The result is an average of more than 3 articles/author per year. Table 2 shows a group of authors’ names that was ambiguous, then, it was not possible to precisely identify who they were: GONZALEZ-A; GONZALEZ-R; RODRIGUEZ-M; PEREZ-M. and PEREZ-R It is seen that both CubaCiencias-DB and ISI-DBs sources are not perfectly suitable to study the productivity of Cuban authors. It is necessary to properly standardize the author field. For bibliometric studies, Cuba needs a database not only for papers published in Cuban journals. A perfect database must include all the papers published by Cuban authors.
164
Scientometrics 65 (2005)
J. A. ARAUJO RUIZ et al.: Cuban scientific articles
Table 1. Comparison of the number of articles per year of publication in the ISI-DBs and the CubaCiencias database Year 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Total
ISI-DBs
%
Cumulative %
236 255 252 237 263 303 306 362 438 443 552 687 670 746 657 734 7141
3.3 3.6 3.5 3.3 3.7 4.2 4.3 5.1 6.1 6.2 7.7 9.6 9.4 10.4 9.2 10.3 100.0
3.3 6.9 10.4 13.7 17.4 21.6 25.9 31.0 37.1 43.3 51.1 60.7 70.1 80.5 89.7 100.0 100.0
CubaCienciasDB 3371 3620 2398 2100 2054 1547 1453 1736 2014 2109 2428 2524 1961 1885 1771 1575 34546
% 9.8 10.5 6.9 6.1 5.9 4.5 4.2 5.0 5.8 6.1 7.0 7.3 5.7 5.5 5.1 4.6 100.0
Cumulative % 9.8 20.3 27.2 33.3 39.2 43.7 47.9 53.0 58.8 64.9 71.9 79.2 84.9 90.4 95.5 100.0 100.0
Table 2. The most productive authors in ISI-DB (more than 45 author-articles each) No. 1* 2 3 4* 5 6 7* 8 9* 10* 11 12 13 14 15* 16 17* 18 19* 20* 21*
Author Mas, R. Elías, A. González, A. Pino J.A. Pérez, R. González, R. Trallero-Giner, C. Ruiz, T. E. Rodríguez, M. Fernández, L. Crespo, G. Hernández, L. Suárez, M. Reguera, E. Ruiz, R. F. C. Valdés Sosa, P. Pómez, R. Sánchez, E. Torres, V. Pérez, M. Comas, F.
Institution CNIC, MES ICA, MES Unknown IIIA, MINAL Unknown Unknown UH, MES ICA, MES Unknown CNIC, MES ICA, MES CIGB III, MINAL UH, MES INIFAT CNIC, MES CNIC, MES CNIC, MES ICA, MES Unknown UH, MES
Number of articles ISI-DB 122 106 93 93 88 80 80 62 58 56 54 53 52 52 51 51 50 48 47 46 46
Number of articles CubaCiencias-DB 36 107 8 12 10 11 4 8 4 10 55 14 10 3 0 4 8 6 30 12 5
Web of Science: Articles = 7141; author-articles = 32364 CubaCiencias: Articles = 34546; author-articles = 89059
Some authors (marked with an asterisk in Table 2) published more articles in mainstream journals than in the low-impact Cuban journals. This also increases the international visibility of Cuban Science, but decreases the impact of Cuban journals. It is necessary that the best Cuban authors publish in national journals, too. At present, the
Scientometrics 65 (2005)
165
J. A. ARAUJO R UIZ et al.: Cuban scientific articles
policy for publishing does not stimulate enough the publication in Cuban journals. A good idea may be to give a relative importance to every article published by this group of authors wherever they may be published. So, the value of the publications of an author belonging to this group will be highly considered. Table 3 shows the journals where Cuban authors have published. It is important to make evident that one journal in Portuguese and seven journals in Spanish are among the most productive journals for the Cuban authors, with 1547 articles published in the period studied. Table 3. The most productive journals (more than 30 articles) No.
