academic-review

分类: 调研与分析 | 上传者: MattArtzAnthro | 下载: 0 | 版本: v1.0（最新）
Use this skill whenever a user needs help writing, evaluating, or responding to peer reviews for anthropological research. Triggers include: any mention of "peer review," "review a manuscript," "write a review," "reviewer comments," "respond to reviewers," "rebuttal letter," "revision plan," "manuscript evaluation," "assess this paper," "reviewing for [journal name]," "R&R response," "how to review," "reviewer feedback," "revise and resubmit." Covers writing constructive peer reviews for anthropology journals, evaluating manuscripts from the reviewer's perspective, and responding to reviewer feedback (rebuttal letters, revision plans). Review types: invited review, desk review, blind review, open review. Do NOT use for grant review panels (use grant-proposal skill) or student work feedback (use teaching-materials skill). This skill handles peer review as a professional scholarly practice and revision as a strategic engagement with reviewer critique. --- # Academic Peer Review Write constructive peer reviews, evaluate manuscripts, and respond to reviewer feedback for anthropology journals. The skill treats peer review as a
更新日志: Source: GitHub https://github.com/MattArtzAnthro/AI-Anthropology-Toolkit
目录结构

当前层级: skills/academic-review/
📁 references/
- 📄 review-writing-guide.md 28.6 KB
- 📄 revision-response-guide.md 12.3 KB
📄 SKILL.md 19.7 KB
SKILL.md

---
name: academic-review
description: >
  Use this skill whenever a user needs help writing, evaluating, or responding
  to peer reviews for anthropological research. Triggers include: any mention
  of "peer review," "review a manuscript," "write a review," "reviewer
  comments," "respond to reviewers," "rebuttal letter," "revision plan,"
  "manuscript evaluation," "assess this paper," "reviewing for [journal name],"
  "R&R response," "how to review," "reviewer feedback," "revise and resubmit."
  Covers writing constructive peer reviews for anthropology journals, evaluating
  manuscripts from the reviewer's perspective, and responding to reviewer
  feedback (rebuttal letters, revision plans). Review types: invited review,
  desk review, blind review, open review. Do NOT use for grant review panels
  (use grant-proposal skill) or student work feedback (use teaching-materials
  skill). This skill handles peer review as a professional scholarly practice
  and revision as a strategic engagement with reviewer critique.
---

# Academic Peer Review

Write constructive peer reviews, evaluate manuscripts, and respond to reviewer
feedback for anthropology journals. The skill treats peer review as a
professional scholarly practice with ethical obligations: reviewers owe authors
a careful reading, specific feedback, and actionable suggestions; authors owe
reviewers genuine engagement with critique, even when disagreeing. Both sides
of the review process are arguments about how to make scholarship stronger.

A good review does three things simultaneously: (a) evaluates whether the
manuscript makes a defensible contribution to anthropological knowledge;
(b) identifies specific, actionable paths to improvement; and (c) treats the
author as a colleague whose intellectual project deserves respectful
engagement. A good revision response does three things: (a) demonstrates that
every reviewer point was read and considered; (b) shows evidence of substantive
change where change was warranted; and (c) provides reasoned justification
where the author disagrees, supported by evidence or argument.

## Quick Reference

| Task | Reference |
|------|-----------|
| Review structure, evaluation criteria, constructive framework, discipline-specific assessment | Read [references/review-writing-guide.md](references/review-writing-guide.md) |
| Rebuttal letters, point-by-point responses, revision plans, handling contradictory feedback | Read [references/revision-response-guide.md](references/revision-response-guide.md) |

## Workflow

### Step 1: Identify What the User Needs

Determine the entry point:

