Assessments and question writing

[ncorbin-id]

Let's think about question writing in terms of Bloom's Taxonomy:

We use Bloom's taxonomy to classify the level of understanding in our learners. For example, are they able to recall only basic facts, or are they able to draw connections among concepts? We can write questions for instructional material based on these levels to:

• Assess the level of understanding the learner has
• Invoke higher levels of understanding by prompting the learner with higher-level questions

The typical recommendation is for chapters to primarily start with questions at the remember and understand levels since we need to ensure these skills are solid before we start adding on additional content. Learners also may not have the necessary skills to make connections to other concepts or ideas at this stage. At the end of chapters, we can start to incorporate more apply and analyze kinds of questions, summarizing content from the entire chapter. This isn't a hard and fast rule, though, and appropriate higher-level questions can be included throughout the chapter.

Without an embedded quizzing mechanism in github, I recommend putting answers to the questions at the end of the chapter for the learners to self-assess. Ideally, learners get feedback in the moment after answering a question, but that may not be possible with our infrastructure.

More resources:
http://faculty.academyart.edu/faculty/teaching-topics/teaching-curriculum/enhancing-teacher-student-interaction/different-types-questions-blooms-taxonomy.html
https://www.mandela.ac.za/cyberhunts/bloom.htm

Example template questions:

• [Remember] What is the output of this function if you change the inputs to these different values?
• [Remember/Understand] What inputs do you need to create this output?
• [Apply/Analyze] Why did/will this code sample fail? (syntax, logic, or runtime errors)
• [Apply] Is this method appropriate for a different dataset?
• [Apply] Here is a new scenario. How can the tool we just used help to solve a problem in this new scenario?
• [Analyze] Why can't we do these steps in a different order? (logical reasoning)

Specific examples from the matplotlib chapter:

After section Line plots:

• [Remember] What change would you need to make to set the Temperature (surface) line black?
• [Remember] How can you change the legend location to the lower right?
• [Understand] What are two ways to turn the grid off in this plot?
• [Understand] Do the x and y labels need to be set before the title?

After Working with multiple panels in a figure:

• [Remember] What inputs do you need in the add_subplot method to create a 4-panel figure with two axes on top and two on the bottom?
• [Understand] Why are methods like set_xlabel and set_title operated on the axes object instead of the figure object?

At the end of the chapter:

• [Apply] Which of the plotting methods shown here is best for displaying satellite imagery?
• [Apply] You are looking for evidence of a diurnal cycle in a new 1D dataset you just received. Which plotting method would be the most appropriate to use?
• [Analyze/Evaluate] You notice that a scatter plot you just created is showing the axes, but you see no data shown. What could be causing your data to not show up in the axes? (pulling on general troubleshooting skills... issues with data: nulls, issues with plot: axes limits, marker color isn't visible, etc)