wk06-TidyTuesday-Commentary-JK

[jkozak180]

TidyTuesday Commentary- Week 06

Code/Tools/Approaches we have seen in class:

Some similar features and/or code, outside of using the library(tidyverse) data package functions;

• Using ggplot() +, geom_boxplot()+ and ggsave.png CodeLink
• Using geom_point facet_wrap and stat_smooth CodeLink
• Using group_by and dplyr::summarise functions CodeLink
• Making plot points faded using alpha= CodeLink
• Using geom_jitter in combination with alpha CodeLink

Code/Tools/Approaches not seen within class:

Some things I’ve never heard of or sounds cool;

• Using an .Rmd file but exporting the output as a github_document: CodeLink
• The type=case_when function, followed by str_detect(code) CodeLink
• Adding fonts from Google!!!! font_add_google() CodeLink
• Displaying specified zoomed in sections of data on a graph as its own section CodeLink TweetLink

Data visualizations that were enjoyed:

1. I didn’t mind this plot by @Wilson_Jaz. It had the basic information that was easy enough to follow. As always, the aesthetic finishings like colour gradients and special fonts are always appreciated.
2. I thought this plot by @cnicault was really striking, but definitely not super wonderful for specificity. I liked how he represented the distributions of weight classes, which provided further insight to the plot on the left.

Data visualizations for improvement:

1. It was NikiThadani’s first #TidyTuesday but I thought this boxplot was ugly. Technically nothing too much wrong with it, its clear enough (despite not displaying all of the data from the dataset) but there is no year(?) on this figure. Maybe its not important or I’m pointing out something not incorrect, but I personally would have added a data range in the title somewhere. Like, pumpkin sizes from 1900-1950 might be very different from 2000-2021, or maybe its from 1900-2021; I don’t know, you can’t tell from this plot.
2. This… thing outputted by @shank4494 was a monstrosity. I hated hated hated everything about it. It was not intuitive AT ALL, took me WAY too long to read/figure out what the data was and/or what he was trying to display. Ew. The only thing that was okay was the ranking of overall heaviest weight by state. Get a real figure title for each data section instead of slapping all three titles in a row on the left, a legend of colours on the US map would be nice, these lines everywhere… figure out a better way of connecting state name to map location (abbreviations on map), again find a better way of connecting the ranked state data to the distribution section. Horrendous. HATE IT.
3. This boxplot figure outputted by Julia Silge, posted by @DataPosts, was very pretty but my main beef with it was that it was missing a title. So, I technically don’t know what you’re plotting (pumpkin weight?), or what year(s) the data is from etc. Also, just for consistency’s sake I would appreciate a title on the x-axis.