"Strengthening Connections with the Audience: Reformation and Exemplification in Mathematics Research Articles"
|
Kristy LesperanceKristy was in her third year of undergraduate studies at the University of British Columbia when this essay was originally written, studying Mathematics under the faculty of Arts. The paper was written for an upper-level, intensive research and scholarly writing course using corpus analysis to investigate discursive features of literature from the student’s chosen major. Contents |
Results & DiscussionsOverview The articles in the present corpus divided into roughly 11,300 words on average per theory article, and 8,600 words per education article. Adapting Hyland’s (2007) approach, the frequency of target word occurrences was calculated per 10,000 words (rather than 100,000 words) for clarity, since code gloss use tended to be overall lower in frequency both in comparison to Hyland’s findings as well as to other discursive features of academic articles (such as hedges, citations, or stance indicators). Due to this relative infrequency, all instances of reformulation and exemplification that were found within the bodies of texts were considered significant, although some patterns emerged with specific target words gaining popularity over others in both sub-disciplines. It seems noteworthy to remind the reader that although Hyland’s analysis did not cover the discipline of mathematics research or mathematics education, it will will be used as a comparative measure in the following discussion, being one of the few available articles on code gloss use. In general, there were nearly 1,200 code gloss indicators found in my corpus, with roughly 41 occurrences of code gloss markers per article. On average, theory articles contained 27 reformulation markers each and 16 exemplification markers. In contrast, education articles used only 16 reformulation markers on average and 22 exemplification markers. This is a fairly notable variation from Hyland’s analysis which found, in general, only about 25 occurrences of code glosses per paper. Also contrary to Hyland’s findings that 67 percent of all code gloss markers occurred within the articles of softer fields, in the present study roughly 61 percent of all code gloss markers appeared in theory (i.e., hard) articles. Additionally, the present study found an almost identical division between exemplification and reformulation in the corpus as a whole: although reformulation was preferred by 10 percent regarding the keywords listed in Table 1 (above), it was preferred by less than 5 percent when compared to all instances of exemplification (which includes more creative variations not listed). This again differs from Hyland’s findings that exemplification was preferred by 16 percent overall, when compared to reformulation. Code gloss markers in theory articles divided into roughly 37 percent indicating exemplification and 63 percent indicating reformulation, with a total frequency of about 38 instances per 10,000 words. In comparison to Hyland’s analysis, which found that roughly 66 percent of all reformulation instances occurred within articles from harder disciplines, the present analysis found roughly only 56 percent of all reformulation markers occurred within theory articles (when compared per 10,000 words). Thus, although partially aligning with Hyland’s findings, these results suggest a more modest disparity between code gloss use in hard and soft fields. In contrast, education articles showed a slightly higher frequency of code gloss markers at around 45 per 10,000 words, dividing into roughly 58 percent exemplification and 42 percent reformulation. Again similar to Hyland’s findings, a notable proportion of exemplification indicators were found in education (a softer knowledge field); however, Hyland notes that over 80 percent of uses occurred in the softer articles of his corpus, whereas, in the present analysis, only 65 percent of uses were found in education. Again, this suggests that the discrepancies between "hard" and "soft" may not be as large when comparing sub-disciplines within a single field of study, such as mathematics. Reformulation – comparisons across sub-disciplines As mentioned previously, of the 11 targeted indicators of reformulation, theory articles tended to use a greater number of reformulation markers than exemplification markers, and a greater amount of reformulation when compared to education articles. Theory articles included about 24 markers per 10,000 words, in comparison to education articles which used on average 19 per 10,000 words. This indicates that theory articles tended to use reformulation markers roughly 26 percent more frequently than did education articles. Although similar, the present results are more modest than Hyland’s findings. Both theory and education articles tended to use the same four markers about 75 percent of the time (in particular, or, i.e., and that is), and the remaining markers were used with relatively equal frequency. The use of in particular was especially salient within theory articles, having the highest frequency at 5.6 instances per 10,000 words, amounting to just over 10 percent of the total code gloss markers found within the corpus. Other commonly used indicators were i.e., with 4.2 occurrences per 10,000 words, and that is, with 3.3 occurrences. Education articles seemed to follow the trend found in theory articles, showing a preference for in particular, i.e., and that is, albeit with slightly lower frequency (3.0, 2.7, and 2.4 occurrences, respectfully). This seems to align generally with Hyland’s findings that i.e., in particular, and that is were among the most frequently used indicators of reformulation. A notable difference, however, is the use of or. In my corpus, both the theory and education articles used this word with greater frequency than the 3.6 percent found in Hyland’s corpus, amounting to 13 percent of all code gloss markers in the present corpus. About 4.7 instances per 10,000 words were found in theory articles; however, education used or at a greater frequency of 6.1 instances. |