SciELO - Scientific Electronic Library Online

vol.78 número4Articles with authors affiliated to Brazilian institutions published from 1994 to 2003 with 100 or more citations: I - the weight of international collaboration and the role of the networks índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados


Anais da Academia Brasileira de Ciências

versão impressa ISSN 0001-3765versão On-line ISSN 1678-2690

An. Acad. Bras. Ciênc. v.78 n.4 Rio de Janeiro dez. 2006 



Articles with authors affiliated to Brazilian institutions published from 1994 to 2003 with 100 or more citations: II - identification of thematic nuclei of excellence in Brazilian science



Rogerio Meneghini; Abel L. Packer

BIREME/PAHO/WHO - Latin American and Caribbean Center on Health Sciences Information / Pan American Health Organization / World Health Organization, Rua Botucatu 862, 04023-901 São Paulo, SP, Brasil
UNIFESP - Universidade Federal de São Paulo, Departamento de Informática em Saúde - DIS, Rua Botucatu 862, 04023-901 São Paulo, SP, Brasil




Articles with 100 citations or more in the scientific literature and with at least one author with Brazilian affiliation, were identified in the Thomson-ISI (Institute for Scientific Information) Web of Science bibliometric database covering a 10-year period, from 1994 to 2003 (see Packer and Meneghini 2006); 248 (0.23%) out of a total of 109,916 articles with Brazilian affiliation were identified. This study was primarily carried out to identify thematic nuclei of excellence in research that had major impact in the international literature. Twelve of these nuclei in the fields of Biomedicine, Medicine, Biology, Physic, Chemistry and Astronomy were considered outstanding and their genesis and development were described. The weight of factors such as international collaboration and network organization are distinct in these areas and the reasons for that are discussed.

Key words: bibliometrics, information services, information storage and retrieval, science, scientometrics, impact factor, scientific policy, scientific institutions, Brazil.


Foram identificados os artigos com um mínimo de 100 citações na literatura científica e com ao menos uma filiação brasileira, segundo Thomson-ISI (Institute for Scientific Information) Web of Science bibliometric database, no período de 10 anos entre 1994 e 2003 (veja Packer and Meneghini 2006). Este levantamento foi feito primordialmente para identificar núcleos temáticos de excelência em pesquisa que tiveram maior impacto na literatura internacional. Doze núcleos, nas áreas biomédica, médica, biológica, física, química e astronomia foram considerados proeminentes e suas gêneses e desenvolvimento foram descritos. Os pesos de fatores como colaboração internacional e organização de trabalho em rede são distintos em cada uma dessas áreas. São discutidas as razões para isso.

Palavras-chave: bibliometria, serviços de informação, armazenamento e recuperação da informação, ciência, cienciometria, fator de impacto, política científica, instituições científicas, Brasil.




Studies on the correlation between scientific quality and citations, at the level of institutions, areas of science, journals and scientists have been performed along the last four decades and many of them resulted in meaningful statistic values for such a correlation. (Garfield and Small 1978, Moed 2005, Aksnes and Sivertsen 2004, Oppenheim 1997, Lee et al. 2002, Saha et al. 2003. One of these studies (Aksnes and Sivertsen 2004) was focused on the highly cited articles of a nation and concluded that they may be useful in creating awareness towards the sub-areas of great influence on the national performance. In a recently published article, David King analyzed the outcomes of investment in research to measure quality at a national level and to compare it with the international scenario (King 2004). As a benchmark, King took the 1% top tier of articles with the highest citation index in the period 1993-2002 for 31 countries, encompassing 98% of the most-cited articles within this sample. Brazil was among these countries.

A similar methodology was applied in the present study (Packer and Meneghini 2006). A list of the articles with 100 or more citations was obtained from data recovered in the Thomson-ISI Web of Science bibliometric database, referring to the 1994-2003 timeframe. A total of 248 articles with at least one Brazilian affiliation were identified. In this paper it is described that 114 of these articles (46%) were found to be within the perimeter of 11 distinct sub-areas, thus defining clusters of high performance in Brazilian science: Amazon forest, cardiac surgery, oxidative metabolism and free radicals, chemical catalysis, genomics, neurosciences, physics of particles, quantum physics, human genetics, infectious diseases and contraceptive-vascular disease correlation.

As discussed at the end, the importance of international collaboration and network mode of operation were fundamental for the shaping of the clusters.



