1 Introduction

My goal in putting this resource is to focus on the practical tools I use and the challenges I (David Reinstein) face. But I am open to collaboration with others on this. (Other contributors so far include Gerhard Riener, Oska Fentem, and Scott Dickerson.)


My focus: Microeconomics, behavioral economics, focus on charitable giving and ‘returns to education’ type of straightforward problems. (Minimal focus on structural approaches.)

What I care about: Where we can add value to real econometric (and statistical, experimental, survey design, and data science) practice?


The data I focus on:

  • Observational (esp. web-scraped and API data and national surveys/admin data)

  • Experimental: esp. with multiple crossed arms, and where the ‘cleanest design’ may not be possible


I will assume familiarity with most basic statistical concepts like ‘bias,’ ‘consistency,’ and ‘null hypothesis testing.’ However, I will focus on some concepts that seem to often be misunderstood and mis-applied, and I will give and link definitions as time permits.

If you are involved with this project, you can find a brief guide (somewhat WIP) on how to add content (HERE)[https://daaronr.github.io/ea_giving_barriers/bookdown-appendix.html]. This is from a different project but the setup is basically the same.

Basic statistical approaches and frameworks

  • Bayesian vs. frequentist approaches

Folder: bayesian Notes: bayes_notes

  • Causal vs. descriptive; ‘treatment effects’ and the potential outcomes causal model
  • Theory, restrictions, and ‘structural vs reduced form’

Causal inference through observation{-#caus_inf_obs}

Causal paths and levels of aggregation

Experiments and surveys: design and analysis