The publications of Brazilian authors: access, distribution and publishers

Science requires its own communication system; advances come from consulting results as previously published in journals. To make a useful contribution to the scientific community, a paper needs to be available to other scientists in a continual process of access and citation (MERTON, 1973; ZIMAN, 1979, 1981).

Publication in a highly prestigious journal is the most accepted way of demonstrating the originality of a study and confirming that its results were robust enough to overcome the scepticism of the scientific community (WHITLEY, 2007; XIA, 2014). Without access to peer-reviewed results, researchers risk duplicating studies, using outdated and/or inappropriate methods, and compromising the core concepts of their study (ABADAL, 2012; SMITH et al., 2017).

Since the progress of science depends on prior research, access to reliable literature is crucial. The journal review process filters the damage that false premises or results could have on the work of other researchers. The concern with providing quality publications is central to the scientific communication system and the selection and review process is the responsibility of journal editors and peer reviewers (LÓPEZ-COZAR; SALINAS; LÓPEZ, 2007).

The system connects research results with shared focus and issues, acting as an arena for conflicts over reputation and interpretations (WHITLEY, 1984). Articles in journals become the main form of establishing social control and competency standards within the scientific community, functioning as a locus in negotiations regarding intellectual goals and priorities.

Scientific journals serve multiple functions (GUÉDON, 2001):

a) as tools for allocating prestige;

b) as keys to ownership of citable elements (results, figures, paragraphs);

c) as instruments for evaluating the individual performance of scientists.

With such functions, well-recognized journals at the core of the community represent the highest standards (BEASLEY, 2016; BJÖRK et al., 2009; BOYLE, 2007; GUÉDON, 2010).

Technological advances have led to questions regarding the value that commercial editors add to the scientific publication system. Three large-scale corporations (Elsevier, Springer-Kluwer, and Wiley-Blackwell) publish 42% of all articles and control the majority of the most highly prestigious articles in circulation. Another 2000 publishers are responsible for the remaining journals, of which none exceed more than 3% of the total (CARVALHO NETO; WILLINKSKY; ALPERIN, 2016; MCCABE, 2001; MCGUIGAN; RUSSEL, 2008; MILLER; HARRIS, 2004; XIA, 2011) despite the rising criticism regarding access costs and commercial oligopoly (HARO, 2017; LARIVIÈRE; HAUSTEIN; MONGEON, 2015a; NOORDEN, 2013).

The barriers to consolidating new journals, the prevalence of highly prestigious journals, and the power that publisher have over indexing and ranking help to maintain the system’s status quo. While well-established journals increase core relevance, newer one’s struggle to break into and achieve recognition, a phenomenon known as the Matthew effect (GUÉDON, 2010). This is particularly harmful to developing countries since it weakens national publications, which remain invisible in the field of international literature, and becomes lost to science (GIBBS, 1995) while their best studies are published abroad, thereby preserving the centrality of the system (GUÉDON, 2010).

In Latin America, a non-commercial open-access (OA) model, the platinum route (MIGUEL; CHINCHILLA-RODRIGUEZ; MOYA-ANEGÓN, 2011; PACKER, 2011), has been successful due to the government sponsored higher-education system. The most internationally prestigious Brazilian journals are financed by public funds (ALPERIN, 2015; CARVALHO NETO; WILLINSKY; ALPERIN, 2015; MUELLER, 2011; PACKER, 2011; RODRIGUES; ABADAL, 2014). This is different to the commercial model prevalent in central countries, where publishers privatize research that has been developed with public funds, control access fees and define the number of journals in each research area according to commercial concerns (LARIVIÈRE; HAUSTEIN; MONGEON, 2015a; McGUIGAN; RUSSEL, 2008), leading to what Larivière, Haustein, and Mongeon (2015b) call “oligopoly”.

Rodrigues and Abadal (2014) indicate that Brazil leads the number of academic journals in Latin America, pointing out that more than 90% of the Brazilian journals indexed in Web of Science (WoS) and SCOPUS are Open Access without Article Processing Charges to authors (APC). This proportion is far higher than in other regions, according to Miguel; Chinchilla-Rodriguez, and Moya-Anegón (2011). For peripheral countries, participation in the international publishing system is complicated due to database restrictions on new journals (ALPERIN, 2015 GUÉDON, 2001; LETA, 2012; LETA; COSTA; MENA-CHALCO, 2017; PACKER; MENEGUINI, 2014).

