The Top 100 Cited Discourse Studies : An Update

s. Abstracts serve as a stand-alone source of information and may draw readers’ attention to the articles they are based on. ‘Research article abstracts are the most effective means of sharing research results.’ (Tankó, 2016). The abstracts in the top 100 articles average 153.6 words, with the shortest abstract containing 62 words, and the longest numbering 275 words. No abstract includes any sub-titles. Most are one-paragraph pieces (98 out of 100). The research aims in the abstracts are stated via set phrases which contain ‘aim’ – both noun or verb – in 9 abstracts; ‘to analyse’ (12 abstracts); ‘to focus’ (11 abstracts); ‘to consider’ (7 abstracts); ‘to examine’ (29 abstracts); to argue (16 abstracts); ‘to explore’ (25 abstracts); ‘purpose’ and ‘hypothesis’ (1 abstract each). On the whole, the top 100 article abstracts tend to follow the move structure applicable in applied linguistics (Pho, 2008) and embracing: (1) Situating the Research; (2) Presenting the Research; (3) Methodology; (4) Results; (5) Discussing the Research. The move ‘Presenting the Research’ (No.2 in the above list) is often enforced by ‘Filling in the Knowledge Gap’. Partly, methodology and discussion of the results were omitted in the abstracts to the empirical articles. The above moves are relevant to empirical or research articles, with theoretical articles sticking to a simpler structure, where methodology is rarely included. As for results and their discussion, these sections also aim to convey as a contribution to a knowledge field. The section titles below are headlined according to IMRAD simplified version (Introduction; Methods; Results and Discussion). Section ‘Introduction’. As ‘the introduction is of prime importance in grabbing the reader’s attention’(Ecarnot et al., 2015), we thoroughly compared the introductions to the top 100 articles with the canonical introduction schema for original theoretical and IMRAD-structured (empirical) papers. It implies that the relevance of the research is followed by a field or literature outlining the gap to be filled by the present research; the next step is the research aim (or in addition, hypothesis/ research questions may be included); on top of all, (for a theoretical paper only) a general description of the main body of the articles displays authors’ logic at the end of the introduction. As most articles under discussion (88 out of 100) do not follow the standard IMRAD format, their introductions are often shorter with additional sub-titles for the steps which tend to be included into introductions to scholarly articles at large. The following themed titles and sub-titles exemplify a kind of ‘extensions’ to the introductions: • Literature Review (Ross & Rivers, 2017; Wu, 2018; Du-Babcock & Chan, 2018; Carden, 2018); • Review of literature on climate change discourses in the context of mobilization (Molek-Kozakowska, 2018); • Aims; Literature Review (Marlow, 2015);


Introduction
Discourse studies actively started in the 20th century.The term 'discourse' appeared in 36,084 document titles indexed in Scopus (as of March 1, 2019), with the earliest of them published in 1838.On average, nearly 2,500-2,800 titles, incorporating the term, enter the Scopus database annually.The discourse field coverage ranges from humanities and social sciences to dentistry and geosciences.As discourses occur only in social and cultural settings, much of the focus in research is given to the relationships between discourse and community.
Most of discourse studies turn to discourse analysis or based on it, as it examines patterns of language within diverse frameworks and settings.The concept of discourse is studied across the fields and approaches.Many articles and books dwell upon such issues as pragmatics in discourse, discourse and conversation, discourse grammar.Extensive methods and approaches evolve and are applied in the field to cover corpus approaches, multimodal discourse analysis, and critical discourse analysis, etc.
As discourse studies lie within the JLE scope, the JLE editors see the recent trends and most promising themes in this area as a benchmark for the journal authors.The review of the 100 most quoted articles published between 2015 and 2019 might extend the guidelines for authors, informing them of the trends.The goal of the review is twofold.In analysing the 100 articles, the review authors also set out to fix the prevailing structures in research Raitskaya, L., & Tikhonova, E. (2019).The Top 100 Cited Discourse Studies: An Update.Journal of Language andEducation, 5(1), 4-15. doi: 10.17323/2411-7390-2019-5-1-4-15

Materials and Methods
The editorial review thoroughly considers the 100 most quoted articles on discourse indexed with Scopus database for the period 2015-2019.The top 100 list was made up according to the criteria below.

Materials
The original search for 'discourse' documents in Scopus database brought 173,153 titles which were filtered to be limited to: subject areas 'social sciences ' and 'arts and humanities'; years 2015-2019; document types 'article' and 'review'; language 'English'; source type 'journal'; keywords 'discourse', 'semantics', 'discourse analysis', and 'critical discourse analysis'.The limitations decreased the number of the documents to 3,452.
Then the documents were filtered and limited to journals that have 'discourse', 'linguistics', 'lingua', and 'semantics' in their titles and 'linguistics and language' as their subject area.The selected documents ultimately decreased to 207.They were published in the following journals: Critical Discourse Studies, Discourse and Communication, Discourse, Discourse Context and Media, Discourse Studies, Journal of Pragmatics, Journal of Multicultural Discourses, Lingua, Linguistics, International Review of Pragmatics, Pragmatics and Society.The top 100 out of 207 by the highest citation made up the final list for further analysis.
The title 'What is policy?21 years later: reflections on the possibilities of policy research' (Ball, 2015) brought the highest citation of 51 (as of March 1, 2019).Sixteen articles with 1 citation each bottomed the list.

