Action Research On The Effects Of Graded Reading-to-write Training On Junior High School Students' English Writing Ability

English writing is a productive process of comprehensive skills most difficult for Chinese students to improve. Given the importance and difficulty of writing teaching,many researches and practices have been done about the teaching and learning of English writing both at home and abroad, but few investigations or researches have been conducted on instructing senior high school students' English writing with reading in accordance with students' specific situation. According to research findings on foreign language teaching at home and abroad, reading-to-write training is an effective teaching method for improving students' language proficiency. Its facilitating effects in writing can be interpreted by comprehensible input hypothesis, comprehensible output hypothesis and alignment theory as helping to integrate input and output and transform to students English ability.In view of the students' writing deficiency in grammar, spelling, words expression,coherence and relevance, the author decided to probe into the effects of reading-to-write training on students' English writing proficiency. The study took action research design to conduct reading-to-write training of graded difficulty levels in three rounds with each round tackling students' different writing problems. The research expected to answer two research questions:(1) Is graded reading-to-write training an effective way to improve senior high school students' English writing ability?(2) Does graded reading-to-write training have different effect on students of different English level in terms of language, relevance and coherence?The 15-week action research had 50 senior high school students as participants under three rounds' training: the first round reading-to-write training aimed at sentence-level output; the second round have paragraph writing as the goal; and in the third round, students were required to summarize the general idea of an article or a paragraph.Both qualitative and quantitative data were collected for research questions. Based on the scores of students' pre-test and post-test, Paired samples T-tests were employed to test whether there existed significant difference in pre-test and post-test. Meanwhile, theinterviews were conducted for teacher's reflection and to adjust the plan in action research. Results show that the overall score of post-test is 11.1400 while that of pre-test is 8.5000. The T-value is-9.543 and the 2-tailed significant difference between two tests is 0.000. And top students' mean score of language in post-test is 7.0000 in comparison to6.7500 in pre-test; and their relevance and coherence in post-test is 3.5000 as against 3.6250 in pre-test. Intermediate-level students' language in post test is 6.2759 while 4.5517 in pre-test; and their coherence and relevance is 2.6897 in post-test and 2.1724 in pre-test.Slow learners' language in post writing test is 4.4615 over 1.8462 in pre writing test; and their coherence and relevance is 1.9231 in post-test and 0.1538 in pre-test.From the findings we may conclude that reading-to-write training conducted in graded level can improve senior high school students' English writing ability and has different effect on students of different English level in terms of language, coherence and relevance. The data-based reflection offers implications on how graded reading-to-write training should be conducted progressively to focus on specific aspects of writing round by round, which has not only improved the author's professional understanding of writing teaching practice, also may be of pedagogical reference to other practitioners'.