This reveals one of the mistaken notions about writing students are often given, that research is a discrete stage prior to writing. The reality is that you may move between research and writing constantly and there's no reason to draw a distinction between the two.
Research is fuel for the writing and so you have to go get fuel whenever fuel is necessary. The reason we (I've been as guilty as this as anyone in the past) teach a process where research happens before writing begins is because that's easy to teach, structurally.
I grew up in the era of writing individual facts down on index cards as part of my research. I had no idea why I was doing it, other than the teacher required me to have 20, 30, 50 index cards before I moved on to the next thing.
As an instructor, I did something similar, telling students to pick a topic, do a library search, then an annotated bib, and so on. I was signaling that stages were discrete. They aren't, writing doesn't work that way.
Once I disaggregated how I teach writing from what is easy to assess and quantify (# of sources, correctness of bibliographic entries, etc...), I could help students see how writing works as an act of thinking, and how writers research as they go, for fuel.
This method makes it much harder to stage writing assignments with discrete checkpoints for grade assignments, which is why I had to move away from traditional grading. We're going to work the process and not overworry about everyone being on the same page simultaneously.
If you want to see how this works practically, it's all in here: penguinrandomhouse.com/books/566892/t…