Procedure
The top 100 most cited documents on discourse studies were sifted to outline the most popular themes in the research area.Both authors broke down the 100 articles into logically set directions, or 'themes'.The results were compared to sum up the key directions in discourse studies.The differences between the authors regarding the thematic breakdown extended to 9 articles and were settled by mutual consent.
The 100 abstracts were analysed to determine their word count and wordings of the aim and findings.The average word count was calculated.
Then, the structure of the 100 articles was studied by canonical schemas (moves and steps) to identify the most common patterns for discourse studies.
LILIA RAITSKAYA, ELENA TIKHONOVA In addition, the types of articles (original article -either empirical/analytical or theoretical; opinion and discussion articles; editorial) were fixed among the most quoted articles, judging by major features of each type.

Results and Discussion
The top 100 articles on discourse were published between 2015 and 2019, with 35 quoted articles in 2015; another 27 and 22 were cited in 2016 and 2017 respectively; 14 quoted articles came out in 2018; only 2 were published and quoted in 2019.The most cited article scores 51 citations (Ball, 2015).The lowest citation is one per article, with 16 documents out of the 100.
The articles came out in seven scholarly journals (28 articles in Critical Discourse Studies; 26 in Discourse; 26 in Discourse and Communication; 12 in Discourse Context and Media; 4 in Lingua; 3 in Linguistics; 1 article in Journal of Multicultural Discourses).Out of the 100 documents, 94 are qualified as original articles by type including 15 theoretical and 79 empirical/ analytical publications.The remaining six cover one editorial serving as an introduction to a special issue and five articles (opinion or discussion articles).
The affiliations of the authors encompass 102 universities and organisations, with Tel Aviv University, Universidade de Macau, and University of Queensland totalling three authors each and another seven universities having two authors each.The geographical breakdown includes 29 countries; with 20 documents coming from the UK, 18 articles from Australia, 16 documents from the USA; China and Sweden gave 8 documents each.Fitzgerald (2015Fitzgerald ( , 2016Fitzgerald ( , 2018)), Talib (2015Talib ( , 2016Talib ( , 2018)), Molek-Kozakowska (2017, 2018), and Schnurr (2015, 2016) authored more than one article.The rest of 156 authors had one article co-authored or on their own.The mean authorship per article is 1.6.

Basic Directions and Methods in Discourse Studies
The 100 most quoted titles were sifted to find out what the major research directions were and what methods were implied.The niches produced most research are discourses in education (23 papers) and press/ media coverage of various events or phenomena (23 articles).
Educational discourses are approached in diverse aspects.Ledin and Manchin (2015) carried out a multimodal study of management language in Swedish universities.Kolleck (2017) examines how changes in social fields shape the concept of education triggered by normative and semantic shifts.The article 'Inequality as meritocracy' is sub-titled 'The use of the metaphor of diversity and the value of inequality within Singapore's meritocratic education system' (Talib & Fitzgerald, 2015) and examines the ways the metaphor brings about a moral background for inequality within an educational system.
Several papers focus on school discourses like e-safety policy discourse related to widespread school Internet access (Hope, 2015); a critical discourse analysis of school bullying (Horton, 2016); analyses of aspiration discourses in a British secondary school (Spohrer, 2016); neoliberal common sense and race as neutral discourses in school policing (Nolan, 2015); discourses of the good parent in attributing school success (Thomas, Keogh & Hay, 2015); a discourse analysis of teachers' body talk (Lester & Gabriel, 2017).
News coverage discourse analyses cover analyses of news discourses on masculinity (2); Israeli media coverage of public examinations (Yemini & Gordon, 2017); British television coverage of the Barclays case in 2012 (Thomas, 2016); a crucial role of the right-wing Brazilian media in the impeachment of President Dilma Rousseff (Van Dijk, 2017); uses of 'you' in Guardian editorials (Breeze, 2015); historical analogies in the coverage of the Israeli-Palestinian conflict (Bourdon, 2015); inquiry in the media context of TV cooking shows (Matwick & Matwick, 2015); empiricist discourse in the talk of broadcast journalists (Reardon, 2018) and others.
Discourses of economics /policies and Internet brought 8 and 9 publications respectively.Few publications THE TOP 100 CITED DISCOURSE STUDIES: AN UPDATE came out to analyse discourses of ecology (3 papers), metaphors (4 articles), discourses of racism (3 articles), religion (3 papers).The rest of articles aimed to cover various themes: semantics and pragmatics of marking temporal progression in an Australian language (Ritz & Schultze-Berndt, 2015); discursive formation of nuclear proliferation (MacDonald et al., 2015); a critical discourse analysis of Swedish public health campaigns (Törrönen & Tryggvesson, 2015); a decision-making discourse analysis (Du-Babcock & Chan, 2018); relevance of materialistic-semiotic approach to discourse analysis (Borrelli, 2018); effects of dominant scholarly publication practices on papers produced by authors worldwide (Larson, 2018); relevance in different types of dialogical contexts, pursuing both cognitive and non-cognitive goals (Macagno, 2018).
The prevailing methods in the articles covered critical discourse analysis (CDA) (over 90 %), corpus-based analysis (10 %).In addition to CDA, the remaining studies applied multi-modal discourse analysis, circumbounded analysis, macro-sociological analysis, social semiotic analysis, case studies.

