Collaborative writing

iStock_000020536048XSmallPaired Writing is a highly structured method of collaborative writing in which one student provides feedback on the content and organization of his partner’s writing. Having an audience for their writing motivates students to make their writing comprehensible and to polish the finished product. Scottish researchers Jennifer A. Sutherland and Keith J. Topping, Department of Psychology, University of Dundee, studied the use of Paired Writing in two classes of eight-year-old students.

They sought to determine whether, with only brief training in Paired Writing, students as young as eight could successfully use this highly structured method to improve the quality of their creative writing. Previous research on collaborative writing is limited and mostly involves older students. Existing studies indicate that training in a clearly structured method seems to optimize learning.

Paired Writing

Paired Writing was conceived as a process in which older, more able writers (parents, older students) would serve as tutors. The tutor provides a supportive framework to develop writing skills through six stages of the writing process: generating ideas, drafting, reading, editing, producing a finished copy and evaluating.

The roles of the tutor and writer are clearly defined at each stage of the process. The emphasis on thinking, planning, questioning, discussion, reorganization and restructuring counterbalances the tendency of beginning writers to focus on mechanics.

Sutherland and Topping compared the effectiveness of two pairings: students with roughly the same ability who took turns being the tutor, and cross-ability pairs in which the more skilled student served as the tutor. The performance of these two experimental groups was then compared to control groups in their classes who received the same training in collaborative writing but worked individually under the guidance of their teacher.

The child serving as the tutor starts by prompting her partner with questions to help him generate ideas while she makes rough notes. Then the tutor writes a first draft dictated by her partner. As the partner’s skills increase, the tutor gradually reduces her support, helping only when needed. If necessary, the tutor reads the writing aloud for her partner and then helps him edit it, first for meaning, then for sequence of ideas, finally for spelling and punctuation. The partner makes a finished copy of the edited work and the pair evaluate it together.

Pairs receive two 40-minute training sessions in the first week, and researchers say it is important that they have at least three more practice sessions. The time needed to become accomplished in the method varies greatly from pair to pair.

Scottish study

Students in these classes had two 40-minute periods per week devoted to writing. In the first week of the study, all students in both classes were trained in Paired Writing. Students who served as controls during the study received the same outline of the six stages of the Paired Writing process as the experimentally paired students.

To minimize the impact of different teaching styles between the two classes, teachers were asked to keep their involvement with the Paired Writing students to a minimum. All students were given a writing test at the beginning and end of the study.

Each experimental student’s pre-test, collaborative writing samples and post-test were compared to control students’ writing before, during and after the experiment. Using a short scale based on standards from the Scottish Qualifications Authority ( researchers assessed the writing samples for language use, selection and organization of ideas, and technical aspects such as spelling and punctuation.


These researchers encountered some difficulty implementing the Paired Writing method with eight-year-olds. Students tended to follow the prompt questions in the order printed rather than using only the ones relevant to the current assignment. Some of the students had difficulty with the concepts and vocabulary of generating ideas and completing a first draft. Many demonstrations and explanations were necessary.

The rest of the process — reading, editing, writing a finished copy and evaluating the final product — was not difficult for these students to understand or implement. At the beginning, students had difficulty completing a writing assignment each week, but this improved with time.

Toward the end of the eight-week study, paired students engaged in a lot of productive talk, especially the cross-ability pairs. However, despite the emphasis during training students still tended to focus primarily on spelling and punctuation rather than on meaning and order when editing their work.

There were no significant differences in pre-test scores between any of the experimental or control groups. Post-test results showed that while both groups of paired writers had improved, only the cross-ability pairs’ improvement was statistically significant.

Samples of collaborative writing also showed improvement over the pre-test individual writing. Tutors in the cross-ability pairs expressed some frustration in having to allow their partner to devise ideas for writing. However, at the end of the study, these cross-ability tutors exhibited significant improvement in their individual writing. The control students, who worked on their own, reported that the Paired Writing outline was useful in helping them organize their writing.

Limitations and recommendations

The Paired Writing outline proved too complex for some of these young students at the beginning of the project, and perhaps should be revised for use with primary students. These young students needed more time to become fluent with the method than originally planned.

While both Paired-Writing groups showed positive results compared to students writing individually, not all differences were statistically significant. Sutherland and Topping believe that the short writing-analysis scale they used was not as sensitive as they would have liked.

Their results indicate that cross-ability pairs may be more effective. Higher-ability students appear to benefit meta-cognitively from Paired Writing, and these benefits show up later on in their individual writing. Further studies with longer-term follow-up are suggested. These researchers conclude that Paired Writing is only one of several ways collaborative writing can be structured in the classroom.

“Collaborative Creative Writing in Eight-Year-Olds: Comparing Cross-Ability Fixed Role and Same-Ability Reciprocal Role Pairing” Journal of Research in Reading, Volume 22, Number 2, June 1999, pp. 154-179.

Published in ERN October 1999 Volume 12 Number 7

Leave a Reply

  • (will not be published)