- **Writing a peer review from scratch.** The user has been invited to review
  a manuscript and needs help structuring a constructive evaluation. Load the
  review-writing-guide and work through the review anatomy section by section.
- **Evaluating a manuscript.** The user wants to assess a paper's strengths
  and weaknesses before drafting a formal review. Load the review-writing-guide
  for evaluation criteria and discipline-specific assessment guidance.
- **Responding to reviewer feedback (rebuttal letter).** The user has received
  an R&R decision and needs to write a point-by-point response. Load the
  revision-response-guide for rebuttal structure and tone calibration.
- **Creating a revision plan.** The user has reviewer comments and needs to
  organize what to change, what to defend, and how to prioritize. Load the
  revision-response-guide for the revision planning framework.
- **Handling contradictory reviewers.** The user has conflicting reviewer
  demands and needs a strategy for navigating them. Load the
  revision-response-guide for the contradictory feedback section.
- **Learning how to review.** The user (often a graduate student) is reviewing
  for the first time and needs guidance on expectations, ethics, and process.
  Load the review-writing-guide and emphasize the AAA ethics section and the
  constructive review framework.

### Step 2: Gather Context

Before generating any content, collect these inputs:

**Required:**
1. **Task type.** Is the user writing a review, responding to reviews, or
   planning revisions? This determines which reference to prioritize.
2. **Journal or venue.** Which journal is involved? AAA flagships (American
   Ethnologist, American Anthropologist, Cultural Anthropology), regional
   journals, interdisciplinary outlets, and open-access venues have different
   expectations for depth, theoretical engagement, and review length.
3. **Manuscript details.** What is the paper about? Subfield, methods,
   theoretical orientation, and approximate length help calibrate the review
   or response.

**Important but can be inferred:**
4. **Review type.** Invited review, desk review, blind review, or open review.
   Blind review requires attention to anonymity in the review text. Open review
   may allow more collegial tone.
5. **Reviewer expertise relative to manuscript.** Is the reviewer an expert in
   the subfield, a methodological specialist, or a generalist? This affects
   what the reviewer can and should comment on.
6. **Decision context (for responses).** What was the editorial decision?
   Minor revisions, major revisions/R&R, or conditional accept? This
   determines the scope and urgency of the response.

**Helpful but not required:**
- Career stage of reviewer or author (affects tone calibration)
- Whether the user has reviewed before (first-time reviewers need more
  scaffolding)
- Number of reviewers and degree of consensus or conflict
- Editor's letter and whether it signals which reviewer to prioritize
- Timeline for revision submission

### Step 3: Load Appropriate References

- **Always load** `references/review-writing-guide.md` when the user is
  writing or evaluating a review. This contains the review anatomy, evaluation
  criteria, constructive framework, and discipline-specific assessment guidance.
- **Always load** `references/revision-response-guide.md` when the user is
  responding to reviewer feedback, writing a rebuttal letter, or creating a
  revision plan. This contains point-by-point response format, tone
  calibration, and strategies for contradictory feedback.
- **Load both** when the user needs to understand the full review cycle — for
  example, a graduate student learning how peer review works from both sides.

### Step 4: Generate Content

Follow the appropriate framework from the loaded reference files:

**For reviews:**
1. **Summary paragraph.** Demonstrate that you read the paper carefully by
   summarizing the argument, methods, and contribution in 3-5 sentences. This
   is not a abstract restatement — it shows interpretive engagement.
2. **Contribution assessment.** What does this manuscript add to the field?
   Be specific: which conversation does it advance, what new evidence or
   framing does it offer?
3. **Strengths identification.** Name what works well before turning to
   critique. Specific praise is more useful than generic approval.
4. **Major issues.** Identify 2-4 substantive concerns that require revision.
   Each issue should name the problem, explain why it matters, and suggest a
   path forward. Separate conceptual problems from execution problems.
5. **Minor issues.** List smaller concerns (clarity, citation gaps, structural
   suggestions) that would improve the manuscript but are not revision-blocking.
6. **Overall recommendation.** Accept, minor revisions, major revisions/R&R,
   or reject — with a clear rationale tied to the issues identified above.

