Follow and learn

Quantitude

This is a must listen! The Quantitude podcast co-hosted and masterfully produced by Patrick Curran (UNC@CH) and Greg Hancock (UMD) is proof that great comedy can also be done sitting down and despite mastering only a bad pirate accent and being of Transylvanian lineage (😎). And it’s about quantitative methods and academia. The dudes are seriously flawed, though; they’re likelihoodists who get a kick out of calling up Bayesians at 5:40 in the morning asking them to explain themselves.

Recent pods

Data Colada

Data Colada – “thinking about evidence and vice versa” – is the meta-science blog of behavioural scientists Uri Simonsohn, Joe Simmons and Leif Nelson. Running since 2013, it has become required reading for anyone serious about research integrity in the social sciences. Posts range from the intuition behind robust standard errors and the misuse of common statistical defaults, to forensic re-analyses that have uncovered fabricated data in high-profile published work. The tone is patient, generous and quietly devastating.

Recent posts

Data Is Plural

Data Is Plural is a long-running newsletter curated by Jeremy Singer-Vine, currently a data editor for the New York Times (previously at BuzzFeed, The Wall Street Journal and Slate). It’s a good idea to subscribe to the newsletter to receive a list of interesting and useful publicly available datasets every once in a while, which are also added to a Google Docs spreadsheet containing all the datasets listed in past editions of the newsletter. It also has an associated podcast, although that has been quiet recently.

Recent newsletters

Andrew Gelman et al.

You know this one. Running since 2004, the Statistical Modeling, Causal Inference, and Social Science blog coordinated by statistician and political scientist Andrew Gelman (Columbia) - but also contributed to by an increasing number of his collaborators - functions like a statistics consultancy desk for the world. Andrew is probably the most influential voice for pragmatic Bayesianism in the social sciences, and a prolific textbook author who has gifted treasures to quantitative educators worldwide.

Recent posts

Richard McElreath

Richard is best known outside the field of evolutionary anthropology for his foundational Bayesian textbook Statistical Rethinking and related YouTube course lectures. Currently directing a Max Planck Institute in Leipzig, his blog Elements of Evolutionary Anthropology is not extremely active, but it makes up for that with the quality of the thoughts conveyed in each post.

Recent posts

Learning Bayesian Statistics

Hosted by Alexandre Andorra, a Bayesian modeller at PyMC Labs, the Learning Bayesian Statistics podcast focuses on long-form conversations with various people who have contributed to Bayesian modelling in various academic fields and industries. Some recent episodes have a hybrid workshop format with a videocast available on the series’ YouTube channel.

Recent pods

Andrew Heiss

Andrew Heiss is a political scientist and public policy researcher at GSU. He has the strongest explicit commitment to making all aspects of his work open and transparent that you’ll find anywhere on the internet today. His blog dives deeply into various technical aspects of social science programming and it’s a wonderful resource for learning about coding in R.

Recent posts

Kieran Healy

Fellow sociologist Kieran Healy (Duke) is a prolific author on both substantive and technical topics. He has written on almost everything from organ procurement, through the institutionalization of mass shootings as a ritual of American childhood, to the problems of digital capitalism. His Plain Person’s Guide to Plain Text Social Science has been a terrific resource for those transitioning to computationally reproducible research workflows, and he is best known in the R social science community for his book on Data Visualization. His blog is just as diverse and interesting.

Recent posts

Scott Cunningham

Scott Cunningham is an economist at Baylor whose Causal Inference: The Mixtape has done more than any other recent textbook to make modern identification strategies legible to applied researchers. His Substack runs in parallel with the Mixtape Sessions workshops and the Causal Inference podcast, mixing long-form essays on diff-in-diff, synthetic controls and event studies with episodes from the podcast and – more recently – field notes on how AI tools are reshaping empirical research practice.

Feed unavailable.

Frank Harrell

Frank Harrell is Professor of Biostatistics at Vanderbilt University and is known in the R community for his popular R packages. His blog Statistical Thinking has been going since 2017 and contains very useful posts on various statistical tests and models implemented in R.

Recent posts

Danielle Navarro

I first encountered Danielle’s work as the author of a very lucid online textbook on Learning Statistics with R from back when she was a lecturer in quantitative psychology at the University of Adelaide. She has since moved into industry and has developed many cool interests in computational methods (including generative art). Her data science blog Notes from a data witch is a treasure trove.

Recent posts

The Test Set

The Test Set is Posit’s podcast, hosted by Michael Chow with regular appearances from Wes McKinney and Hadley Wickham. Each episode is a long-form conversation with someone shaping the day-to-day of data science and analytics engineering – the kind of unglamorous but consequential topics (SQL, reactive notebooks, widgets, the half-life of tooling fads) that working data practitioners actually argue about. A useful counterweight to the more academic podcasts on this list.

Recent pods

Migration Policy Centre

The Migration Policy Centre at the European University Institute in Florence is one of Europe’s leading academic hubs for research on international migration, asylum and mobility, and the natural data-and-policy companion to its sister observatory GLOBALCIT. Posts on the MPC blog draw heavily on the centre’s primary-data projects and comparative European fieldwork – covering irregular migration in the UK care sector, temporary-labour integration regimes, employer demand for undocumented workers, drivers of refugee return, and the conditions that produce irregularity in the first place. Multi-author, policy-literate, and consistently active.

Recent posts

GLOBALCIT

GLOBALCIT is the Global Citizenship Observatory based at the European University Institute in Florence. Its blog is the place to follow what is happening in citizenship and statelessness law across the world – from citizenship deprivation in the UK and EU free movement debates, through the unwinding of colonial citizenship regimes, to climate-related proposals for new forms of political membership. Multi-author, academic in register but accessible, and indispensable for anyone whose research touches naturalisation, dual citizenship or denizenship.

Recent posts

Migration Information Source

The Migration Information Source is the flagship publication of the Migration Policy Institute (MPI) in Washington DC. It is to migration what Data Is Plural is to datasets: a steady, reliable feed of country profiles, policy briefs and data-driven feature articles that cover origin, transit and destination contexts in roughly equal measure. The writing is policy-literate without being parochial, and the depth of comparative coverage is hard to find anywhere else on the open web.

Recent posts