Although caution is necessary to draw conclusions from the ranking data, these are useful for some interesting considerations. Therefore, Table I shows the top most-cited articles: in a decreasing order of citations, the 37 articles from 1994 to 2003 that received more than 250 citations. It shows that as more stringent one becomes with the criteria of selection of articles the higher is the weight of international collaboration and network efforts of organization.



There is a considerable prevalence of medical and epidemiological investigations through multinational networks. They are studies on the incidence of diseases, efficacy and side effects of drugs and results of new treatment techniques or surgeries. There are 18 articles of this nature, accounting for virtually half of the articles included in the ranking in Table I. These 18 articles present an average of 9.4 countries per article, which is much higher than the general average of the 248 articles found in this survey, that is, 3.8 countries per article (Packer and Meneghini 2006). They show an average of 21 authors per article. Several institutions and authors received a high number of citations with that type of article. These are very important researches, but from the point of view of indicating individual citation of the investigators, or institutional citation, they deserve a differentiated assessment.

On the other hand, there are 17 articles on basic research reports, accounting for an average of just 2.1 countries per article and 6.9 authors per article, which is closer to the traditional pattern of scientific publications.

It also draws our attention that only 4 articles (articles 5, 19, 20, 24) were affiliated only to Brazil, that is 10.8%, as compared to 15.7% in the total context of 248 articles of this study (Packer and Meneghini 2006). Again, the stringency degree, from 100 to 250 citations per article, was achieved with an even stronger international collaboration. It is also noteworthy that 67.6% of articles in Table I have authors affiliated to the United States, whereas in the main scenario this figure was 58% (Packer and Meneghini 2006). There is a tendency of higher impact for articles co-authored by American authors. With these indicators, one could not conclude if the participation of American authors brings results with a trend to produce higher quality research or if they involve a psychological-social behavior, in other words, Americans citing more Americans. (Moed 2005, King 2004).



The use of the indicator citations per article enables mapping the impacting research activities in Brazil. It is important to stress, however, that this indicator is not excluding. As already mentioned, the scientific fields are distinguished in terms of tendency of citations due to several reasons. For example, in another study we detected articles with Brazilian affiliation which had fifty or more citations, during the same period of this research (unpublished results). This research identified approximately one thousand articles. In this case the digging went deeper and it was possible to identify fields that were not contemplated in this article, or were contemplated in an incipient manner, such as Mathematics (5 articles), Computer Sciences (8 articles), Anthropology (3 articles), Engineering (26 articles), Veterinary Medicine (6 articles), Agriculture (4 articles) and Biophysics (14 articles). These data stress the relevance of carefully handling scientometric indicators. However, it is important to mention that mapping centers of excellence does not significantly differ in the two surveys.

It is also necessary to consider that this survey encompasses a recent period of ten years, drawing an update picture of the most impacting researches. One could wonder about the absence of research groups that were outstanding in decades previous to that addressed in this survey. These groups continue to be acknowledged, but our goal is to obtain a picture the most updated possible, although, however, this demanded a relatively long timeframe (10 years) to obtain consistent data.

The choice of most-cited articles enables different mappings, which other indicators, such as total number of citations or publications of authors or institutions, would render it more difficult. The great number of citations achieved by an author does not indicate necessarily a high impact within a thematic proposal, since these citations can be diluted in a broad set of approaches. This is the case, for example, of researchers using more refined techniques to study a variety of topics. Or simply the case in which a high number of citations is the result from a high number of publications, each with a small number of citations, not individually reflecting a significant impact (Hirsch 2005).

The present study could detect important trends of the recent Brazilian scientific research in the international context.


