The impact of research carried out with the participation of professors from public universities based in the State of São Paulo has increased in the last ten years, as the proportion of studies carried out with international collaboration also increases. In Brazil as a whole, however, both the impact – measured as the average number of citations per article – and the international collaboration rate of universities appear to be virtually stagnant in the comparison between 2004 and 2014.

These data, compiled in the Web of Science database, from Thomson Reuters, by professor Renato Pedrosa, from the Department of Scientific and Technological Policy (DPCT) of the Institute of Geosciences (IG) at Unicamp, appear in a table published in the September edition of the magazine Fapesp Research.

 “This is a subject that has already been explored, studied and known,” said the researcher. “It is known that increasing international collaboration always increases visibility. You can say that visibility is measured by impact, by citations. It is not necessarily an indicator of quality”, he warns. “But it indicates that this science is being seen more, more used in the world, which in the end ends up being an indicator of quality”.

One of the data highlighted by Pedrosa is the relative impact – the proportion between the average number of citations per publication of an institution or country in relation to the world average. All São Paulo states made progress in the comparison between 2004 and 2014: USP saw its relative global impact go from 78% of the world average to 84%; Unesp, from 59% to 63%; and Unicamp, from 73% to 95%. Of the universities based in the State, the one that achieved the greatest relative global impact was the Federal University of ABC (UFABC), with 226% of the world average.

“One aspect that probably pushes this Federal do ABC number up is that it is more focused on science and engineering – Physics, Chemistry, Biology tend to have more citations than other areas”, explains Pedrosa. “For example, if you look at the Brazilian Center for Physical Research (CBPF), it has a similar global relative impact, of more than 200%. But if you just look at the impact of Physics articles, the CBPF comes closer to the world average”.

“With the Institute of Pure and Applied Mathematics (IMPA), in Rio, which is excellent in mathematics, it is the opposite: mathematics has a tradition of few citations. So IMPA, when compared in relation to the general average, appears at the bottom. But if you only use mathematics, the picture changes. Care must be taken in these comparisons.”

He remembers that the number of publications computed in the data does not include every type of document – ​​books, for example, are not represented. “These data, therefore, do not apply so well to all areas. They apply best to the areas of basic sciences, reasonably well to engineering, to Medicine, but to the human sciences they do not apply well,” he said. “So, when we analyze by area, we have to be careful not to make too strong statements about the area of ​​human sciences.”

PROMOTION POLICIES

In addition to showing that the growth in relative impact was accompanied by an increase in international collaborations – in the case of Unicamp, the proportion of publications that had at least one author from the university and one international rose from 25%, in 2004, to 32% , in 2014 – the numbers show that articles with collaboration from outside the country obtain even greater visibility than average. Throughout Brazil, the relative impact index of publications, in 2014, was 68%. Isolating only international collaborations, the rate jumps to 126%. At Unicamp, specifically, where the relative impact of publications in general is 95%, that of international partnerships reaches 167%.

“The lessons that emerge from this graph are: internationalization and international collaboration help with visibility; there is a growing trend, in Brazil in general and in universities here in São Paulo, to increase international collaboration”, said Pedrosa.

“This, probably, as is happening in all universities, has to do with national and state development policies”, he believes. “In São Paulo, Fapesp always encourages a lot, both through agreements and calls for international projects, but also in its own way of encouraging researchers to have international experience, valuing this when evaluating projects.” 

“Science Without Borders could have some impact on this, but I think the results are not yet visible in these numbers, because the program started sending people in 2011, and the majority are undergraduates,” he said. “Now, the fact of having the program, of exposing Brazilians, exposing the professor here, who ends up communicating with someone abroad – it is possible that it will have an impact, but no one has measured it yet. The program could be improved, it has several problems, but its future seems uncertain, there are no plans for new scholarships for 2016”.

Regarding the stagnation of Brazil's overall numbers – which went from 31% of articles in international collaboration in 2004 to 33% in 2014, and a relative impact of 70% to 68% – Pedrosa says that there is no clear known cause. He recalls that, a few years ago, Web of Science absorbed a large number of Brazilian journals which, because they were published in Portuguese, did not have great international visibility, which may have depressed the numbers.

“It may have a little to do with the recent expansion of federal universities, many are still starting out. Another thing is the Brazilian magazines that entered the base,” she said. “But it is something that should already be changing, because of this phenomenon that is happening, of greater international participation. There is something in the rest of Brazil, which perhaps is not as strong in São Paulo, and which is pulling it down”.

DIALOGUE

In addition to the data in the table produced for Fapesp, which deals with institutions based in São Paulo and figures for Brazil as a whole, Pedrosa has also been collecting information on the effects of the internationalization of scientific production in different countries.  

“Several countries are rising a lot on this scale of percentage of international collaboration, and Brazil is falling a little behind, because the global average is growing,” he said. “Brazil’s average is growing, but the global average is growing along with it.”

“Increasing international collaborations is even more important than increasing the impact number, because it shows that you are managing to talk to the world, in terms of science”, said Pedrosa. “The citations are a result of that. Of course, there are areas where this may not make much sense, but in all areas of natural sciences, and in Economics, in Education, it is important to dialogue with the world and not just look at one's own navel.”

The researcher recalls that Brazil became a place where the international scientific community started to look more closely. “People are interested in doing research in partnership with Brazil. I believe we will see growth in areas of Social Sciences, Human Sciences, energy, social inclusion, public policies”.

CRISIS

 “Now we are in crisis, Brazil has become a country in economic crisis, but there was a very strong interest in Brazil, this thing spread throughout the world,” he said. “Science Without Borders made Brazil much more visible to institutions for which Brazil was nothing more than a name.” 

The current economic scenario should make it difficult to deepen internationalization, predicts Pedrosa. “The crisis will increase the difficulty for us, both of sending people and bringing people, of financing projects, after a few years in which it was reasonably easy to finance comings and goings,” she said. “There is great concern for research and postgraduate studies, with the recent cuts at Capes,” she said. “This will be reflected in the coming years in Brazilian scientific production, which should stop growing as was happening until recently.”  

The researcher remembers that the effects of the cuts made now will only appear in statistics in a few years, when the projects currently underway are completed. “It's not something we're going to see right away, but from 2017, 2018, we could see a stagnation, for example, in the number of new doctors,” he said. 

“The effect on scholarships, on programs, will appear next year. But the effect on publications, on the number of doctors trained, will appear in three, four, five years. And it's bad because, when you see the effect, it's time to fix it, perhaps because there were no resources to fix it. We will then have to make an effort to recover lost ground.”