-hacking or publication bias, in particular for articles using DID or IV as causal identification method. We show that a crucial … then find no evidence for p-hacking or publication bias in the DID and pooled data, but evidence remains for IV …

make no difference. However, the small sample properties can differ. We explore small sample power and estimation bias both …) in the auxiliary model. We find that both power and bias are similar when the number of variables used is the same …. Raising the number of variables lowers the bias but may also raise the power unacceptably because it lowers the chances of …