Investigation of the source text use in EFL argumentative writing

Reading-to-write tasks are becoming more and more common in language teaching (Plakans, 2008; Plakans & Gebril, 2012), and the use of reading texts plays a very important role in these tasks which require the integration of reading and writing skills. The present study aims to investigate the use of reading texts in a reading-to-write task. In the investigation of the source use, the research was conducted to study the amount of information EFL graduate and undergraduate students used from the reading texts, the way the participants used reading texts in terms of functions, the relationship between the use of reading texts in terms of function and quality of text, and whether the use of reading texts varied in terms of topic and academic level or not.In order to answer the research questions stated above, 60 Chinese graduate and 60 Chinese undergraduate students participated in the study. The data were collected from two Chinese universities. Participants were given 25 minutes to read two reading texts with opposing views and take notes while reading. Then, the participants were asked to write an English argumentative essay in 45 minutes. The task required students to take their sides, support their opinions by using ideas from the reading texts, and use their personal experiences when it is needed. A computer program was used to identify and highlight three or more words borrowed from the texts and used in each student?s essay. Three or more words taken from the same idea unit in the reading texts were counted as one borrowed string. The highlighted borrowed strings in each student?s essay were examined and counted by the researcher to study the functional use of the strings. The data were interpreted by using descriptive statistics, a correlation test and t-test results. Findings revealed that the students used the reading texts and some of the participants relied on the reading texts too much. The amount of textual borrowing varied in each group of participants. The use of reading texts by the participants was not affected by their academic level or the topic they read and wrote about. The correlation test results did not find a significant relationship between the source use and the quality of the paper. Based on these results, a number of recommendations for further research, limitations of this study, and teaching implications are provided.