Packer (2011) described the growth of Brazilian journals in WoS (from 19 in 2005, to 71 in 2009, and then up to 2012, whereby this number had grown to 164), highlighting the remarkable decentralization of journals per publisher (approximately 1.2) and the relevance of the mega-publisher SciELO and universities and associations as publishing institutions. The use of the platinum OA model (open to readers without charges to authors) in Latin America, especially in Brazil, has been well documented (MUELLER, 2011; OLIVEIRA; RODRIGUES; MATIAS, 2017; PACKER, 2011; SANTILLAN-ALDANA; MUELLER, 2016; RODRIGUES; ABADAL, 2014; RODRIGUES; PASSOS; NEUBERT, 2018; RODRIGUES; QUARTIERO; NEUBERT, 2015). The high concentration of platinum journals is a result of collaborative efforts among the scientific community, universities, scientific associations and governmental agencies.

The main questions of this study are what are the characteristics of the journals and those publishers where Brazilian authors publish their articles? To answer this question, this study describes the publishers and the principal journals where scientific papers are published by Brazilian researchers. The objectives are:

a) to identify the type of publisher and access to those journals which have published articles by at least one author with a Brazilian institutional affiliation;

b) to examine the research area, origin, and impact factor of these journals;

c) to determine types of publisher distribution.

Using Clarivate Analytics’ WoS database (CLARIVATE ANALITYCS, 2018), we performed the following search “Country/Region CU=Brasil OR Brazil)” to identify publications with at least one Brazilian-based author and then to filter for: (1) 2016 as the publication date; and (2) articles only, which resulted in 52.343 items. Based on this list, we identified the number of articles per journal, and only journals with five or more articles were included, which produced 2.313 journals of which 43.663 articles remained after duplication was removed. The exclusive use of the Clarivate list is a limitation but represents the elite of journals in all knowledge areas.

A search for each journal’s profile in Journal Citation Reports® (JCR) found data on 2.001 journals. We assumed that the remaining 312 journals did not yet have a JCR profile as they are new on the base. Data on journals without a JCR profile was obtained from Ulrich's web and each journal’s website between October 2017 and January 2018. The presence of OA symbols and/or any explicit declaration determined the access type; so-called hybrid journals were classified as subscription. The value of the APC follows Directory of Open Access Journals (DOAJ) information. It's impossible to identify precisely if each author pays for the APCs and the cost of the APCs itself, so these results are approximated.

To classify the types of publishers we considered any association or society as an association while considering teaching institutions as university. Any institution focused mainly on research is a research institute. Journals that fit none of these categories but had a clear commercial focus are commercial. Journals were grouped by publishers’ owners and companies within the same group classified as a single entity. Journals which did not fit into any of these categories, were classified as other.

We used the WoS research area classifications, categorizing the journals into five broad areas: arts & humanities, life sciences & biomedicine, physical sciences, social sciences, and technology. To classify the 312 journals without a profile in the WoS, we used data from the Ulrichs web Global Serials Directory and the website of each individual journal.

The distribution of the 43.663 articles published in WoS-indexed journals in 2016 with at least one Brazilian-based author show 2.313 journals, with an average of 18.88 and a median of ten (10) articles per journal. Table 1 shows 77.5% of the journals published between five (5) and 20 articles, summing 16.220 articles (37.2% of the total). The majority of the articles (62.8%) published in only 22% of the journals. Approximately half of the journals (53.8%) published between five (5) and ten (10) articles (19.2% of the total). The journals with the highest concentration of articles, despite representing only 2% of the total number, published a similar number of articles (19.8%), of which 512 titles (22%) were OA and 1.801 (78%) subscription. The article's distribution shows that 18.520 (42%) were published in OA and 25.133 (58%) published in subscription journals.

Table 1
Distribution of articles’ access by Brazilian-based authors among WoS-indexed journals in 2016

Note: OA = open access. Subs = subscription

Source: The authors (2019).

Table 2 presents those 20 (0.8%) journals with the most articles by Brazilian-based authors. Together, they published 5.385 articles (12.5% of the total) and more than half (13) of these, published in Brazil's titles, mostly in the field of agricultural and life sciences, which are areas of national and scientific expertise (UNESCO, 2015). The remaining journals are from the U.S.A. (3), the U.K. (2), the headquarters of large editorial groups (LARIVIÈRE; HAUSTEIN; MONGEON, 2015b), and New Zealand (2). Among the top journals, most are published by universities (6), and 16 are OA.

Caballero Rivero, Macedo dos Santos e Trzesniak (2017) identified the characteristics of Brazilian publication areas by pointing to Agrarian and Health Sciences, which passed from a predominantly national orientation in the production of articles, to a balance in national and international production.

