Just recorded all my lectures for Quantitative Social Science Methods, I -- the first course in the Harvard Government Dept graduate methods sequence -- and am making them available publicly here: youtube.com/playlist?list=… I hope you find them useful for your course, or yourself
Gary King
5,235 posts
Harvard Professor (social scientist, statistician). Co-founder Crimson Hexagon (now Brandwatch), Learning Catalytics, Perusall, Thresher (now Two Six Tech),...
- Cleaning lady in elevator at APSA says to me "What's with you political scientists?" I say, "?". She explains: "The psychologists come here & its sex sex sex sex. The anthropologists checkin & its drugs drugs drugs drugs. But you political scientists. All you do is dress badly"
- Here's an optimistic thought. After 9/11, we spent trillions of dollars trying to stop terrorism. After this public health crisis, we'll likely spend a fortune on a problem that is easier to solve. If we do, this may well be the last pandemic the world ever sees.
- I'm teaching the first course in Harvard's graduate sequence in political methodology this Fall. Even if you're not at Harvard, you're welcome take the course with us for credit, or use our lectures and other materials if you're teaching it. See j.mp/G2001
- Political scientists know that debates rarely have any measurable effect on voter behavior, but at least they are usually entertaining. This one is excruciating.
- The big transition in graduate school: Every paper you wrote since 3rd grade was for ONE person paid to read it. Now all you have is a large, diverse, amorphous target audience with no reason to read your paper -- unless you figure out how to write it so they have no other choice
- Yesterday, I made public my Quantitative Social Science Methods class lectures. Instead of watching these on YouTube, Harvard students will use our free platform Perusall.com to watch, annotate videos, & interact. To do this in your course: Use code 6AFEYRZH4P
- Did you use Clarify for Stata but would like to do it in R? Or maybe you used {Zelig} for R (before it retired)? Welcome to our new R package {clarify}: Software for Interpreting and Presenting Statistical Results
- New Paper: We show that about half of conjoint survey datasets is measurement error, which can bias estimates in any direction by any amount. We also provide an easy way to correct the bias: GaryKing.org/conjointE w/@katie_clayton14 @YusakuHoriuchi @aaronrkaufman & @MayyaKomis
- Medical & public health folks: enough with the odds ratios already! Real people don't understand them; no one who wants to communicate uses them. There's also no advantage, even in case control studies. Use risk ratios, probabilities, predictions, etc. gking.harvard.edu/files/abs/1s-a…
- 1/3 After making my Harvard class lecture videos public, I'm now starting to release talks on a wide variety of research projects. Here's a playlist with the first 2. I hope you find them useful for your course or yourself.
- I recently made publicly available the lectures from my Harvard class, Quantitative Social Science Methods, I. After lots of requests for info on my video technology & various pedagogical issues, I wrote this summary. Suggestions welcome, updates likely:



