加载中…如何撰写英语科技论文之十：讨论 （Discussion)
(2010-11-29 21:27:14)
标签：
教育 |
5.6 Discussion
Before to decide what to put in the Discussion, re-read the hypothesis and mentally match up its expectations with the principal results might be the start point of writing the Discussion section. In collecting, processing and presenting the data of the research, an author would have formed a number of ideas that might be developed in the Discussion. Such ideas might come to the author almost anywhere and anytime. In this process, many ideas might be perished soon but a few will come through as the important features of the Discussion. These ideas need to be refined and related in a logical way to his/her data and the literature. Such developed ideas are the arguments that the author will have to justify in the face of what is already known of the subject and present their limitations as honestly as he/she can. In such situation, the Discussion then becomes a collection of arguments about the relevance, usefulness, and possibilities or limitations of the author’s research and its results (Lindsay, 1995).
5.6.2 Structure of the Discussion section
There is no standard way to write the Discussion section, giving authors a freer hand to write his/her arguments than in the preceding sections. It is logical, however, to begin with a brief reminder of the most important findings and then to move toward progressively more generalized statements about the relationships among the data, the connections between the present findings and those of the other researchers, and ultimately the principles and inferences that are applicable universally. A typical structure of the Discussion is recommended by James Hartley (2008):
Step 1: restate the findings and accomplishments.
Step 2: Evaluate how the results fit in with the previous findings – do they contradict, qualify, agree or go beyond them?
Step 3: List potential limitations to the study.
Step 4: Offer an interpretation/explanation of these results and ward off counter-claims.
Step 5: State the implications and recommend further research.
(p.49)
Using this structure, some fundamental questions should be addressed in the Discussion section:
¨
¨
¨
¨
¨
¨
For example, in the research paper “Grazing intensity affected spatial patterns of vegetation and soil fertility in a desert steppe” (Lin, et al. 2010), the Discussion section is organized by the following structure:
|
4. Discussion 4.1. Effects of grazing intensity 4.2. Effects of sampling scale 4.3. Relationships between vegetation and soil spatial patterns (Lin et al., 2010, p.288-291) |
In each of the
sub-sections in the Discussion, the fundamental questions are
answered. For example, in “4.1 Effects of grazing intensity”, they
answer the fundamental questions in the following paragraphs.
|
What major pattern the data exhibit? - |
Paragraph1: “The reduction of AGB by intensive grazing led to the dramatic decline of range for AGB and disappearance of its spatial dependence in one HG plot (Table 1 and Fig. 5a) …” “We also found that vegetation patches (roughly circular) became more homogenous under intensive grazing (Fig. 5b)”… (Lin et al., 2010, p.288) Paragraph 2: “Even though plant height was significantly correlated with AGB (Table 3), its spatial heterogeneity was not decreased by heavy grazing (Tables 1 and 2),” … Paragraph 3: “Those responses would likely happen in LG that maintained a similar plant biomass and vegetation height as CK and contribute to its higher C/N ratio, as moderate and heavy grazing strongly decreased biomass and modified spatial patterns of vegetation (Figs. 2 and 5)….” Paragraph 4: “We did observe the decreased heterogeneity of SOC patches with the increasing grazing pressure (Fig. 5b)…” Paragraph 5: “Compared to SOC and TN, soil NH4 + and NO3− only had patchy spatial distribution in a few plots at fine scale and did not respond to the grazing intensity (Table 1), which is consistent with some previous research (Rietkerk et al., 2000; Augustine and Frank, 2001).” (Lin et al., 2010, p.289) |
|
How do the methods, results, and interpretations compare with those of others? - |
