Use this after an EctJ decision letter. Reopen the decision letter, RES review-process page, and current author instructions before setting deadlines.
Classify each requested change before writing:
The response letter should show that the leading case is now stronger, not just longer.
For every major comment, attach exactly one evidence object: theorem change, proof fix, simulation result, application table, compression decision, or replication-file update. If a response has no evidence object, it is probably only reassurance and should be strengthened before resubmission.
| Referee pushback | What it signals at EctJ | Fix that closes it |
|---|---|---|
| "Asymptotic results carry no finite-sample evidence" | Theory-simulation contract broken | Add the one coverage or size panel that pairs with the contested theorem, summarized within the main-text page norm |
| "Simulation DGPs are detached from the empirical illustration" | Applied-value claim looks decorative | Re-anchor one DGP to parameters estimated from the application data and say so in the design preamble |
| "The leading case is too special to matter" | Scope guardrail read as a defect | Show the leading case covers the application class; add a remark stating what an extension would require |
| "Why is this proof only in the online appendix?" | Proof-placement conformance risk | Move the derivation into the printed appendix and re-verify the 20-page budget |
| "Code did not reproduce Table 3" | Replication credibility, decisive after conditional acceptance | Rerun the package, fix seeds or versions, cite the exact script in the response |
Illustrative numbers: a referee writes that the proposed test's size is untrustworthy with few clusters. A strong EctJ response does four things in four sentences: concede the gap ("the original design only simulated 30 or more clusters"), report new evidence ("at 8 clusters the test rejects 5.6% at nominal 5%, versus 12.9% for the incumbent; new Table 2, rows 1-2"), tie it to the theorem ("consistent with Proposition 2's symmetric leading case"), and protect the page budget ("the full grid moved to Supplement S3"). Evidence, theorem, location — nothing else.
[Decision] R&R / conditional acceptance / reject-resubmit / other
[Revision thesis] <one sentence>
[Major issues] <assumptions/proofs/simulations/application/replication>
[Response draft] <point-by-point structure>
[Compression guardrail] <what must not grow>