ALTERNATIVE AND TRADITIONAL INDICATORS OF THE BRAZILIAN FEMINIST

The connection between traditional and alternative indicators of academic impact has been recommended by scholars in the field of metric studies as they are considered complementary and offer a more complete picture of the evaluation of research results. From this perspective, the present descriptive exploratory study aimed to analyze the correlation coefficient between traditional and alternative impact of Revista Estudos Feministas. The sample included the period from 2001 to 2018. Analysis data were obtained from the Dimensions database via API. 1,296 texts with 2,609 citations and 4,984 altmetrics data were analyzed. The results showed that Mendeley obtained a positive evaluation in the correlation of Pearson's coefficient when compared with the quantity of published articles and with the citation and visibility indicators. The study found that altmetrics data is 4.6 times larger than traditional data, although fluctuating is important for indicating the tools and networks in which readers and users share scientific content.


Introduction
Scientific communication follows the trend of open science, envisions new applications and uses for scientific content: visibility, online attention and citation are resources that did not exist with print publishing and without the internet, social and academic networks allow a broad look on the use of scientific content globally which is no longer local and fragmented. Scientific publication still exists, the parameterization methods too, the change happens in how the scientific communication is made and analyzed (Anglada and Abadal 2018).
Although citation analysis is one of the most recurring themes of traditional bibliometric studies and one of national Information Science researchers' favorite approaches (Vanz 2003), it covers only the part of the so-called formal communication. Visibility studies, online attention, scientific and alternative measurement of publications enable to analyze the scenario of communication, circulation and scientific evaluation, permeate the space of quantification and get into the uses given by the society connected to the Internet network when accessing, reading, enjoying, sharing and eventually citing the results of a search.
Measurement systems are evolving as well as the molds of scientific achievement too, as scientific publication is thought as a service, it is possible to think of the reader, user of this information as the key element, in this case the user is no longer just a scientist, as they stop using only formal channels of communication, scientific dissemination is being studied from the perspective of marketing, precisely to meet new needs that were not previously presented. Araújo (2018) makes this observation, in which indicates that "social web resources and social media" modify the idea of impact of what is considered standard, and that the use made of this type of content in the digital environment can provide both the citation regarding the indication of interest for certain content demanded by the readers.
Scientific communication becomes the object of use in a complex information service, which fits new parameters, to which it is still adapting. citations? In this search, the research recognizes these metrics complementarity relationship and considers that they need to be used together, as they measure distinct impacts.

Scientific Measurement in Times of Social Network on the Web
With the vastness of communication systems available on the web and accessible via mobile devices, it is difficult to measure the amount of information being generated and consumed every day, but not everything that circulates is true, verifiable or scientific.
Selecting the information that is true and relevant can become a daunting task.
In science it is not different, the peer review system developed by the first scientific journals around 1660 (Belcher 2009) is still the method to verify the credibility of texts published in a journal, a practice that has not changed, but if adapted from handwritten letters, later typed to typed and sent by e-mail or editorial management systems, published in paper form and now accessed online.
Due to the growing number of publications, the scientific community has developed a quality measurement system called the "impact factor" created by Garfield (1972) in 1955, which presents the idea of citation index and its impacts on the scientific community. As well as the citation indicators metrics, impact factor, the traditional metrics are derived from Bibliometrics and Scientometrics, which use scientific publication as an object of study, intended to complement traditional metrics, such as the impact factor, from scientific publication. (Priem et al. 2010).
Scientific journals that have no data access and use restrictions facilitate altimeter applications and results can be measured immediately after online dissemination. Koon-Kiu and Mark (2011 p. 03) identified that the peak of scientific publication views has a "lifetime", on average two months, decay occurs soon after the first month until stagnant. This is why widespread dissemination after publication becomes crucial, and interaction with the public becomes vital for the spread of scientific achievement, in which everyone can be content producers, broadcasters and receivers, allowing visibility to the produced content (Priem, Groth and Taraborelli 2012 This behavior demonstrates the spontaneity in which scientific information circulates on the web and impacts on society. The path that scientific information goes beyond the walls of academia becomes a way of identifying whether the published document impacts society immediately after its publication, highlighting the scope and influence of each article inside and outside the academy.
It is noteworthy that the author's goal is to be read, and that his research is useful, so the assessment and especially measurement systems need to consider the various uses of scientific information applied today, expanded by the influence of social and academic networks on the web. Fachin

Methodology
The descriptive exploratory study, with the application of the quantitative method of data analysis. A descriptive exploratory research aims, according to Cervo, Bervian and Silva (2007, p. 63) to perform "precise descriptions of the situation and wants to discover the relationships between its component elements" so that it can ascertain the scenario from which the object of analysis constitutes itself.
According to Mukaka (2012 p. 69), the statistical method is "used to evaluate a possible linear association between two continuous variables." Regarding the use of correlation, the author emphasizes that, "the term correlation is used to refer to an association. connection or any form of relationship, bond or correspondence [...] ranges from -1 (perfect negative correlation) to 0 (no correlation) to +1 (perfect positive correlation)" (Mukaka 2012 p. 69). This method has been applied in altmetrics studies in several areas, mainly in health.
Due to the characteristics of the sample and the object of the study, Pearson Correlation Coefficient (r) was applied to the altimetric attention and citation scores.
Following the trend of studies such as Zahedi, Costas and Wouters (2014) The reason for choosing the journal as the object of the study is due to some characteristics: open access Latin American publication with international representation, indexed in SciELO (2) and with established impact factor1. For the choice of the base, one that was free to access the texts and metric data of the publications was chosen, which is the reason for choosing Dimensions.