Paragraph1: “At fine scale, the decrease in AGB and vegetation height along the grazing gradient (Fig. 2) is consistent with a previous study that was conducted in a similar desert steppe in China (Wei et al., 2000)”… “Consistent with our near-surface photographic data which showed that relatively larger vegetation patches were broken up to many smaller ones along the grazing gradient (Lin et al., in revision), our results suggested that vegetation fragmentation increased with increasing grazing intensity in this desert steppe.” “Previous studies have also found that fragmented vegetation induced loss of rare animal species and altered animal community composition (Golden and Crist, 1999). Therefore, range managers should avoid the use of high intensity grazing to maintain vegetation spatial pattern.” (Lin et al., 2010, p.288) Paragraph 2: “We did observe the decreased heterogeneity … increasing grazing pressure (Fig. 5b), which is consistent with the long-term (over 25 years) responses of SOC spatial patterns to grazing in a semi-arid steppe in Inner Mongolia (Wiesmeier et al., 2009). Paragraph 5: “… soil NH4 + and NO3− only had patchy spatial distribution in a few plots at fine scale and did not respond to the grazing intensity (Table 1), which is consistent with some previous research (Rietkerk et al., 2000; Augustine and Frank, 2001)”… (Lin et al., 2010, p.289) |
|
Are any discrepancies with other’s results or unresolved issues addressed? - |
Paragraph 2: “Even though plant height was significantly correlated with AGB (Table 3), its spatial heterogeneity was not decreased by heavy grazing (Tables 1 and 2), suggesting that spatial pattern of height distribution was different from that of AGB.” … “Though plant height has been successfully used to study the spatial pattern of vegetation (Olofsson et al., 2008), these results suggest that plant height only represents one aspect of spatial vegetation structure and it should be used with other parameters (e.g., AGB) for interpretation of data.”… Paragraph 4: “Loss of soil C and N by overgrazing has been found in different types of steppes in Inner Mongolia (Su et al., 2005; Han et al.,2008; Steffens et al., 2008) … However, neither SOC nor TN responded to grazing intensities in this study (Fig. 2).”… (Lin et al., 2010, p.289) |
|
Do the data allow important extrapolations or predictions? - |
Paragraph3: “Those responses would likely happen in LG that maintained a similar plant biomass and vegetation height as …modified spatial patterns of vegetation (Figs. 2 and 5).” “Our results suggest that grazing-induced changes in litter input and erosion could influence soil organic matter content; however, the length of treatment time (4 years) and small sample size might have limited our ability to detect responses of soil total C and N to grazing intensities.” “We will not be surprised to see significant degradation of soil organic matter after longer period of treatments.”… (Lin et al., 2010, p.289) |
|
How do the findings provide answers to the posed problem? Relate the interpretation of results to the hypothesis. |
Paragraph5: This spatial pattern of soil mineral N suggests that their microbial driven dynamics were randomly distributed in this desert steppe. Interestingly, soil NH4 + showed fine scale spatial patterns in LG plots, where the availability and overall variations of NH4 + were higher than heavy grazing (Figs. 2 and 3). Low grazing pressure likely positively influenced soil NH4 + availability by increasing N mineralization, litter input and plant root exudation, especially within vegetation patches (Holland et al., 1996; Frank and Groffman, 1998; Tracy and Frank, 1998). (Lin et al., 2010, p.289) |
|
What conclusions, generalizations, and implications do the results allow? - |
Paragraph5: In this desert steppe, those indirect effect of sheep grazing on mineral N availability were more important than the direct effect of excreta addition (Hobbs, 1996), as HG was supposed to have higher N in put from grazer excreta than LG. Our results suggest that grazing could change the small-scale spatial patterns of soil NH4 + and could consequently preserve soil fertility and forage production, if those influences are consistent over time. (Lin et al., 2010, p.289) |
5.6.2 Development of arguments
The above examples show that each of the arguments in the Discussion can be considered as a separate piece of logical writing and will normally be the substance of a complete paragraph. The technique of developing arguments is identical to that of good paragraphing. A paragraph is the development of one idea. It gives the readers visual help, in breaking up the total bulk of the Discussion and helps them to absorb the key points at a time (Lindsay, 1995).
A good paragraph normally has three components: Topic sentence, Logical development, and a concluding message or a summary of the point reached in the paragraph. These three components can be used to develop the arguments in the Discussion.
