PhDs awarded in selected disciplines, 2006-2016.
Thierry Rossier asked me for the code to produce plots like the one above. The data come from the Survey of Earned Doctorates, a very useful resource for tracking trends in PhDs awarded in the United States. The plot is made with geom_line() and geom_label_repel(). The trick, if it can be dignified with that term, is to use geom_label_repel() on a subset of the data that contains the last year of observations only.
I was playing around with the gganimate package this morning and thought I’d make a little animation showing a favorite finding about the distribution of baby names in the United States. This is the fact—I think first noticed by Laura Wattenberg, of the Baby Name Voyager—that there has been a sharp, relatively recent rise in boys’ names ending in the letter ‘n’, at the expense of names with ‘e’, ‘l’, and ‘y’ endings.
The data from the 2018 wave of the General Social Survey was released during the week, leading to a flurry of graphs showing various trends. The GSS is one of the most important sources of information on various aspects of U.S. society. One of the best things about it is that the data is freely available for more than forty years worth of surveys. Here I’ll walk through my own quick look at the data, in order to show how R can tidily manage data from a complex survey.
I’ve gotten a couple of reports from people having trouble installing the development version of the socviz library that’s meant to be used with Data Visualization: A Practical Introduction. As best as I can tell, the difficulties are being caused by GitHub’s rate limits. The symptom is that, after installing the tidyverse and devtools libraries, you try install_github("kjhealy/socviz") and get an error something like this:
Error in utils::download.file(url, path, method = download_method(), quiet = quiet(): cannot open URL https://api.
A few years ago I wrote a post about the stickiness of college and university rankings in the United States. It’s been doing the rounds again, so I thought I’d revisit it and redraw a few of the graphs I made then.
In 1911, Kendric Babcock made an effort to rank US Universities and Colleges. In his report, Babcock divided schools into four Classes, beginning with Class I:
The better sort of school.
I was asked for some examples of posters I’ve made using R and ggplot. Here are four. Some of these are done from start to finish in R, others involved some post-processing in Illustrator, usually to adjust some typographical elements or add text in a sidebar. I’ve linked to a PDF of each one, along with a pointer to the original post about the graphic.
If you’re interested in learning more about how to making graphs and charts using R and ggplot, then by a staggering coincidence there’s a new visualization book out that can help you with that.
I am stuck at home sick today, so I decided to provide a relational analysis of the Stats Package Wars that have been bubbling away for the past week.
True in all its details.
If you want something slightly more constructive, consider Data Visualization: A Practical Introduction, or The Plain Person’s Guide to Plain-Text Social Science.
Chapter 2 of Data Visualization walks you through setting up an R Project, and takes advantage of R Studio’s support for RMarkdown templates. That is, once you’ve created your project in R Studio, can choose File > New File > R Markdown, like this:
Select R Markdown …
And then choose “From Template” on the left side of the dialog box that pops up, and select the “Data Visualization Notes” option on the right:
Based on the heatmaps I drew earlier this month, I made a poster of two centuries of data on mortality rates in France for males and females. It turned out reasonably well, I think. I will probably get it blown up to a nice large size and put it up on the wall. I’ve had very good results with PhD Posters for work like this over the years, by the way.
Data Visualization: A Practical Introduction will begin shipping next week. I’ve written an R package that contains datasets, functions, and a course packet to go along with the book. The socviz package contains about twenty five datasets and a number of utility and convenience functions. The datasets range in size from things with just a few rows (used for purely illustrative purproses) to datasets with over 120,000 observations, for practicing with and exploring.