Citing and referencing habits in Medicine and Social Sciences journals in 2019
Erika Alves dos Santos, Silvio Peroni, Marcos Luiz Mucheroni
CCiting and referencing habits in Medicine and Social Sciences journals in 2019
Erika Alves dos Santos – School of Communication and Arts (ECA), Department of Information & Culture (CBD), University of São Paulo (USP), Brazil / Fundação Jorge Duprat Figueiredo de Segurança e Medicina do Trabalho (Fundacentro), Brazil / Digital Humanities Advanced Research Centre (DHARC), Department of Classical Philology and Italian Studies, University of Bologna, Italy Silvio Peroni – Digital Humanities Advanced Research Centre (DHARC), Department of Classical Philology and Italian Studies, University of Bologna, Italy / Research Centre for Open Scholarly Metadata, Department of Classical Philology and Italian Studies, University of Bologna, Italy Marcos Luiz Mucheroni – School of Communication and Arts (ECA), Department of Information & Culture (CBD), University of São Paulo (USP), Brazil
Abstract
Purpose – This article explores citing and referencing systems in Social Sciences and Medicine articles from different theoretical and practical perspectives, considering bibliographic references as a facet of descriptive representation.
Design/methodology/approach – The analysis of citing and referencing elements (i.e. bibliographic references, mentions, quotations, and respective in-text reference pointers) identified citing and referencing habits within disciplines under consideration and errors occurring over the long term as stated by previous studies now expanded. Future expected trends of information retrieval from bibliographic metadata was gathered by approaching these referencing elements from the FRBR Entities concepts.
Findings – Reference styles do not fully accomplish with their role of guiding authors and publishers on providing concise and well-structured bibliographic metadata within bibliographic references. Trends on representative description revision suggest a predicted distancing on the ways information is approached by bibliographic references and bibliographic catalogs adopting FRBR concepts, including the description levels adopted by each of them under the perspective of the FRBR Entities concept.
Research limitations – This study was based on a subset of Medicine and Social Sciences articles published in 2019 and, therefore, it may not be taken as a final and broad coverage. Future studies expanding these approaches to other disciplines and chronological periods are encouraged.
Originality – By approaching citing and referencing issues as descriptive representation’s facets, findings on this study may encourage further studies that will support Information Science and Computer Science on providing tools to become bibliographic metadata description simpler, better structured and more efficient facing the revision of descriptive representation actually in progress.
Keywords
Bibliographic references, in-text reference pointers, information representation, bibliographic metadata, FRBR, reference styles.
Paper type
Research paper ntroduction
Providing bibliographic information to make documents searchable and retrievable through several access paths using structured metadata corresponds to one of the purposes of descriptive representation . Descriptive representation is a process in which we identify data related to the editorial production of documents, such as their authors, titles, publishers, years of publication, number of pages, sources, and other bibliographic elements (Galvão, 1998; Lancaster, 2004). Although descriptive representation is still considered the physical description in manual catalogs (Maimone et al. , 2011), it is also a way to materialize and make evident information’s mea ning, context, and relationships (Lancaster, 2004) between documents. One of the products of this type of information representation is the bibliographic reference (Galvão, 1998), considered by Baptista (2007) as the elaboration of records containing the descriptive representation. Bibliographic references are a facet of descriptive representation. They correspond to a specific kind of description that acts like a reference that points the reader to an original source of information cited by an author in the text body of a work. Each bibliographic reference contains the textual representation of a minimal set of descriptive bibliographic metadata which enables the identification of a publication, a speech, a piece of information, or anything else that may be citable, to locate and retrieve it (ABNT, 2018; Cunha and Cavalcanti, 2008; ISO, 2010). This article aims to study bibliographic references and the other contextual entities – namely mentions, quotations, and in-text reference pointers , summarized in Figure 1 and Figure 2 – from several perspectives. Baptista (2007) argues that bibliographic references are now being written by different professionals, according to their interests – scientists, artists, enterprises, negotiators, publishers, libraries, archives, museums, etc. – coming from multiple locations. Bibliographic references are one of the tools that establish links between scholarly works and enable the creation of citation networks. Writing bibliographic references correctly demands previous background and some Information Science skills. It requires that both reference and citation styles guidelines are presented in comprehensive language, with a wide scope of the bibliographic universe, especially considering Baptista’s statement. Information Science should have a primary role in this activity. However, since researchers in all the disciplines usually deal directly with this aspect without involving experts, one can introduce mistakes in the bibliographic references that prevent the clearly and unambiguously identification of the represented (i.e. cited) works. Citation error is an old issue from the XVII Century (Sweetland, 1989), and started before the publication of the first reference style manuals – the Hart's Rules for Compositors and Readers (1893) and the University of Chicago Manual of Style (1906) (O’Connor, 1977). Despite the publication of those style manu als, “citation errors continued to appear, as did an increasing number of complaints about them” (Sweetland, 1989, p. 293). Although Sweetland’s study was published decades ago, his arguments remain updated and most of his conclusions are remarkably similar to the findings of our study suggesting that neither the publication of reference styles nor the technological advances were effective in solving those bibliographic issues. For the sake of clarity, the definitions of all the bibliographic terms mentioned above and the others used in this study can be found in Santos et al. (2020a). igure 1.
Visual representation of the main bibliographic elements involved in citing and referencing author-year system – i.e. when in-text reference pointers referring to mentions and quotations include author’s surnames and year of publication of the cited works described in a particular bibliographic reference. In-text reference pointerAuthor-data citation system
Sweetland ( states that “ bibliographic references are an accepted part of scholarly publication ” .Despite the important role played by citations within scientific works, Sweetland (1989) states about the high percentage of errors found in citations from the nineteenth century to the present and complements that Such errors can be traced to a lack of standardization in citation formats, misunderstanding of foreign languages, general human inabilities to reproduce long strings of information correctly, and failure to examine the document cited, combined with a general lack of training in the norms of citation (Sweetland, 1989, p. 291).
Lastly, Sweetland (1989) argues that despite the publication of style manuals, citation errors continued to appear and point that the real problem, which is the failure on detecting and correcting citations errors, is a consequence of the sharing responsibility in the publishing process.
References
Sweetland, J. H. (1989). “Errors in bibliographic citations : a continuing problem” , The library quarterly,
Run-in quotation Passage authored by the work’s authorMentionBibliographic references session title Bibliographic reference
In-text reference pointerAuthor-data citation system
Double quotation marks
In-text reference pointerAuthor-date citation system
In-text reference pointerAuthor-data citation system
Long quotation
Figure 2.