Five (25%) of the 20 journals are OA platinum, all Brazilian. The processing charges range from USD 109,00 (State University of Feira de Santana and Federal University of Uberlândia) per article to USD 1,760,00 (Scientific Reports from Nature).

Of the USD 2,916,839,00 spent on publication fees in these 20 journals, USD 1,277,595,00 went to PLOS One alone to publish 801 articles, less than 2% of the 43.633 articles published by Brazilians in WoS-indexed journals in 2016.

The fees of Brazilian journals are far lower than those of other publishers. Appel and Albagli (2019) indicate about the value of fees: “[…] internationalization through translations and publications in different languages have increased the APC prices charged to authors.” (APPEL; ALBAGLI, 2019, p. 11), and so therefore, it is possible to identify that APCs cover these specialized services.

Publishing costs are coming under increasing scrutiny due to the changes demanded by authors, governments, universities, and financing agencies (BEASLEY, 2016; MCCABE, 2001; NOORDEN, 2013; PINFIELD; SALTER; BATH, 2016).

The Council for Scientific and Economic Development (CNPq); Ministry of Science and Technology spent approximately USD 688.000 (R$ 2.528.500,00) in 2016 to support the best 154 Brazilian OA journals through federal grants (CNPQ, 2016). More resources came from a combination of local government funds, research institutions, associations, and universities, most of these heavily financed by federal and state governments

The Ministry of Education spent US$ 85 million in 2016 (R$ 357.463.927,00 at the time) on “big deals” and subscription packages from various international publishers according to Budget Report (CAPES, 2018). CAPES´ Portal de Periódicos pays subscriptions to publishers, which allows Brazilian universities and research institutions to access the contents. The libraries of each institution may also have additional subscriptions.

The fact that a significant percentage of articles are published by high quality Brazilian journals with international standards shows the growth in the options to authors. The majority of the journals in Open Access indicate an alternative publishing model, where a journal is a source of prestige for the publisher, not just a profit.

Table 2
Estimated partial value of Brazilian Article Processing Charges in indexed journals ranked according to number of articles

Source: The authors (2019)

Fees in other currencies were converted to USD using the Google tool, October 17, 2018. The cost of publishing in journals with per page fee calculated at ten pages, and where multiple fee options were available, the lowest was used.

Rodrigues and Abadal (2014) report that, according to the number of nationally published journals, medicine and agriculture are the two most significant research areas in Brazil, both classified as life sciences & biomedicine here. UNESCO (2015) reports that the most relevant areas of publication by Brazilian authors are medical sciences, biological sciences, and agriculture, and three of these five are classified as life sciences & biomedicine.

As shown in Table 3, the most common publishers are commercial companies (65.0%), followed by universities (15.5%), associations (14.4%), and research institutes (2.7%). The volume of articles for each type was 50.7%, 22.9%, 20.1%, and 3.6%, respectively. The journals are published in the U.S.A. (34.3%), the U.K. (22.1%), Brazil (13.5%), the Netherlands (12.5%) and Germany (4.5%).

Although Brazil had the third-highest number of journals, it had the highest number of articles, which indicates a significant presence of indexed Brazilian journals in the most prestigious database. The article distribution shows 56.78% published in journals originating in the U.S.A, the U.K, the Netherlands, and Germany, where the big commercial publishing groups have their headquarters.

Among Brazilian journals, universities and associations (28.8%) publish the highest volume of articles, as shown in Table 4. Rodrigues and Abadal (2014) register that universities and associations publish 87% of Brazilian journals, which corroborates these results.

Table 3
Origin and publisher type of WoS-indexed journals with articles by Brazilian-based

Note: Art = articles, Jour = Journal *Other includes: Poland, Argentina, Mexico, Costa Rica, Venezuela, Greece, Belgium, Turkey, Croatia, Romania, Serbia, Czech Republic, China, Portugal, South Korea, Slovakia, Norway, Peru, Ecuador, Hungary, Iran, Pakistan, Taiwan, and Ukraine

Source: The authors (2019)

The distribution of journal impact factors according to publisher type, show that most of the journals’ (60.2) Impact Factor (IF) range between 0.001 and 3.000. A relevant number of journals (13.5%) did not have an IF: 82 in the arts & humanities, and 133 in the social sciences, 98.8%, and 70.4% of the total number of journals in each area. This distribution corroborates the hypothesis that these fields involve several journals with regional and/or specialized content and and/or are new in the base (PURNELL; QUEVEDO-BLASCO, 2013; ROMANOS DETIRATEL, 2003).

