This is a great question. Firstly, there are different types of shared writing.
Sharing your writing
There is quite literally the idea of sharing your writing by crafting mentor texts for your class to discuss and read. These texts should match what you’re expecting the children to produce as part of a class writing project. Research has shown that this kind of practice can yield an effect size of +0.76 (+0.94 for children with SEND). For context, anything over +0.4 is considered to have a significant positive effect on children’s writing development. You can read more about this here.
Incidentally, if you establish product goals for a class writing project in response to studying a variety of mentor texts, this can yield an effect size of +2.03 You can read more about this here.
Next, there is shared writing. Modelling how to use certain craft moves before inviting children to use these craft moves for themselves in that day’s writing time can yield an effect size of +1.75. For children with SEND, this can be anything up to +2.09. Case studies show that the most effective writing teachers use shared writing when teaching ‘craft knowledge’ (Young et al. 2021), ‘sentence-level strategies’ (Young & Ferguson 2022c) and ‘functional grammar lessons’ (Young & Ferguson 2021b). You can read more about shared writing here.
You also have the idea of write alouds. This is the writing version of ‘read aloud’ time. This is an opportunity for teachers and children to come together and write something collaboratively (shared), for pleasure, as a community of writers (Young & Ferguson 2020). This can be done on an IWB or some flipchart paper. Write alouds can be done in a single sitting or over multiple sessions.
Sharing the writer’s life
Finally, there is the concept of sharing your writer’s life. This is about modelling how to be and live as a writer. You can read more about this here. Sharing your writer’s life and writing alongside your pupils during writing time can yield an effect size of +0.54. For children with SEND, this can be anything up to +2.48 (Young & Ferguson 2023b).