Visual representation of the main bibliographic elements involved in citing and referencing citation-sequence system – i.e. when in-text reference pointers referring to mentions and quotations are specified using a number corresponding to a particular bibliographic reference in a bibliographic references list arranged in ascending numerical order Regarding the rate of errors in citations within scientific journals, there is no consenus regarding whose responsibility is to correct them. ‘ ; Publishers seem to feel it is up to the author(s) to provide correct citations; the authors seem to feel it is up to referees to doublecheck them; no one, except perhaps librarians, seems to care very much about the problem ’ ’ . References
1. Sweetland, J. H. (1989). “Errors in bibliographic citations : a continuing problem” , The library quarterly,
In-text reference pointerCitation-sequence system
Run-in quotationMention
In-text reference pointerCitation-sequence system(correspondence with bibliographic reference number)
Bibliographic referenceBibliographic references session titleSingle quotation marksBibliographic reference number(corresponding to the in-text reference pointer)
Starting from the premise that the problems referred by Sweetland still stand, the purpose of this article is to reflect on how recent journal articles published worldwide in all the subject categories of two main academic disciplines (i.e. Medicine and Social Science) organize their citation apparatus. We focus on how bibliographic references are defined in published articles and the way they are denoted ithin the article text through the related in-text reference pointers to support the author's argumentation. In particular, we aim at answering the following research questions (quRQ1-RQ4 from now on): 1.
Considering the introduction of technological tools such as reference manager softwares and the instructions provided in the reference styles adopted by journals, were they enough to fully address all the problems pointed out by the study made by Sweetland in 1989? 2.
Which are the main possible causes for errors in mentioning, quoting, and referencing practices? 3.
Do bibliographic references and in-text reference pointers refer to the same cited entity at a conceptual level in terms of the FRBR specification? (IFLA Study Group on the Functional Requirements for Bibliographic Records, 2009) 4.
What is the potential impact of using FRBR within bibliographic catalogs on information retrieval?
Material and methods ct headings which are based on those available in Elsevier’s Scopus (SCImago, c2020; Guerrero-Bote and Moya-Anegón, 2012). The description of the full methodology we followed for the selection is provided in Santos et al. (2020c) and is briefly summarized as follows. As of 1 November 2019, SCImago contained 24,702 records of journals. Since we were interested in gathering information related to two subject areas, i.e. Medicine and Social Science, we looked only at the 7,224 records related to Medicine journals (17,93% of SCImago journals records) and the 5,855 records related to Social Sciences journals (14,53% of SCImago journals records). We used three criteria to select the journal titles to include in our sample: the “total cites” metric, the publisher, and the subject categories.
The first criterion, i.e. “total cites” metric in SCImago, corresponds to the number of citations received by the articles of a journal that have been published in 2015-2017, coming from the works (of any discipline) published in 2018. We considered this metric as a proxy for the prestige of journals – as pointed by Bonacich (1987, p. 1172), “high prestige is acquired by receiving unreciprocated choices from others”. We ordered the journals in SCImago in decreasing order according to their “total cites” values to consider journals that are the most relevant ones in their particular subject area. The publisher of the journal (i.e. the second criterion) was considered to have heterogeneous coverage. In particular, we avoided having more than one journal per publisher for each subject area to prevent scenarios where a publisher applies the same editorial policy for all its journals. Lastly, we considered journal subject categories (i.e. the third criterion) for the selection. Each SCImago subject area is further classified in categories that define specific subheadings for a particular journal, e.g. Hematology for Medicine and Demography for Social Sciences. We selected only the journals that were classified, in SCImago, under a single subject area, and a single subject category to avoid possible overlaps between the selected journals in terms of areas and categories. e decided a threshold date, i.e. 31 October 2019, as an upper bound to select journal regular issues (special issues were not considered) within the selected journals. In case a journal did not publish in issues, we considered, as a virtual regular issue, all the articles published in the previous month, i.e. between 1 October 2019 and 31 October 2019. For each journal issue selected which was not a special issue, we randomly chose 5 articles. Only the articles introducing original research communications were considered. If the selected issue did not contain all the 5 articles, we considered all the published articles within it. Each article was individually analyzed considering quantitative and qualitative aspects concerning its mentions, quotations, and bibliographic references. We assumed that the citation behavior (i.e. the way mentions, quotations, bibliographic references, and in-text reference pointers are specified) is similar among different regular issues of the same journal. Data
Considering the data available on SCImago in November 2019, we selected 46 journals and 213 articles from both the subject areas in consideration, i.e. Medicine and Social Sciences. The articles contained a total amount of 9,911 bibliographic references and 16,193 mentions and quotations overall. We obtained 27 journals, 132 articles, 5,340 bibliographic references, 8,400 mentions and 5 quotations in Medicine, and 19 journals, 81 articles, 4,571 bibliographic references, 6,953 mentions, and 835 quotations in Social Sciences. The sample comprises journals owned by publishers from 13 different countries. Medical journals are from Canada, China, Egypt, Germany, Hong Kong, Netherlands, New Zealand, Switzerland (each country representing 3.70% of the Medicine sample), United Kingdom (representing 29.63% of the Medicine sample) and United States (representing 40.74% of the Medicine sample). Social Sciences journals are from Brazil, Germany, Portugal, South Korea (each country representing 5.88% of the Social Sciences sample), United Kingdom (representing 41.18% of the Social Sciences sample), and United States (representing 35.29% of the Social Sciences sample). Gathered data included information to support a discussion from multiple points of view, i.e. the publisher, journal, and article perspectives. Also, we collected more granular data about in-text reference pointers and bibliographic references, which provided a view on the citation apparatus. These different viewpoints supported an overview of the variations with which bibliographic data appears and how they relate to each other in articles considering the subject areas we analyzed. All the data used in our study are available in Santos et al. (2020b).
