【Abstract】Using Hylands interactional meta-discourse framework as the model， the study is based on a sub-corpus of CLEC， aiming to make comparisons of stance markers among students with different scores.

【Key words】Argumentative Writing； Stance Markers； CLEC

【作者簡(jiǎn)介】糜卓穎（1994.05.19-），女，湖北武漢人，碩士學(xué)位，武漢理工大學(xué)，研究方向：應(yīng)用語(yǔ)言學(xué)。

1. Introduction

Most students find it hard to make persuasive arguments in argumentative writing. Stance expresses voice. Proper stance markers are vital in argumentative writing. Previous studies were mostly based on academic articles instead of L2 argumentative writing. In view of this， this study will study stance markers in L2 argumentative writing. More specifically， this study will explore following questions：

1.1 What are similarities of stance markers between the higher group and the lower group in argumentative writing？

1.2 What are differences of stance markers between the higher group and the lower group in argumentative writing？

2. Methodology

2.1 Data Collection

The classification of stance in this study is shown in Table 1.

Table 1 The Classification of Stance Markers （Hyland，2008）

Category Function Examples （4）

Hedges withhold commitment and open dialogue perhaps；

Boosters emphasize certainty or close dialogue in fact；

Attitude Markers express writers attitude to proposition unfortunately

Self Mention explicit reference to authors we

This study investigated stance markers used by non-English major students from a sub corpus of CLEC. Coders selected 60 argumentative essays from ST3 randomly and divided those essays into a high-score group and a low-score group according to a rubric developed by Qin. Then， in order to ensure accuracy， the two coders firstly coded stance markers in 10 argumentative essays. Before negotiations， the two coders accordance ratio achieved 88%. Later， two coders coded stance markers in these 60 argumentative essays and argued those controversial ones to reach a consensus. And final results were shown through chi-square tests.

2.2 Tools

The corpus tool used in this study is SPSS 17. Researchers calculated standard frequencies in per 1，000 words. Then researchers observed results through chi-square tests.

3. Findings and Discussion

Final results are shown in Table 2.

Table 2 Distribution of stance markers

3.1 Similarities of Stance Markers between the Higher Group and the Lower Group

There are no significant differences in hedges and boosters as shown in Table 2， but both high-score students and low-score students use more boosters than hedges.

In addition， compared with other 3 categories， the number of attitude markers in both groups is the least. Researchers argued attitude markers make readers explicit about writers attitudes in some degree， but readers can comprehend authors viewpoints through contents. Using too much attitude markers will make the whole passage informal.

3.2 Differences Stance Markers between the Higher Group and the Lower Group

From Table 2， there existed a significant difference in self-mentions. High-score students used more self-mentions in their writings， which may be concerned with writing education. Besides， high-score students used more attitude markers since high-score students have various ways to express themselves.

4. Conclusion

The study reveals compared with low-score students， high-score students use more authorial presence markers. Low-score students have limited ways to express themselves. Neither high-score students nor low-score students use attitude markers frequently， but high-score students are prone to express their views direct and make their arguments persuasive.

References：

[1]Hyland，K. Meta-discourse[M].Beijing：Foreign Language Teaching And Research Press，2008.

[2]Qin J. Applying the Toulmin Model in Teaching L2 Argumentative Writing[J].Journal of Language Teaching & Learning，2013.