26 September 2020
Following up on yesterday’s post on within-state variation in deaths in the United States, here’s a quick look at all-cause mortality rates across twenty countries, courtesy of the excellent work of the demographers who maintain the Human Mortality Database. The panels show death rates across twenty countries. Within each panel you can compare the overall death rate for the first thirty weeks of 2020 (shown in red) with the death rates for each of the previous five years (shown in gray).
24 September 2020
I’ve redrawn the graphs here to add more information about COVID-19 deaths specifically. This post is getting a substantial amount of traffic and some of the feedback I’ve gotten suggests people were confused about what exactly was being shown. The original graphs were drawn from this CDC dataset, which led some readers to think that there was some undercounting happening. In the new versions, I’ve used a merged version of the 2014-2018 data and the ongoing 2019-2020 counts.
I had a very nice chat recently about data visualization with Brian Fannin, a research actuary with the
CAS. We covered a variety of topics from R and ggplot in particular, to how to think about data visualization in general, and what the dataviz community is learning from COVID. You can watch it here:
25 August 2020
If you’re teaching statistics, data analysis, or data visualization with R this semester, especially in the social sciences, I’ve pulled together various bits of data into packages that I use in my own teaching. You might find them useful once you’re sick of Gapminder. They cover a variety of topics and range from single tables of data to whole longitudinal and panel surveys.
The cavax package contains a school-level table of rates of Personal Belief Exemptions (PBEs) in California kindergartens for the 2014-15 school year.
3 June 2020
The wave of protest and unrest in the wake of George Floyd’s killing by the police shows little sign of abating just yet. Unrest nationwide is, if anything, increasing as protesters are met with repression by the police. Civil unrest of this scope is unusual. The conjunction of mass protest and widespread disorder should be worrying to those in authority.
When property damage and theft happens as a side-effect of real mass protest, authorities in a democracy cannot baton, tear gas, or shoot their way to legitimacy.