Read on Twitter
OK y& #39;all, it is now my pleasure to show off some of the truly, genuinely heinous plots students in my Reproducible Data Analysis class made.
Content warning: these plots are f***ing awful.
Content warning: these plots are f***ing awful.
First up, big congratulations to @UTK_EPS Ph.D. student Rihanna Moore, who won by popular vote with this monstrosity
To commemorate the victory, we took an awkward photo of the trophy presentation / handshake. Please appreciate the poor composition and bad lighting.
But the runners-up were amazing too.
Really, everyone was a loser.
Really, everyone was a loser.
This student couldn& #39;t be here to present, so I have no idea WTF is going on here. But I recognize that the data (or, at least, the Substrate categories, notwithstanding "black") come from my lab.
Some VERY strong chartjunk game here. Also props for the fact that the underlying data are real, and pretty interesting.
You can see the data here - three vertical lines, just barely visible against the white background - if you look very, very carefully.
I& #39;m very pleased with the use of one data point per facet here.
I want to highlight the title here, which I find to be subtle yet compelling with its badness
This, I think, wins the prize for being a terrible plot most like one that I might see in real life. There are like two dozen categories, separated by color, which exist in vastly different numbers, and of which you can only see like 3.
You can get a lot of mileage out of
theme(text = element_text(size = 1))
theme(text = element_text(size = 1))
I *think* the horizontal axis is a classic example of factor levels causing categories to be ordered in a way that makes no sense, but the student and I talked for some time about it, and neither of us is totally sure what is happening there.
Polar coordinates: not even once.
Lotta badness here, but note that the color is mapped to a question about gender (Question 29), which is coded as a continuous variable whose values are either 1 or 2.
Fun fact: earthquake magnitudes are continuous! You wouldn& #39;t know it from looking at this plot.
Again, a really superb title here, among other problems.
Bet you& #39;ve never seen the diamonds data set plotted like this before.
(You may also never have seen log-base-pi-transformed data, in which case, you& #39;re welcome).
(You may also never have seen log-base-pi-transformed data, in which case, you& #39;re welcome).
Remember everyone: stat_smooth() calculates error bands, and you should always display them because they are very real and always definitely mean something.
Water_type: no.
