Visualisation in R Post-mortem

28 March 2018

I ran this workshop more interactively than others: with me jockeying at the R console, writing bits of code to introduce ggplot concepts, and asking the group questions as I went, and getting them to offer suggestions and guess as to the outcome of commands as we organically worked towards preparing a plot with a single dplyr/ggplot pipeline.

My plan going in was:

  1. Review the core dlyr verbs: filter(), mutate(), group_by(), summarize()
  2. Explain the basic structure of a ggplot command: ggplot(aes()) + layer + layer + …
  3. Demonstrate more “advanced” features of ggplot, like facetting, which required rearranging our dataset with gather and spread.

Here are my take-aways from this session:

  • People enjoyed this session. I think it was a combination of the change of pace, and general exhaustion at this point in the year meant it was nice to sit back and participate but not have to bang your head against the wall trying to learn something from scratch.
  • I spent far longer reviewing dplyr than I expected. It had been about a month since our session on Data Wrangling workshop and I think most people had forgotten just about everything.
  • Introducing a ggplot command veeerrrrry slowly by building it up from parts was effective. I started with just data %>% ggplot(), introduced aesthetics and got everyone to appearance of axes, and then introduced different geoms and other layers and styling commands.
  • gather and spread are mysterious and seem like magic… I don’t think people came away from this session feeling like they understood the syntax, but I do think they have some understanding of what these commands do and why they are necessary for ggplot. Maybe. I’ll make sure to demonstrate these commands again in other lessons.
  • At the 1.5 hour mark, people were toast. No one went on to do the DataCamp, or Carpentry lessons. I think people are just tired.

In short: the interactive call-and-response lesson format was well-received, but I’m not sure how much in the way of concrete skills people have left with.