The dynamics of changes in the Amazon rainforest due to the intervention by mankind, through deforestation for agriculture purposes, timber exploration and forest fires, has lately become a pungent topic. The discussion encompasses geopolitical issues, changes in the ecosystem, sustainable economic exploration, and gains extensive space in the international press. Not surprisingly, the most outstanding scientific journals also open space for articles in this field. Among the 248 HC-ISI articles, 12 were dedicated to this topic (Table II). Seven were in collaboration with the United States and four were researches conducted solely with Brazilian affiliation. Three articles were published by Nature and three by Science. These researches tend to corroborate the disastrous consequences of uncontrolled exploration of the rainforest. Nine articles result from studies carried out by institutions from the Northern region of Brazil, eight of which are from the Instituto Nacional de Pesquisa da Amazônia (INPA), located in Manaus. This fact is extremely good news, because it demonstrates the possibility of conducting high level researches, on a topic related to issues of an economically challenged region. INPA is an institution established in 1954 and connected to the Ministry of Science and Technology, where Brazilian and foreign researchers streamed to, attracted by the challenge of understanding the biology from this magnificent rainforest. The articles are very rich in terms of themes. Among these investigators is Philip Fearnside, an American that arrived in Brazil in the 1970's, and is considered one of the greatest international authorities in sustainable development in environments such as the Amazon rainforest. His studies put him in a comfortable position to defend what is called the international carbon market. The wealthy countries, at some point, will have to pay for Brazil to preserve the Forest, says Fearnside: "the capability of absorbing carbon will bring more money than selling soybeans".



Some articles have titles that could have been in newspaper headlines: "Physical and human dimensions of the Amazon deforestation", "Collapse of biomass in fragments of the Amazon rainforest", "Large scale impoverishment in the Amazon rainforest due to timber exploration and forest fires". Other titles show the concern with the dynamics of carbon recycling through photosynthesis, breathing and decomposition, or in deforestation and replanting conditions. Since it is an extremely up-to-date topic, the press that reports statements from politicians, economists and environmentalists, should listen more to scientists, whose studies are published in the most acknowledged journals, with far-reaching repercussion.


Another field that stood out for publishing highly cited ISI articles was surgery, and all articles were related to the cardiovascular subfield (Table III). The great majority were works in multinational networks concerning several types of investigation. The classification of these articles in the field of surgery was very flexible since 12 works were related to research on interventions with angioplasty techniques and use of stents. This is a very strong field in Brazil, particularly in the City of São Paulo, with the participation of Incor (Faculdade de Medicina da USP), Instituto Dante Pazzanese and Hospital Unicor. Another research that stood out was the team from the Hospital Angelina Caron, in Campina Grande do Sul, State of Paraná, coordinated by Randas Batista, who introduced the technique to reduce an enlarged heart by partial left ventriculectomy. This technique had far-reaching repercussion; although it is not a procedure established for end-stage cardiac patients, it led to new research opportunities to improve the intervention. Finally, one work that drew attention was published in 2003 and has already achieved 112 citations. It addresses transendocardial autologous bone marrow cell transplantation in patients with severe chronic ischemic heart failure (article 18, Table III). This study reported a new and relatively successful procedure and has prospects for future studies on transplant of stem cells in this type of condition. The groups from the Hospital Procardíaco and the Universidade Federal do Rio de Janeiro participated in this study.




This is a very active field in Brazil and a reflex of significant increasingly interest all over the world. More than 20 groups throughout Brazil study the diverse aspects of cellular oxyreduction processes and their effects in changing molecular structures and control cell regulation mechanisms. Table IV shows ten articles on this research line that achieved over 100 citations. Six of these had Brazilian affiliation, no international collaboration and stood out in this field among the most-cited articles. A work by a group of the Faculdade de Ciências Médicas da UNICAMP, coordinated by Aníbal Vercesi, made success with studies on changes by oxidation of mitochondrial membrane proteins (a cell organelle responsible for oxidative metabolism) that result in cell death (articles 3, 4, 7, 9 and 10, Table IV). The studies also addressed the role of calcium ions in this process. The group was responsible for five articles.



The group of Ohara Augusto, from the Department of Biochemistry of the Instituto de Química da USP, had a successful partnership with the group of Rafael Radi, from Uruguay. This collaboration was one of the few carried out exclusively between two Latin American countries, as already mentioned (Packer and Meneghini 2006), and it is represented by three works in Table IV (articles 1, 2 and 8). These articles resulted in a work in 1999 (article 8), which reported the formation of a carbonate radical, a chemical species not identified in live organisms at that time. The group of Augusto is still successfully investigating the issue.


Chemical catalysis is one the most advanced topics in modern chemistry. Its results reverberate in basic and technological chemistry and resound in Biochemistry. Seven works had far-reaching repercussion as demonstrated in Table V. Five of these articles were written by one group from the Universidade Federal do Rio Grande do Sul, known as the "Catalysis Group". Coordinated by Jairton Dupont and Roberto F. de Souza, the group has studied biphasic organometallic catalysis in ionic fluids, aiming to produce high synthesis yields at room temperature and in conditions to preserve the environment. Their results can be assessed by the high number of citations related to their articles.




