Abstract:
The purposes of this study were to 1) compare the students English paragraph writing before and after learning through corpus-based approach, 2) to compare the students grammatical writing ability before and after learning through corpus-based approach, 3) to compare the students use of vocabulary before and after learning through corpus-based approach, and 4) to investigate the students attitudes towards using a corpus-based approach to develop their paragraph writing ability. Thirty-four Mathayomsuksa 5 students were the participants. The research included both quantitative and qualitative research instruments. The instruments were writing ability tests, semi-structured interview, and students learners diaries. The results were as follows:
1) The mean scores of the writing ability test after learning (x̄ = 15.94) were significantly higher than the mean score of the scores before learning (x̄ = 11.44).
2) The frequencies of the students grammatical errors found in the writing ability test after learning were lower than the errors before learning.
3) The number of types and token ratio of the students vocabulary usage in the writing ability test after learning were higher than before learning.
4) The learners diaries showed students positive attitudes towards learning through corpus-based approach in writing classes and using corpus websites helped students to avoid committing grammatical writing and lexical errors. The results from the semi-structured interviews also indicated the benefits of corpus materials as it was beneficial for error corrections and developing the students English writing ability, grammatical writing ability, and vocabulary usage.