I like to make stuff. This page compiles some of the things that I’ve made that aren’t really worth splitting out on their own. A lot of them are just links to Twitter, where I tend to post the things I make.
A repository of ways to modify msmbstyle to get what The Effect has.
A warning about simplistic ways of “adjusting for trend”
A checklist to use whenever you’re preparing a data visualization.
A graph showing turnaround times to publication in economics by looking at NBER publication rates over time.
A slide on how to interpret the Greek and Latin (English) letters in statistics.
Videos on data wrangling and cleaning in R tidyverse, R data.table, and Python pandas.
A flowchart for how to take a bunch of variables and make a regression out of them.
How to perform a power analysis using simulation in Stata.
A blog post I helped run numbers on showing how the Coronavirus outbreak affected foot traffic at airports.
A guide to how to use and understand the purpose of robustness tests.
Workshop slides about how R can be used in an econometrics classroom, focusing on data manipulation and getting familiar with the language, targeted at economics faculty who may already be familiar with other statistics packages.
Slides from a workshop about teaching research design, econometrics, and applied economics to undergraduates with causal diagrams.
Slides from a workshop about teaching ECON 305, a new causality class designed to go before the rest of econometrics.
A review of the Bryan Caplan book “A Case Against Education.”
An overview of what it’s like working as a statistical consultant, and how people tend to view the purpose of statistics.
A walkthrough of the concept of how instrumental variables estimation can be biased by the process of only performing the analysis if the instrument is strong enough.
A thread on job market advice aimed at those who are going for middle-of-the-road positions rather than R1s.
A graph showing why it’s not a great idea to use R squared to pick your model.
A set of graphs showing why it’s not a great idea to selectively report (or use) your results based on statistical significance, and how this means that in low-power situations, a significant result is very likely to be wrong!
A very preliminary look at how student debt burden and earnings vary by degree type/major.
What else do you need to know?
Data from the College Scorecard 2016-17 with hover-over information.
A graph showing how bad collider bias can get.
Ever wonder what kinds of numbers the lyrics of pop music contain?
Ever wonder how often non-English language songs make the Billboard Hot 100?
Ever wonder how good pop music is?
We sure get a lotta old guys these days.
Where does the money come from, and where does it go?
Some Econ 201 thinking about why there are so many darn TV shows these days.
A graph displaying statistical power for a test in which only a subsample responds to treatment.
In reference to a legendary StackExchange post, what does it actually mean if you sort x and y independently before regressing one on the other?
Graphs of actor names and the kinds of terms that pop up in different movie genres.
When you Google nonsense, what sites pop up?
How happy is EconTwitter? And is happiness related to instructor effectiveness? (no)
What should an economics grad student spend their time figuring out how to do? Other than, like, research and stuff.