Results
33% of the Medicine journals and 35% of Social Sciences journals composing the sample are Gold/Diamond Open Access journals . From an article perspective, 42% of the Medicine articles and 52% of Social Sciences articles can be freely accessed from the journal website without paying any fee – because either the journal is a Gold/Diamond Open Access one or it is a Hybrid journal. Analyzing the works cited by the articles included in our sample, we observed that 55% of the works cited by the Medicine articles are freely available online (either as Green, Gold or Diamond Open See (Piwowar et al. , 2018) for a definition of all the Open Access levels. ccess items) and 50.15% of the bibliographic references referring to them do not provide a DOI URL (e.g. an URL starting with either “http(s)://doi.org/” or “http(s)://dx.doi.org/” followed by a DOI) to access them directly from the Web. Along the same line, only 35% of the works cited by Social Sciences articles are freely available online and only 21% of the bibliographic references referring to them do not provide a DOI URL. Considering all the bibliographic references included in Medicine articles providing a URL that did not include any DOI (even when hidden behind a hyperlink), 20% of them specified a URL, while only 12% of bibliographic references in Social Sciences articles contained an URL. However, 42% of the URLs indicated in bibliographic references of Medicine articles often referred to records within a bibliographic database (e.g ., a library’s bibliographic catalog, like Pubmed ). We observed similar behavior in 2% of the bibliographic references in Social Sciences articles. The data we gathered included the reference styles adopted by the journals, according to the recommendations in the instructions provided for authors. We noticed a huge variety of reference styles among the journals in the sample, as shown in Table I. Table I.
Percentage of the adoption of reference styles in the journals in our sample
Reference style Medicine subject area journals Social Science subject area journals
AMA Reference Style (a.k.a. AMA Manual of Style) 7% 0% AMA Reference Style 6h ed. 0% 5.6% AMA Reference Style 9th ed. 7% 0% AMA Reference Style 10th ed. 11% 0% APA Reference Style 0% 5.9% APA Reference Style - 6th ed. 2010 0% 23.5% ASM Reference Style 4% 0% Author should consult sample issue for guidelines 0% 5.9% BMJ Reference Style 4% 0% Citing Medicine 2nd ed. 4% 0% Harvard Reference Style (a.k.a. Harvard Reference System) 0% 5.9% SAGE Harvard 0% 5.9% The Chicago Manual of Style 4% 0% The Chicago Manual of Style 16th ed. 0% 17.6% The Chicago Manual of Style 17th ed. 0% 5.9% The Chicago Manual of Style 2006 rev ed. 0% 5.9% Vancouver Reference Style (a.k.a. ICMJE) 29% 0% Own reference style 26% 17.6% Not identified reference style (no information available in the journal website) 4% 0% PubMed is a free resource supporting the search and retrieval of biomedical and life sciences literature with the aim of improving health – both globally and personally (available in https://pubmed.ncbi.nlm.nih.gov/about/).
0% of Medicine journals adopted widely accepted reference styles like Vancouver, Chicago, AMA, and APA. Among these, 73.7% customized the adopted reference style guidelines according to their specific needs. Such customization practice was not observed in any Social Sciences journal in our sample, which usually directly reused the original reference style chosen as is. Considering only journals providing own reference styles (Medicine: 26%; Social Sciences: 17.6%), the reference styles of 33% of Medicine journals and 23.5% of Social Sciences journals did not provide clear, comprehensive, and exhaustive guidelines for accurately describing and arranging bibliographic metadata in bibliographic references. The reference styles can be classified into three non-disjoint categories: 1. those which did not provide guidelines for describing some types of publications like grey literature, e-prints, technical reports, speeches, etc.; 2. those which did not provide instructions on how to properly establish correspondences between bibliographic references and the in-text reference pointers used in mentions and quotations and; 3. those which did not provide instructions on how to proceed with particular bibliographic issues, like citing secondary or indirect sources (e.g. quoting quotations), structuring and formatting DOI metadata, etc. In Social Sciences articles, we noticed variations in the title attributed to the section containing bibliographic references. 87.6% of articles name it
References , 6.2% name it
Referências (References, in Portuguese, following the article’s language), and
Notes . 100% of Medicine articles name the bibliographic references section as
References . The analysis of the type of publications cited by the articles in our sample showed 33 different types of works, as shown in Table II.
Table II.
Percentage of the types of the cited works derived from analyzing the related bibliographic references of all the articles in our sample
Cited work publication type Bibliographic references in Medicine articles Bibliographic references in Social Sciences article
Table III.
Percentage of the export formats of bibliographic records of the articles published in the journals in our sample
Exportation format Medicine journals Social Science journals Textual format
ABNT 0% 7% AMA 4.2% 0% human readable)
APA 4.2% 7% CBE (Reference style for the Council of Biology Editors format) 4.2% 7% Chicago (a.k.a. Turabian) 8.4% 7% MLA 4.2% 7% Vancouver 4.2% 0% Text (a.k.a. simple text or plain text) – refer to journals providing bibliographic references with no specification of the reference style in which it is formatted 12.5% 0% Structural format (machine readable)
BibTex 66.7% 13% Bookends 16.7% 0% Easybib 16.7% 0% Endnote 50% 20% Endnote 8 16.7% 0% Endnote Tagged 8.3% 0% Medlars 41.7% 0% Mendeley 20.8% 0% Papers 12.5% 0% ProCite 16.7% 20% Reference Manager 20.8% 7% RefManager 12.5% 0% RefWorks 41.7% 20% RefWorks Tagged 12.5% 0% Zotero 12.5% 13% RIS 62.5% 20% 96% of Medicine articles and 97.5% of Social Science articles provided their own bibliographic metadata in headers, footnotes, or the first page of PDF files they provide. Usually, articles adopt one out of two systems for identifying in-text reference pointers. The first one is the author-date system – e.g. “(Doe, 2020)” – adopted by 3.7% of Medicine articles and 88.2% Social Sciences articles. The second, the citation-sequence system , based on numbers (e.g. “[3]”), was adopted by 96% Medicine articles and 5.9% Social Sciences articles. The remaining 5.9% Social Science articles adopted both styles within the same articles. All these data are introduced in Table IV. Considering articles adopting the citation-sequence system, we observed variations on the format of the numbers associated with each bibliographic reference, as shown in Table V. Within this system, the numerical arrangement of bibliographic references did not correspond to the order in which the respective in-text reference pointers appear in the text in 8.66% of Medicine articles. For instance, let us consider the first paragraph of a Medicine article, shown in Figure 3. The first in-text reference pointer (e.g . “[41]”) refers to the bibliographic reference in position 41 and the second one ( e.g. “[40]”) to the bibliographic reference in position 40. Similarly, the three in-text reference pointers that follow o not follow an ascendant numerical sequence: [1, 21, 31], [13, 27], [31, 34]. The journal in which this article was published adopts Vanc ouver reference style, according to which “references should be numbered consecutively in the order in which they are first mentioned in the text” (ICMJE, c2020), something that is not happening in this case. No additional instructions regarding the numerical arrangement, neither for in-text reference pointers nor for the bibliographic references, were found within the reference style. Table IV.
