The Effect: An Introduction to Research Design and Causality is about research design, specifically concerning research that uses observational data to make a causal inference. It is separated into two halves, each with different approaches to that subject. The first half goes through the concepts of causality, with very little in the way of estimation. It introduces the concept of identification thoroughly and clearly and discusses it as a process of trying to isolate variation that has a causal interpretation. Subjects (...) include heavy emphasis on data-generating processes and causal diagrams. Concepts are demonstrated with a heavy emphasis on graphical intuition and the question of what we do to data. When we "add a control variable" what does that actually do? (shrink)

Structural Causal Modeling (SCM) is an approach to causal inference closely associated with Judea Pearl and given an accessible instroduction in [Pearl, J., & Mackenzie, D. (2018). The book of why: The new science of cause and effect. Basic Books]. It is highly popular outside of economics, but has seen relatively little application within it. This paper briefly introduces the main concepts of SCM through the lens of whether applied economists are likely to find marginal benefit in these methods beyond (...) standard economic approaches to causal inference. The most promising areas are those where SCM's causal diagrams alone offer significant value: covariate selection, the development of placebo tests, causal discovery, and identification in complex models. (shrink)

In Instrumental Variables (IV) estimation, the effect of an instrument on an endogenous variable may vary across the sample. In this case, IV produces a local average treatment effect (LATE), and if monotonicity does not hold, then no effect of interest is identified. In this paper, I calculate the weighted average of treatment effects that is identified under general first-stage effect heterogeneity, which is generally not the average treatment effect among those affected by the instrument. I then describe a simple (...) set of data-driven approaches to modeling variation in the effect of the instrument. These approaches identify a Super-Local Average Treatment Effect (SLATE) that weights treatment effects by the corresponding instrument effect more heavily than LATE. Even when first-stage heterogeneity is poorly modeled, these approaches considerably reduce the impact of small-sample bias compared to standard IV and unbiased weak-instrument IV methods, and can also make results more robust to violations of monotonicity. In application to a published study with a strong instrument, the preferred approach reduces error by about 19% in small (N ≈ 1, 000) subsamples, and by about 13% in larger (N ≈ 33, 000) subsamples. (shrink)