blank

Descriptive and causal inference; survey experiments and experiments in surveys

Causal concepts

Continuous treatments

Regressions of Y on X don't target the ATE

I saw your RCT and I have some questions

A helper for practitioners and others reading analyses of RCTs.

Global vaccination letdown (in German)

On the failure to share vaccinations

Democracy and Climate Action

I enjoyed following the discussion between Ross Mittiga and Alexander Wuttke last week on climate action and authoritarian policies (threads:paper). Both Wuttke and Mittiga engaged constructuively on an issue they care a lot about and the discussion clarified many tricky points. It’s exactly the kind of cross subfield discussion between theorists and empirical researchers that we need more of…

Ambiguous implications of collider bias for the estimation of discrimination

Discrimination Estimands

Conditioning on a collider (a bit technical)

An instrument does not have to be exogenous to be consistent

We often think of an instrumental variable as a random shock that generates exogenous variation in a treatment of interest. But surprisingly if effects are constant, the instrumental variables estimator can be consistent even when the relationship between the instrument and the endogenous variable is confounded

CDD what is it good for?

Contribution to discussion of the benefits of CDD/CDR.