UTS model of learning (writing): Writing for data scientists

At UTS we have a model of learning, that is built on in the ‘learning.futures‘ programme, both of which target course-based graduate attributes:

This runs in parallel with work on assessment futures, which has drawn on examples across the HE sector by different subject areas. I’m currently interested in what it would look like to build an evidence hub to support writing practices in learning futures, thinking about ‘why do we write at UTS’, to articulate the UTS model of learning in the specific context of writing practices – a key professional and academic skill . For example, in our Masters in Data Science and Innovation,”what is writing for data scientists?”, what does it look like/involve, what genres and practices are important for our students, what do the students need to learn to write like a data scientist (learning to write), and how will the writing practices of a data scientists help them develop as professional data scientists (writing to learn).

“Ultimately, and in an important sense, we are what we write, and we need to understand the distinctive ways our disciplines have of addressing colleagues and presenting arguments, as it is through language that academics and students conceptualise their subjects and argue their claims persuasively.”

Genres are abstract, socially recognised ways of using language. Genre analysis is based on two central assumptions: that the features of a similar group of texts depend on the social context of their creation and use, and that those features can be described in a way that relates a text to others like it and to the choices and constraints acting on text producers.

This approach fits into new literacy studies (NLS) analyses of literacy as ‘social practice’ . Building on this approach Lea and Street point out that in university contexts, students are not only being socialised into a particular professional genre, but must respond to institutional contexts that include market pressures, accreditation schemes,and power dynamics. They propose that an ‘academic literacies approach’ should consider the skills and genres students are exposed to and required to address, as well as the particular institutional contexts in which they are working.

The articulation of a broad model of learning for a narrower component of learning, then, should – I hope – help support students and academics in understanding the role of writing in particular subjects and courses. I’m hoping that this articulation will assist in thinking through how we build writing for a purpose into our subjects, embedded in subject-based genres, making use of the rich tools available: in the writing process; in the site, platform, or media of that writing; and in the assessment and feedback mechanisms for the writing. This builds on, for example, the Conference on College Composition and Communication Assessment Committee Statement, the core elements of which might be summarised for our purposes as:

  1. Writing should be embedded in the curriculum, not an afterthought or a means to an end
  2. Writing is genred, writing in learning should be for a purpose (e.g. respecting local and professional needs)
  3. Writing is social, in its process, and its purpose
  4. Feedback is social, we write for readers, assessments should not disconnect writers from readers (for that reason they reject automated assessment, although we’d argue these can support writers, and feedback discussions)
  5. Writing assessment criteria should be explicit, and reflect the range of strengths and areas for improvement across the students (criteria might be called ‘feedback statements’ here)
  6. Writing assessments should be reviewed regularly for potential improvements

So, to take the data science example.

  1. Writing should be considered embedded across a data science curriculum
  2. What types of texts do data scientists consume, and produce?  What is their evidence base?
  3. How are these written, and framed, what is their context, audience, and genre? Tech report, discussion paper, blogs, visual and textual elements – using visualisation to explain text based and vice-versa, scientific reports/experiment outputs.
  4. What kind of specific platforms or spaces do data scientists write in? E.g. writing dynamic variable webpage reports with php, etc., shiny web apps, python, and then collaborative spaces like github, wikis and etherpads/google docs, etc.
  5. Are there particular skills involved in writing for data scientists? E.g. creating infographics and other visual methods, data analysis and reporting methods, data-based decision making, statistical methods and report for a variety of audiences, etc.
  6. The nature of the writing across the subjects should be reviewed as the field develops, in light of emerging needs, and developing assessment practices

So the questions are:

  1. How do we represent this approach to create a framework to explore the particular kinds of writing activities a subject might include?
  2. How do we ensure that this isn’t just a ‘lip service’ (e.g. a shallow NLS approach might ask students to create a social artefact, but outside the genre of the discipline, or to write about content within the discipline but with a very shallow understanding of the genres appropriate to it)?
  3. How do we integrate across the features of the framework, and develop learning information (or learning analytics data) to support our student’s learning?

I’m starting to think about how this could be represented to probe the ways that writing tasks across a curriculum meet the needs of the course/subject. E.g., a simple form and questions could scaffold such a process.

Key features in the target profession Notes or rationale
Evidence base (e.g. Peer review literature, open data sources…
Genre (e.g. Industry white paper, technical report, live data analysis document…
Writing tools (e.g. Shiny web app, collaborative writing tools, word processing tools…
Writing skills (e.g. Visualisation and statistical inference, literature synthesis…


Fill in the grids below keeping the following in mind. (Texts in the grid below are any texts students are asked to create, primarily in assessments, across a course over a set of subjects below)
For each text (in the columns), does each feature belong to that type of text? What features of the text are flexible and which are fixed for this assignment? In each text above, are there any features of the text missing?
Text 1 Text 2 Text 3 Text 4 Text 5 Text 6 Text 7 Text 8 Text 9 Text 10 Text 11 Text 12
Evidence base
Writing tools
Writing skills
Looking across the texts, and the features present in each, consider:

  • Is everything included that should be – are there factors that relate to the particular professional context that are missing?
  • What factors are flexible, or could be amended to reflect the professional genre?   
Notes or comments explaining how ‘writing in this profession’ is being addressed in this course:



Print pagePDF pageEmail page

This Post Has 4 Comments

  1. Simon Knight says:

    As part of my teaching I asked students (yesterday) to do an activity related to my thinking around writing in the professions. I asked them to consider:
    1. Individually, as many types of writing a data scientist might engage in as they could (on post its)
    2. Individually, taking some post its from the table’s pool, consider how uniquely ‘data science’ the example is, and what makes it more or less ‘data sciencey’
    3. As a table, to order the types of writing from more to less academic/scholarly (and what makes it scholarly)
    Obviously there are some very general mediums of writing – emails, tweets, blogs – and contexts of workplace writing like resigning and advertising jobs (although particularly the latter can be worded in terms of data science). I’ve just gone through the post-its and gathered most of the contributions – in no particular order, and without much attempt to filter or define – below:

  2. Customer segmentation/profiling report (and recommender systems based on these)
  3. Analysis of reviews e.g. sentiment analysis, scraping TripAdvisor, etc.
  4. Storyboarding
  5. Multimedia reporting
  6. Campaign report (e.g. sales performance)
  7. Audit report
  8. White paper / discussion paper on emerging trends
  9. Operations efficiency report
  10. Risk analysis
  11. Benchmark report
  12. Needs analysis
  13. Policy and strategy documents
  14. Experiment report
  15. Method validation and tech reports
  16. Specification report
  17. Sales pipeline summary
  18. Root cause analysis
  19. Briefing notes
  20. Presentations
  21. Impact analysis
  22. Business plan/business case
  23. KPI Dashboards
  24. Scenario analysis (what would happen if x…)
  25. Product descriptions
  26. Technical documentation, software or analytic solution guide (and help)
  27. Acceptance testing procedure
  28. Lots of references to use of stats in various types of reporting
  29. Just the facts
  30. PR release kit/media kit, press release
  31. Infographic
  32. Competitor analysis
  33. Software specification
  34. Data access request
  35. Security and compliance report
  36. Case files
  37. Cost analysis
  38. Sales report
  39. Cashflow report
  40. Simon Knight says:

    “genre in the age of multimodality: some conceptual refinements for practical analysis” Bateman

    “alignment strategies and genre variation in students graph commentaries” Sancho-Guinda

Leave A Reply

Leave a Reply

%d bloggers like this: