README and outline deadline: YYYY-MM-DD
As part of the 'outline' process, you will need to complete the following tasks:
- Edit this README by filling in the information for step 1.
- Update the requirements.sh file in your practice pool repository on GitHub. Check with your contact as needed.
- Authoring documentation: https://authoring.datacamp.com/
- Practice admin page: https://www.datacamp.com/teach/content/challenges
- Technical documentation on challenge types: (in development)
- Make the outline of concepts in the template below; the outline lists all concepts that are taught in the corresponding course.
- For each concept, specify the concept parts: the important things to understand about this concept (e.g. nuances, syntax, arguments, data types of arguments / return types, differences with other functions, …)
- For each concept part, what is the exercise that is going to test exactly this?
- Add a challenge type to each exercise, ensuring a mix of various challenge types in each chapter.
- A typical challenge pool contains 10 exercises per course chapter.
Example from a practice pool on Introduction to Databases in Python
Chapter 1 - Basics of Relational Databases
- MultipleChoiceChallenge: does the student know the different steps of getting data out of a database and the order they go in
- make connection
- execute
- get results
- BlanksChallenge: does the student know the parts of a
create_engine
function call, and the format of it (in quotations,engine = create_engine(‘sqlite:///census_nyc.sqlite’
)- OutputChallenge: does the student understand that
repr
output has a particular format, can s/he get the needed information out of it
print(repr({{var1}})
- Var1 → different tables
it is the goal of practice exercises to
- check if the student has a correct understanding of the course
- check whether the student remembers the most important things taught in the course
- allow the student to exercise until s/he can apply what s/he's learned in the course without having to look things up
- each pool should have
- ~ 40 exercises, equally distributed over the different chapters of a course
- a good mix of the different exercise types
- a good mix of conceptual and general questions (e.g. why use this package over that package), and detailed questions (e.g. what is the name of the function of package x to do y?)
- each pool should test
- the most important misunderstandings of a course
- whether the student remembers the most important concepts and their meaning
- is it clear from the outline which concepts will be tested?
- are the most important things tested?
- is it clear what exactly is going to be tested?
- is there a good mix of exercise types?
- does each exercise check exactly what you want to test?
- are all distractors distracting? Is it realistic someone might think they are the correct answer?
- will the student spend most time answering the question, instead of understanding what is being asked?
- do the different versions of an exercise test the same concept?
- do all sentences start with a capital letter?
- is the code is formatted nicely, using newlines where necessary?
- does the output not contain more than 5 lines?
- do tables contain no more than 3 columns?
- are all exercise keys unique?