Ask students to predict

Subscribe to Skilling news on the home page.

Kieran Mathieson

From Brown and Wilson, Ten quick tips for teaching programming, https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1006023:

Surprising research from peer instruction in physics education shows that learners who observe a demonstration do not learn better than those who did not see the demonstration...

The key to making demonstrations more effective is to make learners predict the outcome of the demonstration before performing it. Crucially, their prediction should be in some way recorded or public...

This makes sense. It's also something Skilling can help with, for blended and online as well as face-to-face courses.

Reflect tag

Skilling has a reflect tag. Here's a reflection example, from the demo. This example isn't about predictions. It's to encourage schema abstraction. Here's what students see:

Reflection for schema abstraction

The rest of the page is hidden. Students type in their thoughts, then click Continue to see the rest of the lesson. Students can get a list of all of their reflections, useful for studying.

Here's what authors type to make this happen:

reflect.

    internal_name=reflect_link_arguments



    Compare the two examples. What's different about them?



/reflect.

Skilling takes care of making the text field, recording responses, and such.

The reflect tag could also be used to ask students to make predictions. For example, you could put some code in a lesson, and then:

reflect.

    internal_name=reflect_code_prediction



    What output would that code produce?



/reflect.

You could have a series of these in a worked example.

Multiple-choice and fill-in-the-blank

Another way to implement the research finding about predictions is with multiple-choice and fill-in-the-blank questions. MCQs give more scaffolding than FiBs, so maybe start with MCQs and switch to FiBs.

Skilling helps you make good skills courses. You can take something effective from the literature (making predictions, in this case) and use Skilling's tools to make that happen.

Are there other findings from the literature that you'd like to implement? Please add a comment below. Maybe that can be added to Skilling.