Bringing Writing Research into the Classroom: The effectiveness of Tekster, a newly developed writing program for elementary students

Abstract

The Dutch Inspectorate for Education established that the quality of elementary students’ writing is below the desired level, and that the teaching of writing must be improved. The aim of this PhD research was therefore to improve writing education in upper elementary grades by developing an effective teaching program for writing, including professional development for teachers, and tools for the assessment of writing. Building on the findings of previous research, we developed and tested Tekster [Texter], a strategy-focused comprehensive program for teaching writing in grade 4 to 6. Tekster combines several evidence-based practices, such as strategy instruction, goal setting, text structure instruction and observational learning into one general approach to address both the focus and mode of writing instruction. Tekster was developed in close collaboration with elementary teachers. The effectiveness of Tekster was tested in two large-scale intervention studies with 144 teachers and 2766 students from 52 schools. Students’ writing performance was assessed with multiple tasks in multiple genres in order to draw inferences about students’ underlying writing proficiency. The quality of their texts was rated by multiple raters using a benchmark rating scale. To ensure that the intervention was implemented with fidelity, teachers were provided with a manual and training. In the first study students’ writing performance improved significantly after following the Tekster program, with an effect size of 0.40, while generalizing over students, teachers, and tasks. The improvement was established in all grades and maintained two months after the intervention. In the second study, in which we added a professional development program, the effect size was even higher, i.e., 0.55. This study specifically showed that teachers became more positive and felt more efficacious about the teaching of writing after working with the program. They were highly satisfied with Tekster and the professional development activities. Moreover, it was shown that teachers were able to transfer their newly acquired knowledge and skills to colleagues. In two subsequent studies we specifically investigated the effectiveness of the writing strategy. This strategy divides the writing process into a prewriting, writing, and post-writing phase. We found that explicitly teaching students a writing strategy enhanced their metacognitive knowledge about writing and promoted the use of prewriting activities, which improved the quality of their texts. Further, we examined whether teachers are able to adequately provide feedback to texts of varying quality. Results show that the type and amount of feedback depends more on the style of the teacher than on the actual performance of the student, and that additional training for teachers is necessary to optimize the effectiveness of feedback. Therefore feedback formed a substantial part of the professional development program of Tekster. To summarize, this PhD research demonstrates that a comprehensive writing program, such as Tekster, is a promising approach to improve writing education in elementary grades. The program appeared to be effective for students and teachers. All in all, this research provides valuable clues on how the gap between research and classroom practice can be bridged to improve writing education.