The format used for representing in-text reference pointers within articles
System In-text reference pointers format Medicine articles Social Sciences articles
Author-date
Author’s surnames and year of publication in round brackets, e.g. (Rowley, 2000) 3.7% 88.2% Citation-sequence Superscript numbers, e.g. [ ] 37% 11.8% Superscript numbers in square brackets, e.g. [1] Table V.
Different uses of the numbering systems for ordering bibliographic references in the bibliographic references list.
Numbers format Subject Areas square brackets (e.g. “[1]”) unformatted (e.g. “1”) unformatted with a dot (e.g. “1.”) bold with a dot (e.g. “1.”) superscript (e.g. “1”)
Medicine 11% 9% 68% 12% 0% Social Sciences 0% 0% 50% 0% 50%
Figure 3.
An excerpt from a paragraph in a Medicine article in our sample We observed misuses regarding the alphabetical arrangement of bibliographic references in 40% of Medicine articles and 3% of Social Sciences articles adopting the author-date citation system. Figure Passages referred in Figures 3 to 8 were extracted from articles in our sample.
The World Health Organization (WHO) defines mental health as “a state of well-being in which the individual realizes his or her own abilities, can cope with the normal stresses of life, can work productively and fruitfully, and is able to make a contribution to his or her community.” [41] One in four people worldwide are affected by mental health concerns [40]. includes an excerpt of the bibliographic reference list of one of the articles in our sample, in which its bibliographic references were expected to be sorted in ascending order by authors’ surnames.
Figure 4.
An excerpt of the bibliographic reference list of an article in our sample Some publishers also provided hyperlinks to connect in-text reference pointers to the corresponding bibliographic references. This behavior was observed in 60% of Medicine articles and 25% of Social Science articles. However, the reciprocal hyperlink, i.e. between a bibliographic reference to the in-text reference pointers denoting it, was not a usual feature provided in the Medicine articles. Instead, considering only the articles with hyperlinked in-text reference pointers, 100% of Social Sciences provide backlinks to the in-text reference pointers, while only 10% of Medicine articles implement such functionality. Regarding the uniformity of descriptive metadata, journal titles of cited articles may appear in distinct forms within bibliographic references. These differences may exist either when comparing bibliographic references list of articles from different issues of the same journal or even between bibliographic references included in the same bibliographic references list. Differences regarded to the abbreviation of journal titles were observed only in Medical journals since 100% of Social Sciences articles gave full titles of cited journals in their bibliographic references. Considering the 85% of Medicine articles providing titles of cited journals in the abridged format in their bibliographic references lists, 4% adopts the ISO 4 rules , 9% adopts the ISSN List of Title Word Abbreviations (LTWA) guidelines , 4% adopts the recommendations of the National Center for Biotechnology Information Database, 43% adopts the recommendations of US National Library of Medicine (NLM, also referred by publishers as Pubmed Database and Index Medicus) . For the remaining 39% of Medicine journals, the source in which abbreviations should be based on was not identified. Refers to a Standard published by the International Standardization Organization (ISO), entitled “ISO 4:1997
Information and documentation — Rules for the abbreviation of title words and titles of publications” Maparyan, L. (2012). The womanist idea. New York, NY: Routledge.
Morris, E. W. (2007). ““Ladies” or “loudies”? Perceptions and experiences of Black girls in classrooms.” Youth &
Society, 55(4), 490-515. Ladson-
Billings, G. & Tate, W.F. (Eds.). (2016) Covenant keeper: Derrick Bell’s enduring education legacy. New York: Peter Lang. Johnson, L. (2017). The Racial Hauntings of One Black Male Professor and the Disturbance of the Self(ves): Self-Actualization and Racial Storytelling as Pedagogical Practices. Journal of Literacy Research, 49(4), 476-502. Lorde, A. (1984). Sister Outsider: Essays and Speeches. Trumansburg, NY: Crossing Press. Matthew, P.A. (2016). Written/Unwritten: Diversity and the hidden truths of tenure. Chapel Hill, NC: The University of North Carolina Press. ur data revealed that articles did not usually provide sources of non-textual content – i.e. all that information presented by using visual signs, tables, graphs, photographs and images, illustrations, schemes, verbal communications, audio, and video recordings or any other type of manifestation made without using argumentative text as the main language. Non-textual content was observed in 93% Medicine articles and 72% Social Sciences articles. Of these, 85% of Medicine articles did not provide the source of non-textual content and 4% of articles provided the source for only part of non-textual content. In the Social Sciences articles in our sample, 41% of them provided the source of non-textual content and 9% of articles provided the source for part of it. Quoting and mentioning proved to be more frequent in Social Sciences articles than in Medicine articles. In Medicine, we observed an average of 64 mentions and 0.04 quotations per article, while in Social Sciences we had 86 mentions and 10.31 quotations per article. 71.6% of Social Science articles included at least one quotation, which seemed to confirm the impression that Social Sciences usually quote more compared with Medicine articles, in which only 2.27% of articles used quotations. About the markup used to show and identify quotations within a text, 100% of quotations in Medicine articles used double quotation marks (i.e. “”) to markup the text in run -in quotations. No long quotations were detected in Medicine articles in our sample. In Social Sciences articles, quotations were marked up in different ways. For run-in quotations, 74.4% of the articles adopt double quotation marks, 22.4% of the articles adopt single quotation marks (i.e. ‘’), and 1.7% of the articles adopt both double and single quotation marks. Considering long quotations, 29.3% of Social Science articles used simple indentation, while 3.4% of the articles presented quotations in indented passages and italicized characters between double quotation marks. Usually, providing the page numbers in the in-text reference pointer associated with a mention is optional. However, it was used in 1.28% of the articles in Social Science, while Medicine did not show such behavior. Instead, the page numbers in-text reference pointers referring to quotations were found in 60% of Medicine articles and 86% of Social Science articles. From a FRBR perspective, in-text reference pointers with page numbers were used to refer to a particular FRBR Manifestation of a cited document (i.e. a particular edition). Instead, the remaining in-text reference pointers, i.e. those without the specification of page numbers, actually pointed to the FRBR Expression level of the cited documents (i.e. their content), despite the particular format specified in the metadata of the related bibliographic references, which usually described the cited document at the Manifestation level. The article analysis revealed that 40% of Medicine articles and 48% of Social Sciences articles specified either in-text reference pointers or bibliographic references (or both) related to quotations that did not provide easy access to the text quoted. With easy access, we mean to locate a quoted passage within the cited document without having to perform complementary searches, like queries on indexes and summaries, or to read long excerpts to identify the quotation. One example of these cases is illustrated in Figure 5, which reproduces a passage from a Social Science article containing a quotation of a Collin’s work, published in 2000. We noticed that the in -text reference pointer matches two different bibliographic references in the bibliographic references list, both published in 2000. Since the in-text reference pointer do not provide any specification of which bibliographic references it referred to, e.g. by adding an alphabetical character to the date of publication, like 2000a, and ssuming that the provided pagination is correct, the only way of retrieving the quoted passage on the cited work is by consulting both works referenced by the two bibliographic references. Figure 5.