Prevailing Structures in the Articles on Discourse
Abstracts.Abstracts serve as a stand-alone source of information and may draw readers' attention to the articles they are based on.'Research article abstracts are the most effective means of sharing research results.' (Tankó, 2016).
The abstracts in the top 100 articles average 153.6 words, with the shortest abstract containing 62 words, and the longest numbering 275 words.No abstract includes any sub-titles.Most are one-paragraph pieces (98 out of 100).
On the whole, the top 100 article abstracts tend to follow the move structure applicable in applied linguistics (Pho, 2008)  Partly, methodology and discussion of the results were omitted in the abstracts to the empirical articles.The above moves are relevant to empirical or research articles, with theoretical articles sticking to a simpler structure, where methodology is rarely included.As for results and their discussion, these sections also aim to convey as a contribution to a knowledge field.The section titles below are headlined according to IMRAD simplified version (Introduction; Methods; Results and Discussion).
Section 'Introduction'.As 'the introduction is of prime importance in grabbing the reader's attention' (Ecarnot et al., 2015), we thoroughly compared the introductions to the top 100 articles with the canonical introduction schema for original theoretical and IMRAD-structured (empirical) papers.It implies that the relevance of the research is followed by a field or literature outlining the gap to be filled by the present research; the next step is the research aim (or in addition, hypothesis/ research questions may be included); on top of all, (for a theoretical paper only) a general description of the main body of the articles displays authors' logic at the end of the introduction.As most articles under discussion (88 out of 100) do not follow the standard IMRAD format, their introductions are often shorter with additional sub-titles for the steps which tend to be included into introductions to scholarly articles at large.
'Introduction' as a set-apart and headlined part is missing from a few articles, with one being replaced with a themed title (Corbett & Forsey, 2017), other two starting without any (sub-)titles (Collin, 2016;Larson, 2018).The latter may be caused by APA style.Whitton and Maclure (2017) starts their article with the headline 'Video games in education: a brief overview'; Noy (2017) introduces the research with 'Morality, museums, and the public sphere'; Szabo (2016) begins her article with a quotation followed by untitled introduction; Borrelli (2018) introduces the article in the section entitled 'Structure and purposes of the study'.Talib and Fitzgerald (2018) put a question to focus on the topic -'Critical Discourse Analysis?
Section 'Methods'.Twenty-five articles include the section headlined 'Method(s)', whereas 18 have 'Methodology'.Methods are also described in other sections, often combined with other structural units, e. Section 'Results and Discussion'.As the sample includes only 12 nearly IMRAD-structured articles (IMRAD structure in some of them is revised or updated), the section 'Results' is often replaced with 'Findings': 3 and 8 respectively.28 articles incorporate 'Discussion' section.Three publications have 'Findings and Discussion'.In 4 articles, 'Discussion and Conclusions' combine two sections.There are a few research-tailored sections headlined 'Comparison of Findings', 'Research Findings', and 'Results and Illustration'.
As most empirical research articles in the sample are analysis-based, they followed the patterns best suited to the critical discourse analysis.Most articles include from 2 to 5 themed sub-titles covering the logic of the analysis.They range from background and theoretical issues to specially formed corpuses and databases.
Section 'Conclusion'.The move to conclude the article implies remunerating the main findings and their contribution.53 articles end with 'Conclusion', 10 publications have 'Conclusions'.The section is missing from 14 articles where the section 'Discussion' or other themed sub-titles contain concluding statements.

Conclusion
The top 100 articles on discourse analysis showed that the original empirical or analytical articles accounted for 79 percent of the sample, with the theoretical articles amounting to 15 percent.Structurally, the empirical papers in the sample tended to multiple move variations.IMRAD with minor alterations was chosen only in 12 articles.The other articles in this group, as well as the bulk of the theoretical and opinion articles stick to research-tailored structures.
Though the sample covers the most quoted publications on discourse studies, not all findings relating to rhetorical moves can be considered as relevant to the JLE stance on article structuring.Occasionally, deviated or free structuring may be based on solid rhetorical foundations.But this approach may get elusive and distract readers' attention by irregular and unusual moves and steps.Established genre schemas make a research text accessible, transparent, and reader-friendly.
The JLE Editors opt for genre structures that best outline all research components and display their strong points together with research findings and contribution.As JLE is an international journal, its formats must fit in the global standards and best practices.Our authors represent dozens of countries, so IMRAD and other traditional schemas help them follow similar patterns and align their submissions with the benchmarks in the field.
Discussing the Research.The move 'Presenting the Research' (No.2 in the above list) is often enforced by 'Filling in the Knowledge Gap'.