Genome sequence is a research field of the end of the 20th century and beginning of the 21st century. Brazil has joined this exclusive group in high style by performing the genome sequence of the phytopatogenic bacterium Xylella fastidiosa. This study was published as the cover article in Nature, in 2000. The topic was extensively covered in the lay press both in Brazil and abroad.

This effort was put forth by the State of São Paulo, with financial and logistic support of the FAPESP, which established a real headquarter to make this project feasible in multiple aspects. Scientific and organizational competence was required for its success. A genome sequence network was assembled, named ONSA (Organization for Nucleotide Sequence Analysis) and it has been in charge of many genome sequence projects that followed the study on Xyllela. Although recent, the articles have already achieved a high number of citations, as displayed in Table VI.



The network studies had multiple authors but are not multinational works. The genome sequence of Xyllela was assisted by one researcher from France and another from Germany, but it was the result of the work carried out by Brazilian leaders and over one hundred Brazilian investigators. It could be said it was autochthonous.

Beside these articles, many other genome sequence projects were initiated in Brazil. A healthy competition was established in programs of the Ministry of Sciences and Technology and agencies from other states; since then many genomes have been sequenced. In terms of citations, these other articles will have to wait for a while, considering the results were published very recently.

It is too early to conclude if this was the best manner to achieve proficiency in molecular biology in Brazil. However, it should be acknowledged that there was a fundamental know-how gain to organize research network programs at national level, which is already influencing other fields.


Some strong leaders stand out in neurociences in Brazil. Sixteen articles are shown in Table VII, with over 100 citations.



Three groups are particularly worth mentioning in Behavioral Pharmacology. Frederico Graeff, from the Faculdade de Filosofia Ciências e Letras da USP de Ribeirão Preto, aimed to understand the effect of anxiogenic and anxiolytic drugs in animal behavior using rats as the experimental model (articles 1, 2). Studying anxiety and stress, the group of André Ramos, from the Departamento de Biologia Celular, Embriologia e Genética - UFSC, has investigated an interesting issue, that is, the genetic component of this behavior (articles 7, 8). Finally, a highly productive group, with several articles often cited is coordinated by Ivan Izquierdo, from the Departamento de Biochemistry da UFRGS (articles 3, 6). Izquierdo investigates memory mechanisms with animal experimental models in order to pharmacologically map the memorization process paths. Izquierdo has a long-term collaboration with J. Medina, from Argentina. The great majority of articles are signed by these two scientists and comprises a rare fruitful collaboration between two Latin American countries, like another already mentioned between Brazil and Uruguay.

Xavier Albuquerque is a pharmacologist who investigates the synaptic transmission in neurons. His career was carved out and became successful in a center he founded at the University of Maryland, in the United States. In the 1980's he established a molecular pharmacology center at the Instituto de Biofísica Carlos Chagas Filho - UFRJ. Later, he joined the Instituto de Ciências Biomédicas, Departamento de Farmacologia Básica e Clínica - UFRJ. From then on, he has simultaneously worked at the University of Maryland and at the UFRJ and indicates both affiliations in his articles. His works are strongly focused on the biophysical aspects of the synaptic process, investigating the behavior of ionic membrane channels in relation to drugs (articles 9-14).

One of the authors of article 15, in Table VII, is Brazilian, but the study was conducted in the laboratory of Miguel Nicolelis, of the University of Duke, United States. Nicolelis is Brazilian and internationally acknowledged for his works about sensorimotor connections. He obtained significant results by implanting chips in the brain of a monkey, who enabled him to control a robotic arm. Nicolelis is currently involved in establishing a center of Neurosciences in Natal, Rio Grande do Norte.

Both Albuquerque and Nicolelis were graduate students of César Timo-Iaria, a researcher at the USP and a pioneer in neurosciences in Brazil, who passed away in June 27, 2005.


