Whatever really happened, I'm delighted to see what I believe is the first retraction at the Restud, and perhaps even at a top 5 journal?
REStud retraction
-
It's true an inadvertent coding error can ruin a paper. But with three different datasets where presumably the variables were different and had to be coded separately?
And as others have mentioned, anyone download two of the datasets, so how can they be lost?
And finally, who tried to replicate but couldn't which led to the retraction?
Anyone got the goss?
-
This is great. Hope to see retraction becoming common in economics journals. But what about the 802 citations, according to google scholar? I suspect that the paper has shifted the received wisdom and it won't shift back easily.
Unlikely that this cute result changed any minds.
-
Meanwhile PG makes north of 450k a year …
Interesting it's no longer on her CV. And all of the sudden in the 17 years between PhD graduation and 2020 she has only 1 top 5, with Nunn and Alesina... Ao you're telling me the restud 2014 was not pivotal for her tenure?
P-hackers at Anderson...
-
That's the main problem here. Anderson would never have tenured someone with one top 5 as RA of Alesina and Nunn.
Meanwhile PG makes north of 450k a year …
Interesting it's no longer on her CV. And all of the sudden in the 17 years between PhD graduation and 2020 she has only 1 top 5, with Nunn and Alesina... Ao you're telling me the restud 2014 was not pivotal for her tenure?
P-hackers at Anderson... -
https://academic.oup.com/restud/advance-article/doi/10.1093/restud/rdac085/6982752?login=false
This is a retraction of: Paola Giuliano, Antonio Spilimbergo, Growing up in a Recession, The Review of Economic Studies, Volume 81, Issue 2, April 2014, Pages 787–817, https://doi.org/10.1093/restud/rdt040
The authors and editorial team are retracting this article because the original findings cannot be replicated, likely as a result of an inadvertent coding error. While the original codes and data sets are no longer available, new analysis with a markedly similar data set does not support the original results.I like the reasoning. It might really be simply a coding error rather than deliberate fraud.
"original codes and data sets are no longer available"
This is not "simply a coding error". It is either fraud or incompetence. Either way, good that it is retracted.