teaching principles

Other things to implement

• live feedback from googlesheets for course

Tips from Data-carpentry Instructors training

Seems to be a lot of flexibility still in data carpentry

• see http://www.datacarpentry.org/workshops/ Data Carpentry workshops must include:
• A Data Carpentry lesson on data organization
• Three other modules in the same domain from the Data Carpentry curriculum

http://www.datacarpentry.org/lessons/ - the links don't seem to work at the moment, here are the lessons directly: http://www.datacarpentry.org/spreadsheet-ecology-lesson/ Data Carpentry SQL for Ecology - http://www.datacarpentry.org/sql-ecology-lesson/ Data Carpentry OpenRefine for Ecology - http://www.datacarpentry.org/OpenRefine-ecology-lesson/ Data Carpentry R for data analysis for Ecology - http://www.datacarpentry.org/R-ecology-lesson/ Data Carpentry: Python for Ecologists - http://www.datacarpentry.org/python-ecology-lesson/

• live coding

  • don't spend too much time with slides or doing pre-amble
  • straight into the session
  • really use the sticky notes to keep the class up to date (use red and green sticky notes that learners attach to the back of their laptops to show when they are stuck/ok)
  • do it with pair programming approach

• use concept map for tricky parts

  • e.g. data types

• beware cognitive load

  • ?7 concept limit

• test driven teaching

  • intro lecture should show how ggplot can be used to look at data really quickly in different ways (the test - we want the students to be able to do this)

• red and green stickies for polling the room

• keep the student to helper ratio <1:7

• use MCQs via the stickies at the end of the topic, and use the MCQs to check the understanding of each topic; make sure the answers include 'plausible distractors'; use these to understand where people are going wrong

• run a shared etherpad or similar (google docs?) to keep all the links that people need, notes etc

• peer learning: ask qns and get the class to vote, then discuss amongst themselves the answers chosen (without revealing the correct answer), and then ask again (the class will have taught themselves); scales well even in big classes

• pair programming

  • driver types: in charge of the keyboard and the mouse
  • navigator: more experienced of the pair

• run live etherpad as a copy-paste resource

  • then when live coding, people can copy paste your commands from there
  • this means you need to have another person working with the instructor to write the commands into the etherpad as your proceed

Use some of the scheme from the ecology workshop for data carpentry