This field requires large and very expensive pieces of equipment, particularly the big particle and ion accelerators, which are not available in Brazil and used for study experiments on collisions. Therefore, and also due to the need to extensively collect and prepare data, international networks with many authors started to demonstrate advances in this field. Table VIII displays 13 articles related to this field; seven (1, 4, 5, 6, 7, 8, 9) are from a network with Brazilian participants who are affiliated to the Instituto de Física - USP and four articles (10-13) are from another network, of which the Brazilian participants are affiliated to the Centro Brasileiro de Pesquisas Físicas do Rio de Janeiro. The first network achieved 945 citations. The mean number of countries per article of the seven articles is 9.4, and the mean number of authors per article is 154. One article had 403 authors (article 6). It should be mentioned that the seven articles were published in Physics Review Letters, which is considered the best journal of Physics. More recently, these studies have used a powerful ion accelerator - the Relativistic Heavy Ion Collider (RHIC), of the Brookhaven National Laboratory, United States. Hundreds of physicists from all over the world have used the RHIC to study how the universe could have been moments after its creation. The RHIC has two gold ion beams in opposite directions in a subatomic collision. The other network uses the Large Electron-Positron collider of the CERN, in France. It is an accelerator which makes collisions of electrons and positrons in a 27-km circumference ring. When the beam of electrons and positrons collide, some of them disappear and create, during a second fraction, high energy explosions that are reminiscent for the newly created universe, providing data that complement those from the RHIC. The four works were cited 575 times, have an average of 13.5 countries/article and 456 authors/article, and, in one case, it achieved 593 authors (article 12).



Once again, the comments about credits of researchers and institutions participating in the articles of those giant networks, by means of personal and institutional citation rates and impact factors of the journals, are valid.


Two groups excel in Quantum Physics in Brazil and both are located in Rio de Janeiro - one coordinated by Constantino Tsallis, from the Centro Brasileiro de Pesquisas Físicas and another by Luis Davidovich, from the Departamento de Física - UFRJ (Table IX). Tsallis works with theoretical physics of quantum mechanics and statistical mechanics. His works are very often cited (articles 1-5) and internationally acknowledged due to the approaches that have been named after him, Tsallis entropy and Tsallis statistics. Davidovich belongs to the more experimental segment (articles 6, 7) and was previously supervised by Moyses Nussenzveig, a Brazilian physicist, at the University of Rochester, United States. Today Nussenzveig works at the Centro Brasileiro de Pesquisas Físicas. Table IX has something curious - it shows the only article (number 4) among 248 included in this survey that was published in a national journal - the Brazilian Journal of Physics. This fact demonstrates that Brazilian researchers who are usual contributors to the main stream literature can sporadically submit articles to national journals, and help to enhance their international visibility.




Fourteen articles of human genetics achieved far-reaching repercussion from 1994 to 2003 and the majority was on mutation related to human hereditary diseases (Table X). In particular, the studies by Mayana Zates and Maria Rita Passos Bueno and their team, in international collaboration with researchers from France, United States, England, Italy and Israel (articles 3, 5, 7 and 8), demonstrated the genes involved in a type of human muscular dystrophy, which is hereditary, autossomal and recessive. These findings are very important for genetic diagnosis and counseling, since the mutated genes may be carried by potential healthy parents for it is a recessive disease (the father and the mother must present the mutation). Moreover, the outcome enables a better understanding of the genesis of the disease and may be the path to find drugs and the cure of the condition.



The Unidade de Endocrinologia Genética da Faculdade de Medicina da USP also contributed with two articles on a genetic disease - a type of pseudohermaphroditism (articles 1 and 4), which were published abroad, but with the affiliation of two Brazilian authors from the unit.


The landmark of the beginning of Brazilian sciences took place in Rio de Janeiro, in the beginning of the last century, when Oswaldo Cruz, a physician and researcher coordinated and encouraged microbiological studies of parasitary diseases. The legacy of Oswaldo Cruz still yields fruits; thus it is not surprising that Brazilian investigations about microbial infectious diseases continue to reverberate. In the context of contemporary sciences, this clearly implies in studies of molecular immunology, such as those in Table XI.



Three institutions are outstanding in this field - Departamento de Bioquímica e Imunologia - UFMG (articles 1-4), the Instituto Oswaldo Cruz do Rio de Janeiro (articles 9-12) and the Faculdade de Medicina da USP in Ribeirão Preto (articles 3, 5, 7). In particular, the article 2, by Ricardo Gazzinelli from the UFMG, reporting about the importance of a balance in the production of different cytokines (proteins that establish the communication among cells of the immune system) in resistance to acute infection of Toxoplasma in mice, achieved over 200 citations.


