Gabriella Jenei: Referential Cohesion in Academic Writing • A descriptive and exploratory theory- and corpus-based study of the text-organizing role of reference in written academic discourse
›
8. STAGE 4: THE REPRESENTATION OF ERRORS IN THE COHESIVE REFERENCE ANALYSIS OF STUDENT ACADEMIC WRITING
›
8.5. Results and Discussion

8.5. Results and Discussion

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In Section 8.3, the kinds of errors that were found in the two analyzed corpora were discussed. In the first part of this section, the two corpora (high- and low-rated) will be compared along with each type of error, and then error types will be discussed as they are distributed in the various MA thesis sections in the corpora.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

On average, theses in the high-rated corpus contained 39 errors each, while texts in the low-rated corpus contained 63 errors. As can be observed in Figure 18, in most cases (except for non-cohesive items), low-rated theses contained the same or higher number of errors than high-rated theses. Furthermore, in low-rated theses, twice as many items lacked a referent in the text. Additionally, about three times as many intended generic items were realized by a definite article and a plural noun phrase, also adding to the number of items without a referent.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Figure 18 Types of errors of reference in high and low-rated MA theses in percentages out of the total number of errors
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

These types of errors are rooted in a variety of sources, one of them being the pedagogical nature of the thesis genre, which resulted in over-dependence on the reader’s background knowledge and reliance on the reader’s (thesis supervisor’s) familiarity with the writers’ topic and research project. Students tend to assume that even if their texts are not perfectly clear, their peers and instructors just understand what they mean (Benson & Heidish, 1995). Examples of this from the high-rated corpus are the following items without a referent in the text: the other components of communicative competence, the tools, the intervention. Another typical problem was the long distance cataphoric reference to one of many groups of participants in the research project described in the thesis (e.g., the students, the participants – where it was impossible to identify which specific group was being described). In addition, thesis writers sometimes referred to the researchers without a specific antecedent, as if they were a specific group of people holding particular opinions.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

As regards vague items, differences between the two corpora were observed where items had two or more possible referents. In such cases (for example, two or more questionnaires, analyzed texts, groups of students or target languages) students often failed to make the reference to one of these groups clear. Items with a broad reference were equal in frequency in the two corpora; although it should be noted that there were slightly different items. In high-rated theses, the referent is not specific enough in: the previous sections (meaning: the referent is somewhere in the previous sections), the results, the theory, the findings, the tables. In low-rated theses, errors included the above-mentioned authors, the whole process (referring to some part of the research project), the aim.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Problems with lexical relationships, demonstratives, and agreement have already been discussed in the previous section. Figure 18 shows that they occur slightly more frequently in low-rated theses. Uninterpretable items, especially in low-rated MA theses included items reflecting the misinterpretation of academic style: it was decided to include (meaning: I decided to include) or it was intended to compare (meaning: I intended to compare). These items block the reading process, as it might be interpreted as referential, especially in the latter example. Other uninterpretable items included non-defining relative clauses that occurred in place of an intended defining clause, leaving the noun without specification: the topics, which occurred or the mistakes, which occur. Distant referents were relatively infrequent in both corpora. The category of non-cohesive errors was the only one where high-rated theses scored higher. This category includes all the error types, but it contains problems with sentence-internal cohesion. The majority of the errors were due to the lack of agreement between the referent and the referring item. Possessive of with a person as head (e.g., the opinion of Seliger and Shohamy; the number of them) was one of the most frequent sources of error.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

The distribution of errors in text parts differs considerably in the high- and low-rated thesis corpora. Figure 19 shows the distribution of errors in the five typical thesis sections (Introduction, Review of the Literature, Research Method, etc.) that were found in all the theses in the corpus, although the titles of these sections vary (e.g., Method, Methods, Research Methods, Research Design and Method). Figure 19 shows that for writers of high-rated theses the Review of the Literature is the most problematic. Most of the errors in both corpora were due to the lack of references to cohesive items in the textual context. Surprisingly, high-rated theses writers used more plurals with the definite article for generic reference and had more vague items. Vague items found in high-rated corpus (such as the theory, these perspectives, the purposes, the authors) show that writers of high-rated theses may have found it especially difficult to summarize the relevant theories and approaches in the literature in a way that their readers also see the connections between the concepts, theories and approaches that the writers found. This may be the result of the higher density of information in their review than in the low-rated corpus.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In the Method section the low-rated corpus contained almost three times as many errors as the high-rated corpus. The most frequent problems here included items with no reference, or vague reference (often between two or more possible referents); additionally, lexical problems and the choice of demonstratives were also problematic. Typically, the description of the analytical tools was vague, and it was difficult to identify which group of participants or which research tools were referred to by referring items. Additionally, research results were sometimes referred to quite randomly within the Methods section, without any clear explanation of what those results are and where they originate from.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Figure 19 Reference errors in MA thesis sections
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In the Results (and Discussion) sections in MA theses, Figure 19 shows that there are about twice as many errors in the low-rated corpus. In this thesis section, vague, non-interpretable items and items without a referent (including problems with generic uses) were the most frequent. Apparently, writers of high-rated theses were better at making clear the relationships between research tools, groups of participants and results easy to follow. Observable strategies for this in the high-rated corpus included consistency in terminology for research tools, providing labels for (groups of) participants and using these labels when describing the results of the research.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

The last section, Summary (and Conclusions), was again more problematic in low-rated theses. Most of the errors resulted from using the definite article and a plural noun for generic reference and also from a range of uninterpretable items (e.g., the reliability of the results, the first finding, the above mentioned without specific presupposed items).

8.5.1. Analytical problems and decisions

chevron_left
chevron_right
Tartalomjegyzék navigate_next
Keresés a kiadványban navigate_next
A kereséshez, kérjük, lépj be!
Könyvjelzőim navigate_next
A könyvjelzők használatához
be kell jelentkezned.
Jegyzeteim navigate_next
Jegyzetek létrehozásához
be kell jelentkezned.
    Kiemeléseim navigate_next
    Mutasd a szövegben:
    Szűrés:
    Kiemelések létrehozásához
    MeRSZ+ előfizetés szükséges.
      Útmutató elindítása
      delete
      Kivonat
      fullscreenclose
      printsave