Journals
Number of articles
1
Cuban Journal of Agricultural Science
826
2
Revista de Neurología
390
3
Revista Mexicana de Física
76
4
Interferon y Biotecnología
65
5
Memorias do Instituto Oswaldo Cruz
61
6
Physica Status Solidi B-Basic Research
60
7
Biochemical and Biophysical Research Communications
55
8
Journal of Essential Oil Research
54
9
Physical Review B
54
10
Revista de Metalurgia
45
11
Mycotaxon
45
12
Journal of Applied Physics
44
13
Nefrología
43
14
Afinidad
41
15
Journal of Radioanalytical and Nuclear Chemistry
39
16
Nahrung-Food
36
17
Hybridoma
35
18
Biotechnology Letters
34
19
Acta Crystallographica Section C-Crystal Structure Communications
33
20
Kidney International
33
21
Vaccine
30
22
Current Therapeutic Research-Clinical and Experimental
30
Total articles = 7141 Total of journals = 1647
166
Scientometrics 65 (2005)
J. A. ARAUJO RUIZ et al.: Cuban scientific articles
Table 4. Number of Cuban articles’ references and citations per year Number of references
Average reference/article
Number of citations
Average citation/article
Year
Total articles
1988
236
1964
8.32
705
1989
255
2363
9.27
676
2.65
1990
252
2924
11.60
799
3.17
2.99
1991
237
2511
10.59
455
1.92
1992
263
2941
11.18
834
3.17
1993
303
3636
12.00
929
3.07
1994
306
4093
13.38
968
3.16
1995
362
5491
15.17
1281
3.54
1996
438
7992
18.25
960
2.19
1997
443
9018
20.36
1948
4.40
1998
552
11892
21.54
2596
4.70
1999
687
14426
21.00
2422
3.53
2000
670
15137
22.59
1890
2.82
2001
746
17468
23.42
1784
2.39
2002
657
14606
22.23
722
1.10
2003
734
18551
25.27
182
0.25
Total
7141
135013
18.91
19151
2.68
Table 4 shows the number of references and citations per year, as well as the average number of references and citations per article. As shown, the trend is towards an increase of references and citations per article. References increases from 8 to more than 25 references per article in 2003, and the average of citations per article is 2.68 in the period studied. In Table 5, the Cuban institutions that have contributed the most in relation to the number of articles to the databases in the period studied are shown. The institutions belonging to the Ministry of Higher Education (MES) produce the 50.6 % of the articles published in mainstream journals by the 24 more productive institutions. These 24 institutions published the 90.1 % of all the Cuban articles in ISI-DBs. Figure 2, shows that 89% of the articles published by Cuban researchers in the last 16 years was published in journals in English, what is to be expected, because around 70% of the journals indexed by the ISI are published in that language, as distinguished from the fact that only 30% of Science is originated in countries where English is spoken (GARFIELD, 1995).
Scientometrics 65 (2005)
167
J. A. ARAUJO R UIZ et al.: Cuban scientific articles
Table 5. Number of articles by institution in the Web of Science No.
Institution name
Articles
% of articles
Total % cumulative
1
University of Havana (UH, MES)
1818
25.46
25.46
2
Nacional Center for Scientific Research (CNIC, MES)
835
11.69
37.15
3
Center of Genetic Engineering and Biotechnology (CIGB)
681
9.54
46.69
4
Institute for Animal Science (ICA)
550
7.70
54.39
5
Cuban Academy of Sciences (ACC)
331
4.64
59.03
6
Institute of Tropical Medicine "Pedro Kouri"" (IPK)
309
4.33
63.35
7
International Center for Neurological Restoration (CIREN)
212
2.97
66.32
8
Center of Pharmaceutical Chemistry (CQF)
202
2.83
69.15
9
Central University “Marta Abreu” (UCLV, MES)
172
2.41
71.56
10
University of Oriente (UO, MES)
143
2.00
73.56
11
Institute of Math. Science Applied to Physics (ICIMAF)
119
1.67
75.23
12
Institute for Research of the Food Industry (IIIA, MINAL)
117
1.64
76.87
13
Hospital “Hermanos Ameijeiras” (MINSAP)
111
1.55
78.42
14
Center of Molecular Immunology (CIM)
105
1.47
79.89
15
Institute of Neurology and Neurosurgery (MINSAP)
95
1.33
81.22
16
University “Jose A. Echevarria” (ISPJAE, MES)
92
1.29
82.51
17
Cuban Research Institute on Sugar Cane (ICIDCA)
91
1.27
83.79
18
Center of Applied Studies to Nuclear Development (CEADEN)
78
1.09
84.88
19
National Museum of Natural History (MNHN)
73
1.02
85.90
20
Higher Institute of Nuclear Science and Technology (ISCTN)
72
1.01
86.91
21
National Institute of TropicaI Agriculture Research (INIFAT)
68
0.95
87.86
22
Ministry of Science, Technology and Enviroment (CITMA)
58
0.81
88.67
23
Institute of Hematology and Immunology (IHI, MINSAP)
51
0.71
89.39
24
Institute of Nephrology (IN, MINSAP)
50
0.70
90.09
6433
90.09
90.09
708
9.91
100.0
7141
100.0
First 24 Institutions Other Institutions Real Total of Articles
As can be seen in Table 6, the trend of the Associativeness of Cuban authors is different when they publish in Cuban journals or in major foreign journals. In the first case, the Associativeness decreases, and in the second, it increases. So, for Cuban authors, the teamwork increases when they publish in the mainstream journals, but in Cuban journals they generally avoid teamwork. This different trend may be explained by the fact that the author who publishes in Cuban journals is generally not the same who publishes in higher impact journals, and the cooperation between multiple institutions is more frequently in articles published in mainstream journals.
