What are some situations that can occur if you make a graph that shows summary statistics of your data (such as averages)
without showing the actual data? (For ideas, see the papers linked previously, as well as
the Bar Barplots campaign or
Anscombe's Quartet.)
In section 3 of Politzer-Ahles & Piccinini (2018), we discuss "inferential" vs. "communicative" functions of graphs. What is
a situation where you might want to create an "inferential" graph? What is a situation where you might want to create a
"communicative" graph?
Suppose you do a study in which you have 2 groups, with 10 people in each group. For a study like this, imagine at least
three different possible outcomes that would have the same barplot but different informative graphs. Draw graphs of each
of these possible outcomes. (You can use plotting software like R or Excel, or you can just draw by hand.)