Feminist Studies Journal
The Dimensions is one of Digital Science's database reference services, which indexes bibliographic data from publications and crosses altimetric data for possible sources in which content may have been mentioned.
The good thing about this bank is that it is open, meaning that anyone can access and search all its content. Another issue is the crossing of data from various academic and social sources, facilitating the researcher to obtain the scenario of Scientometric, bibliometric and altmetrics analysis, as with this research. One negative aspect is that accessing and downloading advanced search reports requires requesting an API key.

Data Presentation and Analysis
According to the research data, which can be seen in table 1, in the analyzed period,  Chart 1 -Correlation of Variables.

Correlation Linearity
Year (18) x Published Articles (1.296) = r 0.79 Year (18)  to Altmetrics and Mendeley data, which represents that these indicators impact the sample more than the other indicators.
For the journal, negative indicators may indicate that publications have had a significant visibility in social and academic networks, presenting numbers that are discrepant with the scenario of traditional metrics, so they did not obtain perfect linearity in the analyzed sample.
Another issue is that the altimetric analysis tools started to measure the publications from the creation of social and academic networks and the tool itself, it is not a retroactive analysis, it aims to ascertain the current scenario using relationship indicators to try to draw a profile impact (traditional and alternative) in future publications.
For the analysis of data dispersion, the closest positive perfect correlation indicators were used, those approaching the 1+ and analyzed throughout the 18 years of publication (See graphs 1, 2 and 3). gender" alone received 189 alternative mention out of the 425 in the entire sample. If we compare with the total of citations (308) from this same period, we can see that this text alone obtained more than half of the mentions. You can see the data in table 1. In the 2008 sample, it was observed that some texts obtained a significant number of alternative mentions, helping to increase the alternative data, and thus measuring a negative linearity for the sample of the observed period.  2004, 2008, 2011, 2012, 2013, 2014 and 2017, in which Mendeley's data outperform the citation data, presenting a negative linearity, which means that the texts of this period had a higher visualization index than the citation, even thus, this coefficient presents a positive degree, since in 11 periods the data remained closer to a positive than negative linearity.  2001,2002,2005,2006,2008,2012,2013,2014 and 2017, during this period Mendeley's visualization data stand out, much more than in periodic, this means that the texts had good visibility and readability. Fachin The studies by Erdt et al. (2016) indicate that Mendeley has a great coverage of altmetrics data, however, these data do not infer any relationship that Mendeley can leverage traditional citations, but the fact is that if the text is being viewed and read the chances for this material to be cited increase due to the high visibility index.
Silva Filho and Vanz (2018) collaborate with this view by emphasizing that social media helps to share and disseminate research, increasing the visibility capacity that this material will have. Moreover, they indicate that Mendeley proves to be a significant tool for scientific dissemination.
The analysis of the sample obtained from the REF, presented in table 1, indicates that the altmetrics data are representative and that the use of social and academic networks and media effectively contributes to the visibility of the scientific dissemination of the content produced by the analyzed journal.

Conclusions
The application of alternative metrics is also due to changes in scientific production and publication, from the high cost of scientific publication to open access, the restrictive visibility of publications in formal dissemination channels to informal web channels and social networks, and citation counting as the only indicator of the impact of a scientific production, ignoring the other uses of knowledge in electronic media. Fachin This study identified that academic tools and social media present intriguing data about the uses made of scientific content and that altmetrics data, as for this sample, are 3,8 times larger than the traditional one. It was also apparent that the Mendeley tool, widely used by the journal, presented a high Pearson coefficient correlation index, indicating that the dispersion between publication and visibility by this tool is 1.9 times greater than any other.
The low use of other media such as Twitter, Facebook cannot be ignored, these elements may indicate the profile of readers of the journal and which the preference of access and use of the content offered by the journal is, here is a detailed mapping that enables to track where users are and who the potential influencers of this particular content are. This type of study helps to glimpse the uses and if they influence impact in any way, whether traditional or altmetric.
The coefficient analysis with all Altmetrics x Citation data showed low linearity, this issue is widely emphasized by Erdt et al. (2016), when they indicate that, although both indicators (altmetrics and citation) have a strong correlation, they follow different paths, while complementary, when one strengthens the other and present scenarios about scientific communication that was previously ignored by scientific society, serving as contributions for various purposes. This is the relationship between negative and positive dispersion coefficients, it represents to the journal that traditional and altmetrics indicators are not aligned and that the peak of visibility and interaction in social and academic networks are higher than the traditional indicators of measurement. This is what altmetrics presents, how scientific publication performs on the internet, as well as its uses by indiscriminate subjects with access to the material.
Finally, survey studies need to be done in order to identify similarities, patterns or otherwise between indicators, formulas, methods or sample types to be researched and used for the purpose of dissemination and use of scientific publication and accessible to all.

Notes
(1) The calculations and diagram were made by Jean Carlo Gengnagel -Bachelor student in Scientific Mathematics at UFSC is GETMORE Brazil. email: jeangen00@gmail.com. We are grateful to the National Council for Scientific and Technological Development -CNPq for aid to research (Process: 426777 / 2016-6).