Contrastive Analysis of Results and Discussion Sections in Medicine and Education Research Articles

In the last twenty years, the study of different text types in the light of genre-based analysis has become a central issue for linguists and English language teachers.  This has been partly due to the dominant role of English as the language of international research literature and to the ¨North-South imbalance in the world¨ (Swales, 1987, p.43) by which nonnative speaker academicians from underdeveloped countries have not been able to actively participate in their discourse communities at an international level.  Given these circumstances, many recent studies have focused on the analysis of the structure and linguistic features of the Research Article (RA).

  Even though most journals from diverse scientific fields have adopted the Introduction, Methods,Results and Discussion ( IMRAD) format for structuring their RAs, it is noteworthy that ¨scholarly discourse is not uniform and monolithic…. It is an outcome of a multitude of practices and strategies, where what counts as convincing argument and appropriate tone is carefully managed for a particular audience.¨ (Hyland, 2004, p.3).  Consequently, numerous attempts have been made to analyse and compare the sections of RAs within specific fields with the aim of helping nonnative speaker scholars better understand the specificities of academic discourse.  However, rather less attention has been paid to the potential of comparatively analyzing sections of RAs from different fields as a tool for studying written discourse for academic publishing ends.   Great gap!

The aim of this paper is to explore the possible variations in the Results, Discussion and Conclusion sections between two RAs in the fields of medicine and education:  Bhatt et al. (2014) and Crossley and McNamara (2013).  This paper seeks to enrich the understanding of similarities and differences in these sections pursuant to disciplinary variations.  Specifically, we analyze the sections based on the genre analysis models proposed by Swales ( 1990) and Swales and Feak (2004).  The plan of this paper is as follows: first, there is an analysis of the Results Sections of the two RAs; next, a comparison of the Discussion and Conclusion Sections is provided; and finally, some conclusions are offered. 

            As far as the analysis of the Results Section (RS) is concerned, Swales (1998) states that the first paragraphs are used to introduce the general results of the researcher's findings, tables and figures. Regarding the medicine RA, the title of this section appears in bold with capital letters on the left margin. Not only did the authors  separate the RS from the rest of their work, but also they subdivided it in two subtitles: ¨Door-to-Needle Times¨ and ¨Clinical Outcome¨, where they expose their main findings and mainly use past simple and active voice to describe the procedure of their study.

Although it can be assumed that Bhatt et al. (2014) adhered to the style conventions set by the American Medical Association Manual (2007) in order to get their RA published, it should be pointed out that the manner in which results are displayed seems to mirror the majority of the rules established by the American Psychological Association (APA) (2007).  However, if seen from the perspective of the latter conventions, there are some inconsistencies: the titles are not italicized and  each table does not begin on a separate page. The authors use two types of figures: a bar and a line graph. They include the words ¨Figure¨ and the corresponding number but they are neither in italics nor double spaced, as required by APA (2007). 

In the education RA, the RS is separated from the rest of the article and the title appears in bold with capital letters on the left margin. Another difference that can be noticed when comparing the two RAs is that the education one does include tables but not figures.  The most striking difference with the medical RA, however, is that the tables are not only present in the RS, but also in other parts of the paper. Thus, it can be assumed that this RA does not comply with the APA (2007) rules as these conventions establish that the tables and figures should be written in the RS and that each table and figure must be separated on different pages. In this case, the section is divided in several subsections and they all contain tables exposing the results.

            Even though genre analysts do not distinguish between the Discussion Section (DS) and the Conclusion Section (CS) , it should be noted that the two RAs do establish these separate sections.  Such distinction ¨ is partly conventional, depending on traditions in particular fields and journals¨ (Swales & Feak, 2004, p.268).  According to these authors, there are three discussion moves:  move 1, by which RA writers consolidate their research space; move 2, in which the limitations of the study may be indicated; and move 3, in which areas of further research or courses of action are recommended. 

Regarding Bhatt et al.’s (2014) RA, it is noteworthy that while moves 1 and 2 are developed extensively (there are six paragraphs devoted to summarizing results and stating conclusions on the clinical benefits of rapid administration of intravenous tissue plasminogen activator and one paragraph under the section ¨Limitations¨), move 3 is absent.  Crossley and McNamara (2013), in turn, fully develop their findings and claims in fourteen paragraphs (move 1) and explain the need for further studies in the last paragraph of the conclusion section (move 3).  However, references to the limitations of their study (move 2) are scarce in the latter RA. 