An excerpt from a Social Sciences article The excerpt in Figure 6, extracted from an article from the Social Science subject area, illustrates some of the issues mentioned above. In the original source, the year of publication of each cited work within the in-text pointer is connected to the respective bibliographic reference in the bibliographic reference list through a hypertextual link. By clicking on the link provided in the in-text reference pointer for “Okoh and
Hilson, 2011”, we were sent to the bibliographic reference shown in Figure 7.
Figure 6.
An excerpt of a Social Sciences article
Figure 7.
The bibliographic reference of one of the in-text reference pointer introduced in Figure 6 None of the works authored by Hilson described in the bibliographic reference list was co-authored with Okoh. The only work dated from 2011 considered in the bibliographic references list was the one reproduced in Figure 7. By checking the work represented in the bibliographic reference in the webpage of its publisher, we confirmed that Okoh was not an author of the work referred by the bibliographic reference. Thus, there is no evidence that the in-text reference pointer and bibliographic reference referred to the same work, neither that the quoted passage is actually a quotation or a mention incorrectly marked up as a quotation. Figure 8 illustrates a situation in which the author does not use the same approach to specify in-text reference pointers within the text. Indeed, since the in-text reference pointer referring to the bibliographic reference number 41 appears after the period ending a quotation and no page number was provided in it, readers have to infer autonomously to which passage in-text reference pointers 40 and 41 refer to since it is ambiguous whether 41 refers to the previous sentence or not. These are only In the text-body: It also develops space for Black women researchers to do work that Collins (2000) describes as activating “epistemologies that criticize prevailing knowledge and that enable us to define our own realities on our own terms...” (p. 292).
In the bibliographic references list: Collins, P.H. (2000). Black feminist thought: Knowledge, consciousness, and the politics of empowerment. New York, NY: Routledge.
Collins, P. H. (2000) “What’s going on? Black feminist thought and the politics of postmodernism."
In E. A. St. Pierre & W.S. Pillow (Eds.) Working the ruins: Feminist post-structural theory and methods in education (41-73). New York: Routledge. “This is the case for many countries such as Ghana, Tanzania, Senegal, and Mozambique (Fisher,
Mwaipopo, Mutagwaba, Nyange, & Yaron, 2009; Aizawa, 2016; Bryceson & Geenen, 2016; Hilson & Garforth, 2012; Okoh and Hilson, 2011; Persaud, Telmer, Costa, & Moore, 2017).”
Hilson, G. (2011). Artisanal mining, smallholder farming and livelihood diversification in rural Sub-Saharan Africa: An introduction. Journal of International Development, 23(8), 1031 – Figure 8.
Another excerpt from a Social Science article.
Discussion
In this section, we discuss the results introduced in the previous sections focusing on answering the four research questions (R1-R4) presented in the introduction. In particular, the discussion is organized in two parts. In the first part, we address RQ1 and RQ2, while in the second part we discuss RQ3 and RQ4.
Issues in citing and possible causes for errors r’s awareness o f promoting the access to freely available content may result in the higher reliability of the hypertext links provided in their bibliographic references. Second, we did not find detailed instructions regarding the provision of URLs and DOI hyperlinks within bibliographic references in reference styles (RQ1). In particular, while usually encouraging authors to provide these bibliographic data, the reference styles analyzed generally did not indicate clear guidelines regarding the description procedures neither on the location where those hyperlinks should point to, such as the actual file containing the cited work nor to the landing page from where it may be downloaded (RQ2). These behaviors suggest a resistance – by authors, editors, or both – on providing enough metadata in the bibliographic reference to facilitate the access to the cited works (RQ2). Although the articles’ content and its accuracy are authors’ responsibilities, there did not seem to be an adequate verification of the bibliographic references before their publication, which may suggest that the efforts devoted by publishers of the documents in our sample in providing trustful bibliographic data in their …community”. [41] One in four people worldwide are affected by mental health concerns [40]. rticles, especially concerning bibliographic references, were not as meaningful as those related to content quality. In this sense, Sweetland (1989, p. 300) considers that “the role of citations is not taken very seriously by the scientific community”. Cronin (1982, p. 71) complements that “journal editors and referees could pay greater attention to the quality and quantity of references". Our analysis showed the adoption of 20 different reference styles within the journals of the sample. According to Table 1, the average number of reference styles adopted by journals composing the sample is 45% higher in Social Sciences journals than what we observed in Medicine journals. Such diversity in the adoption of reference styles in the same subject area weakens the argument that the existence of multiple reference styles is justified by the specific needs in different disciplines (Gratz, 2016), specific rules for a particular audience (Barbeau, 2018) and by tradition (Barbeau, 2018; BibMe, 2017; Gratz, 2016), for instance: “social sciences tend to use current research, so the publication date of a source is very important. For this reason, people in the social sciences tend to use APA style; APA style puts the date before all other information, aside from the author’s name, which makes it much easier for researchers in that field to find valid, up-to- date information.” (UCM Writing Center, representing the same information. As stated by Sweetland, “complaints about lack of uniformity are common in the literature, whether from authors or librarians”. The author complements that “given the variety of formats for citation and the lack of any real agreement among journals or authors, the chance of misunderstanding is high” (Sweetland, 1989, p. 298). Some journals recommend more than one bibliographic reference managers within the instructions provided to authors for writing their manuscript. Updating the format used by such reference managers to return bibliographic references according to the reference styles’ guidelines can prevent journals from receiving differently formatted articles. Theoretically speaking, that could decrease the editorial work on checking bibliographic metadata accuracy.
