Skip to content

SKILL.md

Latest commit

 

History

History
79 lines (62 loc) · 5.22 KB

SKILL.md

File metadata and controls

79 lines (62 loc) · 5.22 KB
Raw
name ecj-rebuttal
description Use when a The Economic Journal (EJ) decision letter (R&R or reject-with-encouragement) has arrived and you need to draft the response-to-referees letter and revision plan. Structures the response and the diff; it does not redo the analysis (route back to ecj-identification / ecj-theory-model / ecj-robustness first).

R&R Response & Rebuttal (ecj-rebuttal)

When to trigger

  • An EJ R&R (or reject-with-encouragement-to-resubmit) letter has arrived
  • You have referee reports and need a point-by-point response plan
  • You are deciding which requests to implement, which to push back on, and how
  • The revision is drafted and you need the response letter to accompany it

Do the revisions before writing the letter. The letter documents changes already made; it should never promise work that is not in the revised manuscript.

Reading the decision letter

  • The editor's letter is the binding document (at EJ a handling editor coordinates the report; editor-in-chief Francesco Lippi, verified 2026-06-20). Identify which referee points the editor flags as essential vs. optional and prioritize accordingly.
  • Separate requests into: (a) must-do for the editor, (b) reasonable and cheap, (c) costly or wrong-headed. Negotiate (c) with evidence, not assertion.
  • An EJ editor and referees care most about the economic argument and broad interest: expect demands to strengthen the mechanism, broaden the relevance / bound the generality, or sharpen identification. Treat those as first-order.
  • Plan replication readiness now: acceptance triggers the EJ Data Editor's pre-acceptance reproducibility check and a Zenodo deposit (see ecj-replication-package).

Response-letter structure

  1. Cover note to the editor: thank them; summarize the 3–4 substantive changes; state how the paper is now stronger economically and more broadly relevant. One page.
  2. Point-by-point, per referee, in the referee's order:
    • Quote (or paraphrase) the comment.
    • State the change made, with the new text/exhibit and where it now lives (section, table, page).
    • Where you disagree, respond with evidence and economic reasoning — concede what is fair, hold the line where you are right, and never be defensive.
  3. A summary of changes mapping each major comment to the revised location, so the editor can verify at a glance.

Tone: respectful, concise, confident. Every response is either "done, here it is" or "we respectfully differ, and here is why." Avoid hedging and avoid over-promising future work. Because EJ review is single-blind, you already know the report is read by referees who see your identity — keep the tone collegial.

Handling the EJ-typical demands

  • "Broaden the relevance / it reads too narrow" → route to ecj-topic-selection / ecj-literature-positioning; make the general lesson and external relevance explicit, or bound the claim honestly.
  • "Strengthen the mechanism" → route to ecj-theory-model; add the model result or the discriminating test, then report it.
  • "Rule out alternative X" → route to ecj-robustness; add the discriminating test, summarize in text, table in appendix.
  • "Identification is not convincing" → route to ecj-identification; add diagnostics/falsification or reframe the claim to what the design supports.
  • "Improve the exposition" → route to ecj-writing-style; at EJ this is substantive, not cosmetic — take it seriously.
  • Conflicting referees → flag the conflict to the editor and explain the path you chose and why.

Checklist

  • Every referee point has an explicit response (none skipped)
  • Each "done" response points to the exact revised location
  • Disagreements backed by evidence/economic reasoning, framed respectfully
  • Editor's essential requests all addressed first
  • Broad-interest / mechanism / identification demands treated as first-order
  • Exposition improvements taken seriously (EJ bar)
  • Replication package on track for the EJ Data Editor if acceptance follows
  • Revisions are actually in the manuscript before the letter is written
  • Summary-of-changes table maps comments → locations
  • No new claim is over-promised without being delivered

Anti-patterns

  • Writing the response letter before doing the revisions
  • Skipping or merging referee points to seem responsive
  • Defensive or dismissive tone toward a fair criticism
  • Conceding a correct result because a referee pushed hard (defend it with evidence)
  • Promising "future work" in place of an analysis the editor asked for now
  • Dismissing an exposition request as cosmetic when at EJ it carries weight
  • Ignoring the editor's prioritization and treating all comments as equal

Output format

【Decision】R&R / reject-with-encouragement
【Editor's essential points】1... 2... 3...
【Per-referee plan】R1: [done / push-back + evidence] ; R2: ... ; R3: ...
【First-order (broad-interest/mechanism/ID)】how addressed
【Exposition】improvements made? [y/n]
【Revisions complete before letter】[y/n]
【Summary-of-changes table】drafted? [y/n]
【Replication readiness】code ready for EJ Data Editor on accept? [y/n]
【Next】resubmit via Editorial Express (ecj-submission for format re-check)