I am Professor of Sociology at Duke University. (→ More about me.)
10 October 2020
Some of my Work
Data Visualization: A Practical Introduction. Princeton University Press. /
Buy on Amazon
“Fuck Nuance.” Sociological Theory 35:118-127. /
“Seeing Like a Market.” Socio-Economic Review, 15:9-29. /
“The Performativity of Networks.” European Journal of Sociology, 56:175–205. /
Last Best Gifts. University of Chicago Press. /
Buy on Amazon
8 January 2021
I don’t know what happened. But here’s my current theory of what the White House thought was going to happen. I don’t have any more information than you do, and here I’m not concerned with the broader question of how the country came to this end. I am just trying to make sense of what happened on Wednesday.
From the moment he knew he’d lost the presidential election, Trump absolutely wanted to get the result overturned.
Jurisdiction figures updated on October 25th with data up to week 37 (ending September 12th).
Updated again on December 4th to include data up to week 45.
Recent posts on the CDC data have been getting a fair amount of traffic. This page is an overview of the things I’ve been doing with the data.
Summarizing excess mortality in the United States this year so far An overview of mortality in the US in 2020
8 October 2020
Although yesterday’s excess deaths plots by cause graph was for the whole of the United States only, the table we made did the same calculations on the whole CDC dataset, so the resulting df_excess table has numbers for all U.S. states and several other jurisdictions, such as New York City.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 > df_excess ## # A tibble: 4,428 x 9 ## # Groups: jurisdiction, cause  ## jurisdiction cause year deaths baseline baseline_sd excess pct_excess pct_sd ## <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> ## 1 Alabama All Cause 2015 24261 25278 750.
6 October 2020
As I was saying the other day, calculating excess deaths can be a tricky business, especially if your focus is on understanding counterfactuals like how many people died of some cause who would not have died due to some other competing risk over the period of interest. Moreover, even setting the counterfactuals aside, the whole business of accurately counting and classifying deaths on the scale of a country as large and variegated as the United States is an enormous challenge in itself.
1 October 2020
The other day I was looking to make a bunch of graphs showing some recent data from the CDC about excess mortality due to COVID-19. The idea was to take weekly counts of deaths over the past few years, both overall and from various important causes, and then show how the weekly counts from this year compare so far. The United States has a very large population, which means that a fairly predictable number of people die each week.