Multicenter case-control studies comprise a medical field that requires professional competence of the participants, capacity to manage the network and proficiency in statistical analysis. Several studies of this nature were detected in this survey; the discussion in Packer and Meneghini 2006 already mentioned these works. Table XII is another example of multinational networks - in this case, one group of the Escola Paulista de Medicina took part and the studies gathered approximately 40 researchers from 20 countries. All articles were published in the renowned journal The Lancet, since the effects of oral contraceptives in vascular diseases are still an issue of concern and it is important to be investigated in studies that are carefully planned and competently conducted. Nevertheless, one cannot deny that the real participation of each author in these investigations is diluted and not very perceptible.




The survey of the most-cited articles has enabled a more focused evaluation of fields and competence groups regarding the number of publications or citations of individual researchers. It is sure that other ingredients beside the content of articles may contribute to that (Van der Vorst 2004). Not considering anecdotic aspects, there is a consensus that the impact of an article, measured by the number of citations, indicates one of the following possibilities: (i) seminal contributions; (ii) relevant original investigations; (iii) reviews prepared by distinguished authors showing mastery, clarity and putting forward significant new perceptions to other authors of the field. To draw attention only to total number of citations of one researcher could be useful in many contexts, but it is ideal to detect contributions that are worth mentioning. Furthermore, total citation assessment may be biased when applied to one author, such as production of many articles but none with a significant number of citations (nothing against those who are very productive, since this could deserve merit) and mastering knowledge or expensive breakthrough techniques that result in a great demand for collaborations, which is another issue that deserves merit but does not provide to such author a major intellectual contribution in the categories mentioned above (Hirsch 2005).

Recent works investigating the fields of impact on science and technology have used a methodology to search most-cited articles. Recently, King (King 2004) used this approach to assess to what extent the different countries achieved a scientific output, in terms of science and technology expenses. To this end, normalization was made for citations in distinct disciplines, as well as for year of publication of the articles. This normalization is explained because the author intended to compare different countries during two periods. In the present study, such normalization was not complied with, and the citation parameters were based on the ISI-Thomson database, including articles with Brazilian affiliation during a ten-year period (1994-2003). Therefore, articles with 100 or more citations were listed. The intention was not to carry out a study on temporal transitions, but to take a picture of a recent period that cannot be a snapshot, considering that to understand its purpose, a longer period (10 years) was necessary. On the other hand, it was impossible to not establish a timeframe, and by doing so, some authors were not included, although, in previous periods, they had an impact equal to or greater than that of those herein detected. These cases could be quantified by means of other methodological approaches regarding citations. It could not be denied that the 10-year period was arbitrary. However, this decimal numeric bias is generally acceptable and protects the authors of this article against any accusation of favoritism.

In this timeframe, our primary objectives were to diagnose the fields that have major Brazilian scientific contributions at international level, to understand their genesis and consequences. We could wonder why other fields that are considered a priority in contemporary sciences and technology were not represented (information technology, nanotechnology, biosphere, materials, structure of biological and inorganic macromolecules, geological prospection). Nevertheless, it is important to mention that, to initiate this discussion, normalization by trends of citations in diverse disciplines would be needed. Thus, some fields such as Mathematics and Engineering have not been considered, as already mentioned. This normalization requires updated information from the ISI-Thomson database, which is commercially available.

A spin-off of this survey was to check the importance of multinational collaborations in works conducted or not in networks. (Aksnes and Sivertsen 2004, Meneghini 1996, Leta and Chaimovich 2002, Harvey and McMeekin 2005) In both cases, the participation of authors with foreign affiliation in ISI journals was much greater than the total number of articles with Brazilian affiliation. Furthermore, these collaborations were primarily with authors from developed countries. Hence, an important issue concerning scientometrics and bibliometrics arise: to what extent does this fact results from collaborations with scientifically strong groups from these countries or from an impact favored by greater influence of these international authors in the scientific community? There is no single answer and it definitely demands data that are not currently available.

Another scientometric issue resulting from this survey refers to the credit of citations of authors who belong to networks with several researchers as opposed to those operating more traditionally, that is, with a limited number of authors. Certainly, this is a concern of those studying information science who do not find support in more consensual rules. A recent work has advanced possibilities to deal with such situations (Hirsch 2005).

The limited set of articles and of groups in charge of authorship in the present study shows that a few can make a huge difference in an institutional or regional context. Taking into account the dual function of citations as a means of recognition by the peers and of symbols established for specificscientific achievements (Small 2004), one group of excellence detected by this approach have outstanding visibility, so that if moving from one institution to another, they bring their prestige to the new place of work. This migration is not frequent in Brazil, but it occurs quite often abroad, mainly in the United States and, to a certain extent, it is similar to transfer of famous soccer players between teams. The likelihood of few investigators unbalancing a collectivity may be observed in Figures 1 and 2 (Packer and Meneghini 2006), both at institutional and regional levels. The lines connecting the most-cited articles to the number of ISI publications are not parallel and divert upwards or downwards.

The inclusion of some groups or fields of research detected in this study is not surprising for those who closely follow up the development of Brazilian sciences. The scientific media has progressed much in the past ten years, and particularly the major national newspapers now dedicated more space to sciences. It means that the lay readers can have access to and get interested in scientific matters. We did not conduct an extensive survey of the achievements herein mentioned that resulted in reports in recent newspapers, but can state that at least 55% of them became news, which is a reasonable percentage considering what the scientific journalism values and the scientific community detects. The remaining 45% are more related to a limited sphere of scientists. These comparisons also deserve a more elaborated study, since scientometrics is interested in gathering impact detection tools not only in academic settings and in ISI database, but also in the general public. Above all, these individuals contribute most to conduction of scientific investigations and are very much interested in their results.



The authors thank Maria Regina Rangel Braga and Rosely Fátima Pellizzon and their team for assisting in the searches on ISI database.



AKSNES DW AND SIVERTSEN G. 2004. The effect of highly cited papers on national citation indicators. Scientometrics 59: 213-224. n/1/6/7/n167344l22728752.pdf         [ Links ]

GARFIELD EM AND SMALL H. 1978. Citation Data as science indicators. In: ELKANA Y, LEDERBER J, MERTON RK, THACKRAY A AND ZUCKERMAN H (Eds), Toward a metric of science: the advent of science indicators. New York: J Wiley & Sons, p. 179-207.         [ Links ]

HARVEY M AND MCMEEKIN A. 2005. Brazilian genomics and bioinformatics: instituting new innovation pathways in a global context. Economy and Society 34: 634-658.        [ Links ]

HIRSCH JE. 2005. An index to quantify an individual's scientific research output. Proc Natl Acad Sci USA 46: 16569-16572.        [ Links ]

KING DA. 2004. The scientific impact of nations. Nature 430: 311-316.        [ Links ]

LEE KP, SCHOTLAND M, BACCHETTI P AND BERO LA. 2002. Association of journal quality indicators with methodological quality of clinical research articles. JAMA 287: 2805-2808.        [ Links ]

LETA J AND CHAIMOVICH H. 2002. Recognition and international collaboration: the Brazilian case. Scientometrics 53: 325-335. contributions/5/h/4/b/5h4blf60q82kj83v.pdf         [ Links ]

MENEGHINI R. 1996. The key role of collaborative work in the growth of the Brazilian science in the last ten years. Scientometrics 35: 367-373. searcharticlesresults,2,2        [ Links ]

MOED HF. 2005. Citation analysis in research evaluation. Netherlands, Springer.         [ Links ]

OPPENHEIM C. 1997. The correlation between citation counts and the 1992 research assessment exercise ratings for British research in genetics, anatomy and archaelogy. J Doc 53: 477-487. EmeraldAbstractOnlyArticle/Pdf/2780530502.pdf        [ Links ]

PACKER AL AND MENEGHINI R. 2006. Articles with authors affiliated to Brazilian institutions published from 1994 to 2003 with 100 or more citations: I - The weight of international collaboration and the role of the networks. An Acad Cienc 78: 841-853.         [ Links ]

SAHA S, SAINT S AND CHRISTAKIS DA. 2003. Impact factor: a valid measure of journal quality? J MedLibr Assoc 91: 42-46.        [ Links ]

SMALL H. 2004. On the shoulders of Robert Merton: Towards a normative theory of citation. Scientometrics 60: 71-79. x/6/v/t/x6vtvm1209131570.pdf        [ Links ]

VAN DER VORST. 2004. How to write a frequently cited article.        [ Links ]



Manuscript received on May 2, 2006; accepted for publication on May 31, 2006; contributed by ROGERIO MENEGHINI*



*Member Academia Brasileira de Ciências
Correspondence to: Rogerio Meneghini

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons