See also Teaching statistics, Writing.
Statisticians often write about data analyses—as reports for clients, as presentations, as blog posts—but there is fairly little research on the content, form, or teaching of writing in statistics.
Radke-Sharpe, N. (1991). Writing as a component of statistics education. The American Statistician, 45(4), 292–293. doi:10.1080/00031305.1991.10475825
A very brief paper arguing that writing should be included in statistics courses because “(1) It improves writing skills; (2) it focuses internalization and conceptualization of material; (3) it encourages creativity; and (4) it enhances the ability to communicate methods and conclusions.”
Parke, C. S. (2008). Reasoning and communicating in the language of statistics. Journal of Statistics Education, 16(1). doi:10.1080/10691898.2008.11889555
Describes how writing was integrated into a statistics course in a graduate program in education. The course scaffolded writing carefully: students started by writing template sentences about their results, moved to longer concluding paragraphs to edit and improve, started making their own analytic decisions and justifying them, and so on. Each course section also showed example excerpts from journal articles that used the statistical methods and had students review their presentation. And there were peer review activities using rubrics the students developed. Overall, a thoughtful way to incorporate writing about results without simply throwing in “Write a report!” with no other scaffolding.
Woodard, V., Lee, H., & Woodard, R. (2020). Writing assignments to assess statistical thinking. Journal of Statistics Education, 28(1), 32–44. doi:10.1080/10691898.2019.1696257
Discusses a four-step structure for statistical writing (give the answer, state the facts that led to it, state the information implied by the facts, show how the implications lead to the conclusion) and shows how the author used this structure in giving short writing assignments to intro students.
???
Argues for programs to have students build portfolios of work exhibiting “strong communication”: reports, reflections, blog posts, press releases, and so on. Communication is “a crucial part of data science practice”, and so it should be taught as an integral component. Portfolios also help students demonstrate their skills when they graduate. But to do them, you have to integrate different communication activities into every part of the curriculum; they can’t just be tacked on at the end.