From the reference manager administrator’s perspective, the more the number of reference styles available is huge, the greater the challenge of creating algorithms for recognizing and stylesheets for formatting bibliographic references appropriately. Besides, an incredible effort is needed to update such algorithms and stylesheets when, for instance, particular reference styles have variations on their citing systems (Barbeau, 2018) and multiple versions and editions of the same reference style are introduced in time. The diversification of guidelines may confuse researchers. A clear example of this situation is the name used by publishers to refer to reference styles in their homepages. According to the data we gathered, we noticed several variations – e.g. AMA Reference Style was often mentioned as AMA Manual of Style, Harvard Reference Style was also referred to as Harvard Reference System, and Vancouver Reference Style was also referred as ICMJE and Citing Medicine. We also noticed that the name attributed to the section containing bibliographic references in Social Sciences articles may change. For example, sections containing bibliographic references are named differently within articles adopting the 16th edition of Chicago reference style: 66% name it References” and 44% name it “Notes”. Besides, two different citation styles were observed on this same sample slice: citation-sequence and author-date, each one adopted by 33,3% of these articles. The remaining portion of articles adopts both citation systems. No misuses of adopted reference styles were observed in medicine articles, considering specifically these aspects. We speculate that, in time, failures regarding the interpretation of bibliographic guidelines description cannot be exclusively attributed to the availability or use of multiple reference styles (RQ2). Indeed, our analysis revealed that reference styles content could be clearer, including the ones authored or adapted by journal publishers. We observed shortcomings and omissions, as discussed in the results session, which may increase the probability of making mistakes like those introduced in Figures 3-8, as well as those pointed by Sweetland (1989). Most of the Medicine journal publishers in our samples adapted reference style guidelines to their particular purposes. However, even with these changes, the reference styles remained vague and imprecise in some respects and may result in errors in mentioning, quoting, and referencing cited works (RQ2). Similarly, although Social Sciences publishers generally did not recommend any adaptation to the existing reference styles adopted by their journals, often the guidelines for writing and formatting bibliographic references and in-text reference pointers related to mentions and quotations were not clear and easy to understand. For instance, while the most adopted reference styles focused on instructions for bibliographic references content and formatting, they generally did not provide guidelines for describing all the types of publications cited in the articles in our sample such as engravings and lithography (RQ1). These shortcomings, also mentioned by Sweetland (1989), often are the reason that denies a precise identification of the cited works by a reader. This situation seemed particularly relevant in Social Sciences articles, which cited many types of publications. Also, 5.9% of the Social Sciences journals did not adopt explicitly a reference style and, instead, recommended the authors to consult the bibliographic references of the journal sample issue and to consider them as a model for writing and formatting bibliographic references. However, the coverage of the types of cited works in such sample issue was not complete. In these cases, the author had no alternative unless writing the bibliographic reference according to his/her own beliefs on what could be the best way of describing the cited works. This may explain some cases in which required metadata for the identification of the cited work are actually missed (RQ2). Along the same lines, we observed that some of the reference styles provided by the publishers of the Medicine articles in our sample, which adopted a citation-sequence system, did not mention the rationale in which the bibliographic references should be sorted in the bibliographic reference lists, thus creating even more confusion for the author who wrote them. Also, we observed that four variations of the style of the numerical character of such bibliographic references adopted in citation-sequence system in Medicine articles, while for Social Science articles there were found only two variations. Regarding how quotations are marked up within the article content, we found cases within Social Science articles in which double and single quotation marks were adopted for run-in quotations, while long quotations were indicated using indentation which was, sometimes, accompanied by italicized characters and double quotation marks. Besides, we noticed that there were no shared rules used to dentify a quotation as run-in or long. When specified, the main strategies adopted in the reference styles of the journal in our sample were: 1. to classify quotations considering the total numbers of words quoted (usually, 80 words at maximum for run-in quotations); 2. to classify quotations according to the length of the quoted passage in the citing work (usually, 3 or 4 lines at maximum for run-in quotations). Considering the bibliographic references included in Medicine articles in our sample, the names of journals in which cited articles were published are provided in abridged formats, which are usually based on lists providing standardized abbreviations to journal names. In the meantime, we observed five different sources from where these abbreviations can be extracted. In principle, the same journal can be abridged in at least five different ways, which reinforce the claim by Sweetland (1989, p. 298) in which “differences in journal title a bbreviation have been commonly noted as a particular source of error” (RQ2). Figure 8 reproduces one of the bibliographic references of an article composing our sample (Uzunalli et al., the abbreviation “Ann” can be understood by th e reader as “Annals”, “Annual”, “Annalen”, “Annales”, and so on. As an additional source of ambiguities, the guidelines for authors provided in the journal’s webpage did not mention the source from which journal titles abbreviations should be based on.
Figure 8.
A bibliographic reference with an abridged representation of the cited journal. Thus, abridging journal titles within bibliographic references in Medicine articles seemed not to be fully accomplished. Thus, one possible strategy to adopt to disambiguate the journal would be to include compli mentary information within bibliographic references, e.g. the cited journal’s ISSN, to assure the uniform interpretations of journal titles abbreviations. Our analysis also revealed a high percentage of articles not including the source of non-textual content included in the articles, e.g. tables and figures, especially in Medicine articles. Although non-textual content is also citable, we observed that reference styles rarely provide guidelines on how to proceed with the presentation and description of this kind of content. Summarizing, all the information presented so far allowed us to conclude that the issues raised by
Sweetland’s’ study in 1989 are still valid today (RQ1). We do not need thousands of reference styles if we do not have clear guidelines on how to inject bibliographic metadata into bibliographic references. In this context, neither the standardization nor the technologies developed were able to fully support the creation and management of bibliographic metadata to write bibliographic references. Reference styles do not provide sufficient and clear information to authors regarding the procedures of citation, metadata description, and formatting. Looking at the articles in our sample, reference styles seemed to be a list of suggestions used by authors and publishers to support their own decisions regarding referencing and citing rather than a concise and precise instrument of guidance compliant with standardized behaviors on citing habits.
Fanning AS, Anderson JM. Zonula Occludens-1 and -2 Are Cytosolic Scaffolds That Regulate the Assembly of Cellular Junctions. Ann N Y Acad Sci. 2009; 1165:113 –
20. https://doi.org/10.1111/j.1749-6632.2009.04440.x. [PubMed] s anticipated by Galvão (1998), our analysis confirmed that the terminology used to refer to bibliographical concepts was ambiguous. Terms like “reference styles” and “citation styles”, and “bibliography”, “bibliographic references”, “references”, and “notes” referred to the same thing on different occasions, thus confirming Sweetland’s statements regarding the “lack of uniformity in the literature” and “of re al agreements among journals and authors” (Sweetland, 1989, p. 298).
