Important modifications to be applied the presentation of Workshop 8

[katherinehebert]

Before developing and modifying this workshop, please read through the presenter and developer protocol and refer back to it regularly as you work.

Summary of presenter and participant feedback (2020-2021)

This workshop received very positive feedback last year, thanks to the contributions of previous QCBS members to the construction and improvement of the presentation. Participants and presenters consistently felt that the workshop's structure was clear and easy to follow, that the workshop's content is well-suited to understand how GAMs work, how to implement them in R, and how interpret their outputs, which is the objective of this workshop. The feedback also notes that this workshop is an appropriate length for remote delivery. Consequently, the workshop would most benefit from enhancements, rather than corrections, as outlined below.

General issues to be addressed related to the presentation of this workshop

Adapt exercises to better suit the remote workshop format

Exercises should be informative and engaging, and should build on the material shown during the workshop. Here are some suggestions to achieve this goal:

• Review the quantity and structure of exercises. Use exercises efficiently, to avoid overcrowding a workshop with exercises that do not actively build on the concepts and tools shown in the workshop. We generally recommend implementing up to a maximum of 2 longer-form exercises in breakout rooms, to allow ample time to thoroughly go over challenge solutions. Of course, there may be exceptions, and we ultimately leave this decision to our contributors. When adapting or construction exercises, think about the time management required to run an effective exercise in a workshop setting. Please remember to leave time for students to work through their solutions and ask questions, and consider the additional time it takes to use breakout rooms.
• Ensure that breakout rooms are used effectively. If breakout rooms are used, provide or add prompts to stimulate collaboration and/or discussion between participants, such as questions to discuss about the challenge or specific elements to compare in the solutions developed by participants in the room. You may remove certain recommendations of breakout room use if you do not find them relevant, or propose breakout rooms if they would be beneficial.
• Construct or edit exercises to promote active learning. Refer to these guidelines to find suggestions of exercises promoting active learning. You may use additional resources to create the exercises.

Facilitate the presentation of the slides with presenter notes.

Presenter notes are particularly helpful to guide presenters through slides that have been developed by many QCBS contributors over several years. For example, you may add notes to direct presenters to the key message of a complex slide, to remind presenters to spend more (or less) time on a difficult concept, to offer additional information or explanations, to propose check-in questions to seek more active engagement of the participants, to remind presenters to call back to a previous concept in an earlier workshop, and more. To add presenter notes, you can add ??? at the bottom of the slide, and write your presenter note underneath as such:

```
# Slide title: Coding in R is fun!

R allows us to clean, analyze, and plot our data!

???
Take the time to explain that R allows us to program tasks, so we do not have to manually repeat them. 
To engage the participants, ask them if they think R is fun too.
```

• Annotate slides with presenter notes, to help presenters navigate the previously developed workshop material more fluidly.

Remaining issues to resolve from previous contributors:

• Mixed models using real-world data Mixed models section with real-world data #17
• Plotting of smooths for generalized and mixed models Plotting smooths (package "itsadug") #15

View the comments and contributions implemented during the 2020-2021 QCBS R Workshop Series here: #4

Specific to slide content, structure, grammar and style

• Demonstrating the package gratia for model visualisation and checking could be beneficial, particularly for people who prefer working with ggplot2. For example, we can make helpful diagnostic plots with the function appraise() and plot models with draw(). We can also easily pull out the EDF with edf(), rather than k.check() as is shown in the presentation.
• Slide 19 (EN, FR): Overflow of the code output beyond the slide boundaries.
• Slide 44 (EN, FR): The content is very close to the bottom edge of the slide. Making the plots slightly smaller would remedy this.
• Slide 46 (EN, FR): Same as above.
• Slide 50 (EN, FR): Overflow of the code output beyond the slide boundaries.

[katherinehebert]

Slide overflow issues resolved in commit 86cee01

[katherinehebert]

Unfortunately a lot of the Winter 2022 development time for this workshop was needed to create the book (#20), and match the content between the book and the presentations. Though I did not modify the challenges, I do think this workshop is well-structured (only 3 challenges), so I would probably not change this structure much, although I'm sure there are many ways to make the exercises even more pedagogical and interactive. Breakout rooms are also not mentioned in any prompts, leaving the decision to use breakout rooms up to the presenters during the workshops, which I think can be great because this is more flexible. Final edits were completed in 1ec4a52.

The remaining issues listed here will need to be addressed in future development sessions. Leaving this open for future contributors!