Pedagogical principles for the book

• Problem-based learning
  • Supporting learning around real-world challenges
  • The walkthroughs were designed with actual problems edu. data scientists face in mind
• Differentiation
  • Chapters 5 and 6 for those completely new to R
  • Appendices as 'reaches'
• Working in the open
  • R View blog post
• Building mental models
  • Foundational Skills Framework: Projects, functions, packages, data
• Universal design
  • Inclusive and accessible to all individuals
  • Did not adequately address (based upon our aim)

Strategies for teaching data science

• Provide a home base
  • Reduce some of the work that learners have to do that is unrelated to reasoning about and using code
• Write code early and often
  • Most everyone can learn to code; begin with reasoning about the output of code and writing code early
• Don't touch that keyboard!
  • Supporting learners to fix typos and become familiar with syntax is well-worth the effort
• Anticipate issues (and sacrifice accuracy for clarity early on)
  • The curse of knowledge is hard to overcome; be generally accurate but emphasize being clear and pragmatic early on
• Start with "early wins"
  • It is important to consider the end point; what will learners want to do or make?
• Consider representation and inclusion
  • Doing data science and working with data in education make these even more important than in other domains in that we can make impressions upon learners about who does data science and what data we create
• Draw on other resources
  • There are great resources in computing education and K-12 education to draw on

What's next?

• There is a lot of growth and interest in data science education
• But, learning to do data science is still really hard!
• What are big gaps in resources?

Thank you!

Joshua Rosenberg
jmrosenberg@utk.edu
@jrosenberg6432