Will just add to point about the decline in reporting null results: some evidence (in econ) that this problem is concentrated high-ranking journals (top-5s) and not elsewhere. Within sample of all registered RCTs on AEA site (2013-2016), we found reporting nulls in abstract. . .
Reposted from
Prashant Garg
While the (quasi-) experimental methods increase credibility of causal results, there are two related concerns that make studies hard to replicate:
1. decline in reporting of null results
2. increase in use of private sector data
a mini thread 🧵based on work with @trfetzer.com
1. decline in reporting of null results
2. increase in use of private sector data
a mini thread 🧵based on work with @trfetzer.com
Comments
@violaasri.bsky.social + Imai). Meaning our perception of most important or influential results may be particularly shaped by bias against nulls
https://aeaweb.org/conference/2025/program/1293