The journals’ IF ranged between 0.062 and 72.406. A total of 309 journals (13.4%) had impact factors over 4.0, publishing 5.527 (12%) articles by Brazilian-based authors. Only 31 journals had impact factors over 10.0, the highest being the Massachusetts Medical Society’s New England Journal of Medicine (72.406), which published 16 articles. The journals with IF over 10.0 were in life sciences & biomedicine (21), technology (4), and physical sciences (3).

The IF of the majority of journals produced by commercial publishers ranged between 1.0 and 4.0; for journals published by associations and universities, the most frequent IF ranges were 0.001 to 3.000 and 0 to 1.000, respectively. The distribution is equilibrated for journals published by research institutes.

Table 4
Impact factor, according to publisher type, of WoS-indexed journals with articles by Brazilian-based authors in 2016

Source: The authors (2019).

Life sciences & biomedicine, physical sciences, and technology were the main research areas of commercially published journals with articles by Brazilian-based authors in 2016, while social sciences and arts & humanities were the main areas of those published by universities, as presented in Table 5.

Table 5
Distribution of research areas according to publisher type in WoS-indexed journals with articles by Brazilian-based authors in 2016

Note: Art = articles, Jour = Journal

Source: The authors (2019).

The commercial publishers with the most journals are Elsevier, Springer, Wiley and Taylor & Francis, groups with a strong presence in WoS. Elsevier had the highest number of journals among these publishers, with 25.9% of journals and 22.4% of articles (Table 6).

IEEE had the highest number of journals for an association, with 35, while the seven journals published by the American Physical Society had the highest number of articles (793). The university with the highest number of journals was Oxford (47), while the highest number of articles (621) appeared in the four journals published by the Universidade Federal de Santa Maria.

Among research institutes, the Institute of Physics Publishing had the highest number of journals (26) and articles (465), while EMBRAPA (The Brazilian Agricultural Research Corporation) was in second place with 412 articles in two journals.

These results could promote discussion regarding the migration of research within the scientific publishing systems, in which peripheral countries send their best research to big commercial publishers and, in return, are required to pay high fees to access them. Governments are the primary source of research funds and should ultimately be responsible for the distribution of results. Brazil's government supports high-quality international journals published mainly by local universities and scientific associations. This alternative system is resulting in an increase in both quality and visibility of the journals and the articles.

Table 6
Frequency of publisher type among WoS-indexed journals with articles by Brazilian-based authors

Source: The authors (2019)

An essential part of any research project is to publish the results in a recognized journal. The articles published indicate the scientific activity of a country and its institutions, thereby identifying its authors and the state of the art for each area. Publication in indexed journals plays an important part in the evaluation of researchers, programs and institutions, as well as reflecting the ethos of each area. The publishers who own the journals play a crucial role in the scientific behaviour of each area.

The distribution shows that around 75% of journals are responsible for less than 40% of articles, and almost half of the articles are published in Open Access journals. This high percentage of open articles is a result of the publication of Brazilian titles; most of which open without APC and financed by universities, research institutions, government, and the new Mega Journals from the U.S. and the U.K., mainly PLOS One in the U.S. and Scientific Reports in the U.K.. The low APCs or lack of taxes from Brazilian titles contrast with the prices charged by commercial publishers, around US$ 1.500,00 for each article. The relevance of the mega-journals introduces a new player into the publishing system, with new rules and costs. The impact of this new journal model deserves further studies, especially with the permanence of high-cost subscription journals in the same system.

The Impact Factor of the majority of journals is between 0.1 and 3, with commercial titles showing the highest, and university titles showing the lowest. The differences among the areas reflect the predominant type of publisher, with Life Sciences and Biomedicine accounting for around 60% of journals and articles, with more than half of these published by commercial companies. Physical Sciences and Technology are also dominated by commercial publishers and represent around 20% of journals and articles. Social Sciences and Arts and Humanities show the dominance of universities as publishers and collectively represents around 13% of journals and titles. This distribution is equivalent to the global scenario, showing an alignment with other countries’ publication patterns.

Commercial publishers are responsible for more than half of all articles and journals. Elsevier with almost a quarter of the journals and articles in this study, Springer with 12% of journals and 9% of articles, followed by Wiley with 10% of journals and 6% of articles. Journals from open access companies follow, highlighting Public Library Science and PLOS One as the titles with the highest number of articles. The associations and universities play an equivalent role, with around 15% of journals and 20% of articles. It is important to note that the Open Access platinum is a characteristic of Brazil and Latin America, as the majority of universities and associations in other countries charge higher amounts for subscription or APCs.

Studies of a country’s publication scene rarely mention the costs behind them, focusing mainly on rankings and impact factors. This article is an attempt to expand the discussion regarding the availability and the costs to a country of accessing its own articles.