The rate of errors in citations in respected scientific journals in Medicine and Social Sciences is high (RQ2). Sweetland (1989) states that there is little consensus about who should be responsible for correcting citations: publishers may think it is up to the authors while authors would like to have referees and editors to double-check them. The author complements: “no one, except perhaps librarians, seems to care very much about the problem. […]
We spend considerable time and effort in training catalogers in both the theory and the methods of descriptive cataloging. It would be good to spend at least some effort on training all information workers [including authors and publishers] in the theory and methods of the citation.” (Sweetland, 1989, pp. 301 -302)
Conceptual representation of references and citations
Toward the end of the 20th Century, librarians began discussing possible changes in the description, access, and encoding of bibliographic information. Typically, especially in bibliographic catalogs which were based on the Anglo American Cataloging Rules (AACR), documents were described out of context and their descriptions usually referred to a particular edition published by a specific publisher. In modern times, this approach has been no longer sustainable and compliant with the fulfillment of the functions of bibliographic catalogs in the new information scenario (Joudrey et al. , 2015; Tillet, 2003b, OCLC, c2020). Between 1992-1995, the IFLA Study Group on Functional Requirements for Bibliographic Records (FRBR) developed an entity-relationship based on a conceptual model for describing bibliographic records for all types of materials. This conceptual model should not only consider the function of the catalogs, which should enable users to find, identify, select, obtain bibliographic resources and navigate within the catalog, but also allow the performance of user tasks associated with bibliographic resources and the conceptual model of the bibliographic universe: the entities, their relationships, and attributes (Tillet, 2003a, 2003b; IFLA, 2009). FRBR determines that items (embodiment of works in physical or electronic publications) must be described in a context in a manner sufficient to relate the item to the other items comprising the work using the four-level bibliographic structure: FRBR Work, FRBR Expression, FRBR Manifestation and, FRBR Item (OCLC, c2020). Even adopting new perspectives, the essence of descriptive representation in bibliographic catalogs still focus on describing a publication as stored in a particular information support, be it physical or electronic. Indeed, despite all these new aspects introduced by FRBR, cataloging, and the preparation of bibliographic references, according to IFLA Study Group on the Functional Requirements for Bibliographic Records (2009), should aim at: a) ident ifying all the background information that have supported an author’s ideas conception, or the information which reading is being recommended by him as a complimentary content of his ideas; b) using the information contained in the bibliographic reference (e.g. an author, the title, or a journal) to select an entity that is appropriate to one’s needs; ) finding the entities described. Since the new approach of descriptive representation materialized in the FRBR guidelines considers document content (FRBR Expression) rather than the format in which it is embodied (i.e. FRBR Manifestation or, more concretely, FRBR Item), the relationship among mentions, quotations, bibliographic references, and in-text reference pointers seems to collide to some degree. Quotations are excerpts of a textual content available in the cited works (FRBR Expressions), while mentions are the author’s written interpretations of a textual content of the cited works (again FRBR Expressions) that should convey somehow the original idea of the cited author (FRBR Work). Instead, bibliographic references usually focus on one particular embodiment (FRBR Manifestation) which could be available in a single and physical (or electronic) exemplar of a publication (FRBR Item), e.g. the copy of a PDF file. In this context, it should be considered that a FRBR Expression can be (and actually is) published in different information supports and formats. Since most of the scientific production is available on the Internet (sometimes free of charge) and since this content can assume different FRBR Manifestations, referring to specific FRBR Manifestations within bibliographic references may restrict the possibilities of access for the reader, if he does not have the perception (reading the bibliographic reference) that certain content to which he does not have access to (e.g. in the publisher website) may be available in a different FRBR Manifestation (e.g. in a preprint/postprint repository). Besides that, in-text citations usually refer to the content of a bibliographic resource, that is, the FRBR Expression of a FRBR Work, while the respective bibliographic references usually refer to its embodiment (FRBR Manifestation). This richness represents also an obstacle to facilitating and improving the ways of accessing information, which also corresponds to one of the purposes of descriptive representation. In our sample of Medicine and Social Sciences articles we found that the guidelines above were not always followed (RQ3). In particular, the descriptive elements provided in the main part of in-text reference pointers that accompany quotations pointed to the FRBR Manifestation layer of the cited work due to the presence of the pages where the quoted text is contained. However, a smaller portion of such in-text reference pointers did not provide any descriptive element for referring to the FRBR Manifestation layer and actually seemed to relate to the pure content of the cited works (i.e. their FRBR Expressions). Besides, we found that the main part of bibliographic references referred to a particular FRBR Manifestation of the cited works. However, there were a few cases where it was not possible to identify even the FRBR Expression of the work defined by a bibliographic reference. For instance, the bibliographic referenc e “World’s Work. 1909. "The March of Events.". World’s Work, December.” found in one of the articles analyzed did not provide enough descriptive information to classify it properly. Usually, the reference styles of the journals considered in our analysis encourage authors to provide page numbers in the in-text reference pointers referring to quotations, which can be used to locate the excerpt within the cited work if we strictly consider the particular FRBR Manifestation of the cited article referenced by the particular bibliographic reference. Although considering a broader context where each cited work can be fully characterized according to FRBR, that scenario can be restrictive, since does not allow one to consider the possible various kinds of print-like embodiments (printed within a volume, PDF in the publisher website, PDF in an institutional repository, etc.) the work may assume. In addition to that, we found there are no clear guidelines when the cited works cannot be aginated, such as in HTML versions of articles, speeches, digital media content, and tridimensional objects. The evolution that has characterized the universe of information sciences may have brought some additional challenges in specific processes, such as the bibliographic normalization activity. In this work, bibliographic normalization was addressed as a facet of descriptive representation. Therefore, it has been inevitable to consider the evolutions and improvements that have been carried out in the cataloging domain, including the introduction of FRBR. Conceptual changes in cataloging should potentially impact the way bibliographic metadata is written and managed, the relations between in-text reference pointers, bibliographic references and the cited works, and bibliographic catalogs. Cataloging description level concept under FRBR, which corresponded to the descriptive comprehensiveness, needs to consider different aspects of the same work, which may switch the interpretation of the metadata available in the catalog. This is something to be considered in bibliographic normalization activities, to understand and, eventually, foresee how these changes can impact the way bibliographic references should be presented (RQ4).
