W 108 M

Introduction

This paper aims at the description, analysis and evaluation of the structure of three randomly selected research papers that have been published in different journals. Such an analysis may provide useful for educational purposes, revealing mistakes and setting examples for structuring the information acquired by research and presenting it in a research paper. The underlying assumption is that any research paper has to meet certain criteria in order to communicate its message clearly and accurately. The expected result of analysing texts published in learned journals is that they adequately fulfill the formal and structural requirements of research papers.

Methods

The method of selecting three journals was partly random, as the choices were limited by the number of journals in English avaliable at the library I used. I wanted to use texts of different subject areas in order to allow an analysis of their differences, and I wanted to avoid technical or scientific areas, which require a special form of presentation (Berry, 1994 p.2) as well as texts which are not to be considered research papers -- for example, one that is a subjective opinion rather than a presentation of the findings of a research.

The selection of the articles was also only partly random, as I chose texts which I found interesting and were of appropriate length. A text which is too short would obviously fail to provide the needed raw material for analyses, while a very long article would not only exceed the scope of this paper, but would mean additional work without furthering the aims of the research. The selected articles are between 14 and 19 pages, including appendixes and references.

The method of analyses included the identification of the structural units, while evaluating the texts at the structural level was based on the assumption that "a research paper should be circular in argument" (Berry 1996 p. 99), that is, it should be introduced by stating what it discusses, discuss exactly what was stated and then draw conclusions that can be supported on the basis of the discussion. The title should clearly reflect the actual data the paper discusses (Berry 1996 p. 99). Along with the organisation of material, an important criterion is the appropriate ratio of its sections; one specifically under- or over-developed part causes the whole paper to be shapeless (Berry 1996 p. 102). The texts are never compared to the abstract notion of, say, everyday language use or journalistic style; instead, every comparison is made to the other articles. This way of comparison seemed to be the most convenient method as the texts deal with different subject areas that provide sufficiently diverse grounds for comparison, and it is a more direct method to make comparisons with a specific, relatively short text than with an undefined and unlimited set of discourse.

The nature of this research and the method of analyses implies that the data -- the examples from the text -- and their evaluation should not be presented in separate units, so the results and discussion are included in a single body of text for each of the analysed papers.

Results and Discussion

I.) Language Aptitude Testing: Unveiling the Mystery (Otto 1996)

The title is aimed at creating anticipation and interest, but it is not too informative: all the reader learns from it is that the paper is about language aptitude testing, which, being a topic far extending the limits of a research paper, needs further specification. The Introduction seems to specify the purpose clearly: it is to answer general questions about language aptitude and to describe a research aimed at developing such a test for Hungarian learners (p. 6). The problem with this specification is that the Conclusion names different goals that the paper has achieved: the description of the Hungarian language aptitude test (not the process of compiling it) and the theory on which the test is built (p. 17). Following this ambiguity, one might be anxious to see what is actually discussed in the paper.

Given the ambiguous frame of the paper, it might not be surprising for an experienced reader to find sections in the text which are not mentioned either in the Introduction or the Conclusion, while expected parts are left out. The longest section, about the utilization of the Hungarian language aptitude test (Menyet), takes up four and a half of the 11 pages of the main body of the paper (pp. 13-17). On the other hand, of the announced attempt to "describe my research to develop a language aptitude test" (p. 6) only a brief paragraph is presented (p. 11), from which the reader learns the following: the development, which was completed in 1996, and included the piloting of several hundred items, took three years and was supervised by Dornyei Zoltán of ELTE, Deptartment of English Applied Linguistics. The description of the Modern Language Aptitude Test (MLAT) (p. 9-10), the Conclusion (which says that the theory on which the Menyet is built on is described in the paper) and the description of the Menyet (which contains tasks that are similar to MLAT) (p. 12, and the appendix) indicates that the process of designing the test involved the adaptation of MLAT tasks to the needs of Hungarian learners, and the reader may assume that the test items accumulated by Sinkovics (p. 11) had also been used, but these are never stated. Assumptions and indications are not enough to fulfill the expectations the Introduction creates.

The skeleton outline of the paper consists of four parts: a long description of language aptitude testing in general and current language aptitude tests (pp. 6-11), two short paragraphs on language aptitude testing in Hungary and on the development of the MenyÈt (p. 11), a description of the Menyet, Utilisation. The parts are not out of balance; it is not the proportion of the sections that should be changed, but the Introduction and the Conclusion. The title, the introduction with the conclusion and the main text can be constructed as three parts of different size, but similar structure and content, as it will be shown in the next example.

II.) The effect of school-based arts instruction on attendance at museums and the performing arts (Kracman 1996)

This research paper is structured similarly to a pyramid, in which the different levels are identical in shape, colour and material and the only difference is in their size. The block at the top is the title, precisely stating the purpose of the research and prescribing the method followed in the course of its completion: data was collected on school-based arts instruction and attendance at museums and performing arts, the discussion established a link between the two and the conclusion identified a cause and effect relationship between them. Since the journal features an Abstract at the beginning of each article, there is an additional level; the Abstract is a summary of the paper, its structure is identical to that. The introduction does not only specify the exact source of data and the predicted results of its analyses, but it briefly summarises the previous findings on the relevance of museum and elite performing art attendance. This is necessary in order to be able to make predictions of the results.* The conclusion drawn is that the predicted effect of school-based arts instruction does exist and it is a positive one. This conclusion states nothing that is not supported by the assembled data, which is the lowest and widest level of the pyramid.

Compared with the previous text, this one is so clear that I felt no need to indicate the page numbers of the different sections (or levels of the pyramid); their place is self-explanatory after a single reading. Still, it can be said that identifying the structure of both articles was relatively easy, partly because they both have clearly separated and named sections. The case is very different when one reads a single body of text, only divided into paragraphs but without subsections and -titles, as the next example will show.

III.) Dante and the Modern Subject: Overcoming Anger in the Purgatorio (Tambling 1997)

The first questions I had to answer about this paper were whether it can be considered a research paper or it is too subjective (which literary studies are often accused of). Very briefly, I find that a paper which analyses a text and contrasts it with other texts, applying different theories (of literary theory, literary history, psychology, linguistics and philosophy) to interpret it, cannot be considered only imaginative writing.

The title of the paper does not precisely narrow down the area of research to a very specific field, but it introduces the keywords. Only gradually does the reader learn the purposes of the paper, there is not a section -- the first paragraph, for example -- which could be identified as the introduction. It does not mean, though, that there is none. The purposes are stated: "This paper concentrates on the first use of the word 'moderno' in Purgatorio" (p. 401), and: "The link between modernity and anger (...) which I want to pursue" (p. 402). The method is also specified: the paper applies "theory that applies to the modern subject" instead of a historical approach, with the predicted result of making the text "speak its modernity in ways unobtainable through an older critical theory" (p. 402).

Even though the organisation of the paper is far from obvious for the first sight (or for the first reading, for that matter), the circularity in argument, Berry's requirement is followed through. The discussion is centered on the keywords of the title, the initial aims are reached, and the conclusion summarises the relation of the tropes of modernity and anger in Purgatorio on the basis of the discussion.

Conclusion

The three papers seem to adequately fulfill the criteria of research papers, even though the misleading introduction and conclusion of the first one is to be considered a mistake, and the rather formless nature of the last one is likely to irritate the more accurate researchers. The different subject areas seem to cause many of the structural differences in the papers. The first one, being a text related to the field of methodology, uses the correct form (introduction, subsections) even though the Introduction and the Conclusion do not function properly. The second, a research in sociology, which features the processing of a considerable mass of numerical data and the checking of the validity of these data, is meticulously accurate. The paper on a specific approach to a piece of medieval literature is less concerned about form, but the structure is logically strict.

I assumed that analysing and comparing the structure of three different research papers which have been published in learned journals may provide helpful information for somebody who wants to compile a research paper. In fact I could apply the results of this research in the composition of my own paper: I tried to avoid the mistakes my research revealed and follow the example of those features of the papers that I found appropriate. Hopefully my findings can be helpful for others too in designing a well-structured research paper.

* The part I call introduction here is sections 1. and 2. ("Introduction" and "Defining high status culture: The meaning of 'cultural capital'"), because they form one logical unit.