andrew66882011
commented
6 years ago Good catches!
Comments on the three issues you reported:
-
It's important to create a folder ("data" here but could be named something else - point out to learners) before using the function download.file() to download a data file because for most R novices, this is not obvious and it's also a good habit to save raw (original) data in a different folder than for temporary data generated during a R session as mentioned later in the lesson.
-
It's important to tell learners that there are two ways to specify "aesthetics" (color, size, etc): a variable in the data and a specific value. It might be a good idea to mention (not necessarily go through) the site: https://cran.r-project.org/web/packages/ggplot2/vignettes/ggplot2-specs.html so that learners can get a good resource (reference) when needed.
-
It's important to learn how to transform a data between "long" and "wide" forms (reshaping), especially for people doing longitudinal data analyses (time series analyses, panel data analyses, repeat measures, and the like. Combining with visualization might reinforce learners' impression on the effects and differences but its importance make it worth an individual section.
Jianjun
anacost
tobyhodges
I've found three small issues in this lesson:
Issue 1 - creation of data folder
In the lesson "Starting with data" the download.file() function is used to download the CSV file that contain the survey data.
According to the code, the destination of the file is the "/data" folder that has never been created so the student will most likely have an error. I would suggest to create the folder first, using the dir.create() function.
dir.create("data", showWarnings = FALSE)
download.file("https://ndownloader.figshare.com/files/2292169", "data/portal_data_joined.csv")
Issue 2 - denity plot
When we describe how to visualise data with ggplot2 we explain that "we need to define define a mapping (using the aesthetic (aes) function), by selecting the variables to be plotted and specifying how to present them in the graph, e.g. as x/y positions"
I think it would be nice to show them a plot wher only aestethic is needed (i.e. geom_density()). We can also show them how to add a line for the mean using the function geom_vline() as another example of adding geoms.
We first calculate the mean of the weight and we store it in an object surveys_mean -> surveys_complete %>% summarize(mean_weight = mean(weight, na.rm = TRUE))
ggplot(data = surveys_complete, mapping = aes(x = weight, color=sex))+geom_density()+ geom_vline(data=surveys_mean, aes(xintercept=mean_weight),linetype="dotted",size=1)
Issue 3 - reshaping data
I think that the section "Reshaping with gather and spread" in the Manipulating data frames lesson is a bit difficult to understand at first. Students don't really understand why we want to reshape the data. They don't see the point. It is much easier to understand the concept in the context of ggplot when they see that you need a column for each axes.
I would perhaps move this session to the "Visualising data" lesson. We could first spread the data and show them that in the wide format is not possible to use the genuses and the mean weight for a plot.
Alessandra