Conclusion
This study was fully based on the analysis of bibliographic metadata of a subset of articles of Medicine and Social Sciences subject areas. While the conclusions we reach cannot be considered final and of broad coverage, considering we included only two scholarly disciplines in the study, our findings suggest that reference styles do not fully accomplish with their role of guiding authors and publishers on providing concise and well-structured bibliographic metadata within bibliographic references to allow the easy and accurate identification of the referenced works. In particular, our study confirmed that several of the issues highlighted by Sweetland (1989) still hold today. The main explanations behind the issues that we have identified in our study are: (a) reference styles do not provide clear instructions on how to write bibliographic metadata description expressed via bibliographic references, in particular when different types of work, in addition to articles and books, are considered; (b) the training of authors in bibliographic reference writing and in-text mentions and quotations of passages of a cited works seemed poor; (c) in some cases, the importance of having clear and clean citation apparatus seemed not being a primary concern by both the authors and the publishers; (d) journal editors did not seem to spend a huge effort on revising bibliographic metadata within their articles. Also, the adoption of FRBR in bibliographic catalogs may affect the way one stores bibliographic records since each catalog could follow its own approaches to describe a certain document (e.g., only the FRBR Manifestation level of the first edition) which may differ from the others. Taking into consideration these issues in descriptive representation activities is important since the concept of bibliographic metadata description and management is under revision within Librarianship. Facing the need for considering a broader and more accurate view of this subject, we are currently running a new study to complement the insights presented here by considering other subject areas and may encourage other related studies in metadata management to the benefit of the scholarly domain as a whole. eferences
ABNT. (2018),
ABNT NBR 6023 - informação e documentação - referências: elaboração.
ABNT, Rio de Janeiro.
Baptista, D.M. (2007), “O impacto dos metadados na representação descritiva”,
Revista ACB: Biblioteconomia em Santa Catarina,
Vol. 12, No. 2, pp. 177-190, available at: https://revista.acbsc.org.br/racb/article/download/529/663 (accessed 17 August 2020)
Barbeau, E. (2018), “Why thousands of citation styles exist”, available at:
BibMe (2017), “Why are there so many citation styles?”, available at: , P. (1987), “Power and centrality: a family of measures”,
American Journal of Sociology,
Vol. 92, No. 5, pp. 1170 – Social Science Information Studies,
Vol 2 No. 2, pp 65-77, available at: https://doi.org/10.1016/0143-6236(82)90001-1 (accessed 17 August 2020) Cunha, M.B. da. and Cavalcanti, C.R. de O. (2008),
Dicionário de Biblioteconomia e Arquivologia,
Briquet de Lemos, Brasília, DF.
Galvão, M.C.B. (1998), “Construção de conceitos no campo da ciência da informação”,
Ciência da Informação,
Vol. 27 No. 1, pp. 46-52, available at: https://doi.org/10.1590/S0100-19651998000100006 (accessed 17 August 2020)
Gratz, A. (2016), “Why are there so many different citation styles?”, available at: https://libraries.mercer.edu/research-tools-help/citation-tools-help/DifferentCitationStylesHandout.pdf (accessed 17 August 2020) Guerrero-Bote, V.P. and Moya-
Anegon F. (2012), “A further step forward in measuring journals’ scientific prestige: the SJR2 indicator”,
Journal of Informetrics,
Vol. 6, pp. 674 – ICMJE (c2020), “Preparing a manuscript for submission to a medical journal”, available at: ords (2009), “Functional Requirements for Bibliographic Records”, available at:
FLA. (2009) “Statement of International Cataloguing Principles”, available at:
ISO 690 - information and documentation - guidelines for bibliographic references and citations to information resources,
ISO, Geneva. oudrey, D.N. et al. (2015),
Introduction to cataloging and classification,
Indexação e resumos, et al. (2011). “Reflexões acerca das relações entre representação temática e descritiva”,
Informação & Sociedade - Estudos,
Vol. 21, No.1, pp. 27-35 available at: https://periodicos.ufpb.br/ojs2/index.php/ies/article/view/7367/5596 (accessed 17 August 2020)
O’Connor, M. (1977). “Style manuals and guides for authors an d editors: prescriptive or descriptive?”, in Balaban, M. (Ed.),
Scientific information transfer: the editor's role: proceedings of the first International Conference of Scientific Editors, April 24-29, 1977, Jerusalem,
D. Reidel Publishing Company, Dordrecht. pp. 287-290, available at: https://link.springer.com/content/pdf/10.1007%2F978-94-009-9863-6.pdf (accessed 17 August 2020)
OCLC. (c2020). “OCLC Research activities and IFLA's Functional Requirements for Bibliographic
Records”, available at: et al. (2018), “The state of OA: A large-scale analysis of the prevalence and impact of
Open Access articles”,
PeerJ , Vol. , pp. e4375, available at: https://doi.org/10.7717/peerj.4375 (accessed 17 August 2020) Santos, E.A., et al. (2020a), “Glossary of the article "Citing and referencing habits in Medicine and Social Sciences journals in 2019"”, Version 1.0, Zenodo , available at: https://doi.org/10.5281/zenodo.3996578 (accessed 17 August 2020) Santos, E.A., et al. (2020b), “Raw and aggregated data for the study introduced in the article "Citing and referencing habits in Medicine and Social Sciences journals in 2019"”, Version 1.0,
Zenodo , available at: https://doi.org/10.5281/zenodo.3996582 (accessed 17 August 2020) Santos, E.A., et al. (2020c), “
Workflow for retrieving all the data of the analysis introduced in the article “ Citing and referencing habits in Medicine and Social Sciences journals in 2019 ” ", available at: https://dx.doi.org/10.17504/protocols.io.bbifikbn (accessed 26 August 2020) SCImago (c2020), “SJR - SCImago Journal & Country Rank”, available at:
Sweetland, J.H. (1989). “Errors in bibliographic citations: a continuing problem”,
The library quarterly,
Vol. 59, No. 4, pp. 291-304, available at: https://doi.org/10.1086/602160 (accessed 17 August 2020)
Tillet, B. (2003a), “The FRBR Model (Functional Requirements for Bibliographic Records)”, paper
Tillet, B. (2003b), “What is FRBR? a conceptual model for the bibliographic universe”, Library of
CM Writing Center (2016), “Why are there so many citation styles?”, available at: http://ucmwriting.weebly.com/blog/why-are-there-so-many-citation-styles et al. , (2019) Dynamic transition of the blood-brain barrier in the development of non-small cell lung cancer brain metastases,