I was reading up on research on argument writing recently (in preparation for a textbook I am working on) and came across a very interesting thesis by Braaksma (2002). She investigated one aspect of what makes for the best approach to teaching writing, and in particular whether writing is best learned by observing others’, or by writing original texts. The argument in favour of observational learning is that it avoids cognitive overload (Alarmagot & Chanquoi, 2001); having to learn about writing through writing is more challenging than reading and reflecting on others’ text (Couzijn 1999) although it may not seem like this to the student at the time. Braaksma’s study showed that observational learning can indeed be more beneficial than learning by doing but this applied mainly to unfamiliar tasks. When learners have more experience with a particular type of writing (such as argument writing), they benefit more from practising their own writing. With new tasks, however, learning by observation was particularly helpful and resulted in the development of metacognitive strategies (such as the ability to plan ahead and to reflect on the writing process). For weak learners, the use of weak models was most beneficial (such as examples of poorly structured essays or those lacking good supporting evidence – of course with pointers and explanations to guide the learners!) nd for good learners, the use of good models. Braaksma suggests that weak learners gain insight by being able to identify the (incorrect) elements of a text as these are similar to their own level. For teachers this means that especially with inexperienced learners the use of observational learning is important, and that a mix of good and poor models could be used to support learners of all levels.

Alarmagot, D. & Chanquoi, L. (2001). Nature and control of processing. In G. Rijlaarsdam, D. Alarmagot, & L. Chanquoy (Eds.), Studies in Writing: Vol 9. Through the models of writing. (pp. 125-154). Dordrecht: Kluwer.

Braaksma, M. (2002). Observational learning in argumentative writing. Amsterdam: University of Amsterdam.

Couzijn, M. (1999). Learning to write by observation of writing and reading processes: effects on learning nad transfer. Learning and Instruction, 2, 109-142.