An action research for the improvement of academic writing of international learners

The aim of this study is to develop a course model to improve the academic writing skills of international students studying at undergraduate/graduate level. With this course model, it is aimed for international students to prepare qualified research papers, reports, scientific writings by identifying structural organization and metadiscourse features in accordance with the communicative purpose and to transfer this writing culture to more comprehensive texts such as articles, papers and theses. For this purpose, the content analysis of 171 articles, 496 papers and 229 essays from different disciplines, genres and subjects was carried out with MAXQDA software and MaxDictio, an add-on of this programme. Firstly, 605 different word forms consisting of 210 word bases and their different inflections and 136 different forms belonging to 59 collocations of the most frequently used word bases were determined. Secondly, by analysing the different forms of word bases, 25 sentence structures frequently used in academic texts were identified. Thirdly, 71 in-sentence connectives and 69 sentence-head connectives were listed under 8 headings: ordering, stating cause and effect, stating purpose, comparing, adding, exemplifying, restating-explaining and stating condition. Fourth and finally, writing genres and sub-models belonging to these genres were determined through the structural analysis of 229 academic texts in the essay genre. In the academic writing course model prepared on the basis of these findings, grammatical structures, vocabulary, basic phrases, connectors, paragraph structure and planning stages of thoughts specific to texts in comparative, cause-effect, advantage-disadvantage, argumentative, problem-solving and expository discourse genres are included. In this study, which was conducted as an action research, the implementation of the course model was carried out in a 7-week period with 15 students who had completed C1 level, and the academic writing self-efficacy scale and academic writing achievement test were applied as pre-test and post-test. Descriptive analysis, categorical content analysis and frequency analysis were used to analyse qualitative data, and dependent samples t test was used to analyse the quantitative data. The quantitative data showed that the course model prepared based on the analysis of qualitative data contributed to the development of students' academic writing skills. Both academic writing self-efficacy perceptions and achievement levels of the students participating in the application increased. Among the prominent findings of the study were that students' awareness of academic discourse improved and they started to analyse texts on the basis of genre according to the purpose of writing. In the analyses, it was determined that students showed progress in all criteria of form, spelling and punctuation, syntax, grammar and sentence structure, vocabulary, content and paragraph organisation and metadiscourse markers. These findings indicate that teaching the course model developed based on academic text analyses with a process-genre-oriented approach is more effective in the development of writing skills. When the expectations, needs and opinions of international students and academic Turkish language instructors whose opinions were consulted within the scope of the study were examined, it was understood that the current programmes and contents were inadequate and the approach put forward by this study was found to be useful and successful. In conclusion, it can be said that the planned and systematic transfer of the course model prepared based on the student needs, instructor opinions and the findings provided by the data pool including 892 texts in the teaching of writing skill, which is included in the production, interaction and mediation modes of communication, at the academic level will increase students' academic writing achievement and academic writing self-efficacy perceptions.