The topic sentence of a paragraph is normally at the beginning or the end of the paragraph. It is a brief statement of what is discussed in that paragraph. Writers can start their logical sequence by starting the paragraph with a mini-summary of what is to follow. Such a mini-summary in fact is the topic sentence that can paraphrase the main point the writer wish to make in the paragraph. It immediately attracts attention and puts the reader on the right mental wave-length to receive the ideas on the logical developments which follow. The topic sentence, therefore, should also act as a link with the previous paragraph. This enhances the coherence of the whole discussion effectively.
Logical development
In a paragraph, the logical development aims to draw a sound conclusion by deduction, induction or a mixture of both. Writers can use facts from research results and combining these facts with others to make their points. Each argument is unique and is supported by its own set of facts so there can be no specific rules here but should avoid a number of fallacies of logic, such as generalizations, authority, and expressions of confidence.
Premature generalizations based on a few preliminary results, or on results obtained under a limited set of conditions could lead to a disaster later on. A valuable use of statistical analysis is to minimize the chances of making foolish generalizations. The writer should not draw rash conclusion by the early evidence. Even when statistical analysis shows that the results are unlikely to be due to chance, the generalization must be guarded.
Most scientific arguments are based on one or more authoritative sources. Writers should carefully choose the authoritative sources to sure that the source of authority is not out of date, or controversial, or simply wrong. The writers also need to ensure that the principle on which the argument is relying is currently accepted and recognized.
Moreover, the conclusion articulated in a discussion must be supported closely by the specific observations and the relevant patterns in the data. For instance, Lin, Hong, Han, Zhao, Bai and Chang (2010, p.289) support one of their conclusions — “that grazing could change the small-scale spatial patterns of soil NH4 + and could consequently preserve soil fertility and forage production, if those influences are consistent over time” — by concisely enumerating their relevant methodology and data patterns, as well as the findings of other research:
Compared to SOC and TN, soil NH4 + and NO3− only had patchy spatial distribution in a few plots at fine scale and did not respond to the grazing intensity (Table 1), which is consistent with some previous research (Rietkerk et al., 2000; Augustine and Frank, 2001). This spatial pattern of soil mineral N suggests that their microbial driven dynamics were randomly distributed in this desert steppe. Interestingly, soil NH4 + showed fine scale spatial patterns in LG plots, where the availability and overall variations of NH4 + were higher than heavy grazing (Figs. 2 and 3). Low grazing pressure likely positively influenced soil NH4 + availability by increasing N mineralization, litter input and plant root exudation, especially within vegetation patches (Holland et al., 1996; Frank and Groffman, 1998; Tracy and Frank, 1998). In this desert steppe, those indirect effect of sheep grazing on mineral N availability were more important than the direct effect of excreta addition (Hobbs, 1996), as HG was supposed to have higher N in put from grazer excreta than LG. (Lin et al., 2010, p.289)
The conclusion sentence in a paragraph should be expressed according to the force of research data. The expressions, such as “It may be possible that …’ or ‘The possibility exists that…’ should be avoided since such sentences show insufficient confidence of the data. On the other hand, over-statements such as “there is a clear indication…’ or ‘there was a marked response…’ are neither not permitted in the conclusion. The safest expression is honest, factual, and eminently scientific (Lindsay, 1995).
Concluding message: To round off the argument and the paragraph the writer can emphasize in the final sentence the key point of what have been developed in the argument. For example:
“…our results suggest that small-scale responses of spatial patterns to grazing should be emphasized in the future to understand the influence of grazing on ecosystem processes and services in arid and semi-arid Eurasian steppes.” …
To summarize, each of the sentences in a paragraph has a clear message with emphasis. Thus, a well developed argument can be made in the paragraph and the reader can be convinced by the logic of reasoning. In practice it is a good idea to play with the sequences of information in each argument in the form of notes. In this way, a writer can decide finally on what sequence seems most logical and can therefore be understood quickly by a reader (Lindsay, 1995).
喜欢
0
赠金笔