168
Scientometrics 65 (2005)
J. A. ARAUJO RUIZ et al.: Cuban scientific articles
Figure 2. Language distribution
Table 6. Cuban authors’ Associativeness in the studied period Year
ISIDBs
No. authors
1988
236
837
3.55
3371
8995
2.67
1989
255
965
3.78
3620
9693
2.68
1990
252
937
3.72
2398
6862
2.86
1991
237
903
3.81
2100
5639
2.69
1992
263
1100
4.18
2054
5872
2.86
1993
303
1291
4.26
1547
4309
2.79
1994
306
1298
4.24
1453
3997
2.75
1995
362
1778
4.91
1736
4429
2.55
1996
438
1861
4.25
2014
5027
2.50
1997
443
2016
4.55
2109
5298
2.51
1998
552
2257
4.09
2428
5969
2.46
1999
687
3251
4.73
2524
5983
2.37
2000
670
3063
4.57
1961
4581
2.34
2001
746
3783
5.07
1885
4619
2.45
2002
657
3092
4.71
1771
4285
2.42
2003
734
3932
5.36
1575
3501
2.22
Total
7141
32364
4.53
34546
89059
2.58
Scientometrics 65 (2005)
Associativeness
CubaCienciasDBs
Authorarticles
Associativeness
169
J. A. ARAUJO R UIZ et al.: Cuban scientific articles
Conclusions The number of articles published in the Cuban scientific journals decreased from 3620 in 1989 to 1453 in 1994. However, the number of published articles by Cuban authors and indexed in ISI’s databases has increased, from 236 articles in 1988 to a total of 734 in 2003. We can see that the CubaCiencias-DB and ISI-DBs are not perfectly suitable to study the productivity of Cuban authors. It is necessary to properly standardize the author’s and author’s address fields. For bibliometric studies, Cuba needs a database, not only for papers published in Cuban journals, but also for the papers published by Cuban authors in foreign journals. The most productive authors (Table 2) publish more articles in international journals than in Cuban journals. It is very good for increasing the visibility of Cuban Science, but not so good for increasing the impact of the Cuban journals. It is necessary to encourage the most productive Cuban authors to publish in national journals, too. The number of Cuban articles’ references per year, as well as the average of citations per article, increases from 8 to more than 25 citations per article as an average in 2003. The institutions belonging to the Ministry of Higher Education (MES) produce the 50.6% of the articles published in mainstream journals by the 24 more productive institutions. These 24 institutions published the 90.1% of all the Cuban articles in ISIDBs. The 89% of the articles published by Cuban researchers in the last 16 years was published in journals in the English language, what was to be expected, because around 70% of the journals indexed by the ISI are published in English. The trend of the Associativeness of Cuban authors is different when they publish in Cuban journals or in higher impact foreign journals. In the first case the Associativeness decreases, and in the second case, it increases. So, for Cuban authors, teamwork increases when they publish in mainstream journals, but in Cuban journals, they avoid teamworking.
* The authors wish to thank The University Library of Ghent for the help in consulting all the Data Bases; and to Professor Jesús A. Núñez Romay and translator Luz M. Rodríguez Cabral for the revision of this article.
170
Scientometrics 65 (2005)
J. A. ARAUJO RUIZ et al.: Cuban scientific articles
References ARAUJO RUIZ, J.A., ARENCIBIA J ORGE, R., GUTIERREZ CALZADO, C. (2002), Ensayos clínicos cubanos publicados en revistas de impacto internacional: estudio bibliométrico del período 1991–2001, Revista Española de documentación científica, 25 (3) : 254–266. BIBLIOTECA NACIONAL DE CIENCIA Y TÉCNICA (1988), Cubaciencias, la base de datos de la producción científica cubana. Ciudad de La Habana: IDICT. GARFIELD, E. (1995), SCI Journal Citation Reports; A bibliometric analysis of science journals in the ISI database, Philadelphia: Institute for Scientific Information, Inc. MINISTERIO DE EDUCACION SUPERIOR, DIRECCIÓN INFORMATIZACIÓN, MES (2004), Revistas Digitales para la Popularización de la Ciencia y la Tecnología para América Latina y el Caribe. [on line] available in: http://169.158.24.166/bives/ [Cited: 15 January 2005] LANCASTER, F. W., PORTA, M. A., PLAGENZ, K., SZYMBORSKI, K., KREBS, M. (1986), Factors affecting sources cited by scientific/ a case study of Cuba, Scientometrics, 37(5-6): 243-257. LICEA DE ARENAS, J., VALLES, D. W. (1994), Investigación cubana en agricultura: un enfoque bibliométrico, Ciencias de la información, 25 (3) : 136–141. LICEA DE ARENAS, J. (1994), Medios y mensajes de la ciencia. La revista científica. Ciencias de la información, 25 (1) : 2–12. LICEA DE ARENAS, J., VALLES, D. W., CAMPOS, J. (1995), Las bibliotecas universitarias de América Latina: un estudio bibliométrico, Ciencias de la información, 26 (1) : 13–16. ROUSSEAU, S., ROUSSEAU, R. (1997), Data envelopment analysis as a tool for constructing scientometric indicators, Scientometrics, 40 (1) : 45–56. SHRUM, W. (1997), View from afar: “visible” productivity of scientists on the developing world, Scientometrics, 40 (2) : 215–235. SPINAK, E. (1996), Diccionario enciclopédico de bibliometria, cienciometria e informetria. Caracas: UNESCO.
Scientometrics 65 (2005)
171