Because ¨the structure of the discussion section is closely correlated to both the number and kind of research questions posed in the introduction sections of the paper¨ (Belanger, 1982, as cited by Swales, 1990, p. 171), DSs tend to present cycles through which each research question is passed.   Swales (1990) identifies the following eight-move scheme: background information, statement of results, (un)expected outcome, reference to previous research, explanation, exemplification, deduction and hypothesis, and recommendation. 

In relation to this, Crossley and McNamara (2013) begin the DS by generalizing their results and highlighting their implications in the first paragraph and then dedicate all the other paragraphs to discuss their findings from the angle of the different, specific variables considered during their study (each of them introduced by their respective subheading).  Although some of these subsections are succinct, others are more developed and, thus, evidence some of Swales’ (1990) cycle moves.  For instance, under the subheading ¨Speaking Proficiency¨,  there are two paragraphs: In the first paragraph, some theoretical information is provided in order to define speaking proficiency (move 1: background information).   The second paragraph begins with a statement of results (move 2) when it is stated that ¨Our analysis focuses solely on language organization and demonstrates that …¨ (Crossley and McNamara, 2013, p. 187).  Having said that, the authors go on by making reference to previous research with the aim of providing support to their claim (move 4) : ¨An important component of our study is that the tested features adhere to the relations hypothesized … (i.e., speaking proficiency or communicative competence; Shin, 2005)¨ (p. 187).  Immediately after this, Crossley and McNamara (2013) close the paragraph by making a claim about the generalizability of some of the reported results (move 7: deduction and hypothesis) when they state that: ¨Given this, we have confidence that our models have not only predictive validity, but also face validity.¨ (p. 187).

Bhatt et al.’s (2014) RA, on the contrary,  does not display as many cycles as the article on education.  The reason for this may be that Bhatt et al. (2014) posed a single research question in the introduction related to the administration of a drug within a specific time frame after hospital arrival.  Consequently,  the DS is more linear and refers to the authors’ results (move 2), though references to previous initiatives and guideline recommendations are quoted in this section (move 4). 

The analysis carried out in this paper has revealed that even though there are some similarities in the generic structure of the RSs, DSs and CSs of the two RAs, many differences surfaced when examining them in detail.  As it was explained early in this paper, some differences found might have been related to the different style conventions followed by the journals in which these RAs were published.  Another significant finding was that the DS and CS moves of the two RAs did not completely correspond to the genre analysis models proposed by Swales and Feak (2004).  However, the fact that so many traces of such models could be identified in two RAs which belong to very different disciplinary communities confirmed the claim that Swales' (1990) and Swales and Feak’s (2004) models of analysis contribute to a better understanding of how information and knowledge are structured in RAs.

  Because this paper has compared only two RAs from different fields, the results of this analysis cannot be generalized to all the RAs in the medicine and education disciplines.  In spite of this, it should be possible to gain greater awareness on how different disciplines write by contrasting a larger corpus of RAs from the medicine and education fields in terms of Swales’(1990) and Swales and Feak's (2004) genre analysis models.

References 

American Psychological Association. (2007). Concise rules of APA style. Washington,DC: British Library Cataloguing-in-Publication Data.

Bhatt, D.L., Fonarow, G., Hernandez, A.F., Peterson, E., Reeves, M.J., Saver J.,  … Zhao, X. (2014). Door-to-needle times for tissue plasminogen activator administration and clinical outcomes in acute ischemic stroke before and after a quality improvement initiative. JAMA, 311(16),1632-1640.doi: 10.1001/JAMA.2014.3203

Crossley, S., & McNamara, D. (2013). Applications of text analysis tools for spoken response grading. Language Learning & Technology, 17(2),171-192. Retrieved from http://llt.msu.edu/issues/june2013/crossleymcnamara.pdf

Hyland, K. (2004). Disciplinary discourses: Social interactions in academic writing. Ann Arbor, MI: University of Michigan Press.

Swales, J.M. (1987). Utilizing the literatures in teaching the research paper. TESOL Quarterly, 21(1), 41-68. doi: 10.2307/3586354

Swales, J.M. (1990). Genre analysis: English in academic and research settings. (Cambridge Applied Linguistics Series). Cambridge, UK: Cambridge University Press.

Swales, J.M. (1998). Other floors, other voices: A textography of a small university building. Mahwah, NJ: Erlbaum

Swales, J. M. & Feak, C. B. (2004) Academic writing for graduate students: Essential tasks and skills (2nd ed.) Ann Arbor, MI: University of Michigan Press