**For rebuttal letters and revision responses:**
1. **Cover letter to the editor.** Brief, professional summary of how the
   revision addresses the key concerns. Acknowledge the reviewers' labor.
2. **Point-by-point response.** For each reviewer comment: quote the comment,
   state the action taken, and cite the specific location of the change in the
   revised manuscript. For points of disagreement: acknowledge the concern,
   provide evidence or argument for the author's position, and explain any
   partial accommodations made.
3. **Revision summary table.** Optional but effective: a table mapping each
   major reviewer concern to the specific changes made and their locations.

**For revision plans:**
1. **Triage reviewer comments.** Categorize each point as: (a) agree and will
   change, (b) partially agree and will accommodate, (c) disagree and will
   defend with argument. Prioritize major issues over minor ones.
2. **Map changes to manuscript sections.** Identify which sections need
   revision and what kind (rewriting, adding, cutting, reorganizing).
3. **Timeline and sequencing.** Determine the order of revisions — usually
   address the largest structural changes first, then work through smaller
   modifications.

### Step 5: Generate Output

Produce one or more deliverables depending on user needs:

- **Full peer review.** Complete review following the six-part anatomy:
  summary, contribution, strengths, major issues, minor issues,
  recommendation. Calibrated to journal tier and review type.
- **Evaluation memo.** Structured assessment of a manuscript's strengths and
  weaknesses, organized by evaluation criteria (theory, evidence, methods,
  writing, contribution). Useful for pre-review assessment or reading group
  discussion.
- **Rebuttal letter.** Cover letter plus point-by-point response to all
  reviewer comments, with evidence of change or reasoned disagreement.
- **Revision plan.** Organized plan mapping reviewer comments to planned
  changes, with triage categories and timeline.
- **Review framework.** For first-time reviewers: a structured template with
  prompts for each section of the review, calibrated to the specific journal
  and manuscript type.

### Step 6: Quality Check

Before presenting output, verify:

- [ ] Every section of the review or response serves a clear function (no
      filler or repetition)
- [ ] Critiques are specific, actionable, and tied to evidence from the
      manuscript (not vague impressions)
- [ ] Positive contributions are acknowledged with specificity, not just
      generic praise
- [ ] Major and minor issues are clearly distinguished and hierarchically
      organized
- [ ] The tone is constructive and collegial — critiques address the work,
      not the author
- [ ] For blind reviews: no language reveals knowledge of author identity
- [ ] For rebuttal letters: every single reviewer point is addressed (none
      skipped)
- [ ] For rebuttal letters: disagreements are supported with evidence or
      argument, not defensive dismissal
- [ ] For revision plans: changes are prioritized by importance, not by ease
- [ ] The recommendation (if a review) follows logically from the issues
      identified — no mismatch between severity of critique and recommendation
- [ ] Discipline-specific evaluation criteria are applied: ethnographic
      evidence, theoretical argument, methodological rigor, ethical practice,
      positionality, and writing quality
- [ ] The output evaluates the manuscript as written, not a different paper
      the reviewer wishes had been written

## Parameters

- **Review type:** Invited review, desk review, blind review, open review.
  Determines anonymity requirements and tone conventions. Blind review requires
  no identity-revealing language; open review allows more collegial and
  conversational tone.
- **Journal tier:** AAA flagship (American Ethnologist, American
  Anthropologist, Cultural Anthropology), regional journal, interdisciplinary
  outlet, open-access venue. AAA flagships expect significant theoretical
  contribution and deep ethnographic engagement; regional journals value area
  expertise and engagement with local scholarship; interdisciplinary outlets
  need accessibility and cross-disciplinary relevance.
- **Reviewer stance:** Constructive (balanced assessment, improvement-oriented),
  developmental (mentoring tone, especially for early-career authors), critical
  (rigorous evaluation for top-tier venues with high standards). All stances
  require actionable feedback — the distinction is emphasis and tone, not
  whether critique is offered.
- **Output type:** Full peer review, evaluation memo, rebuttal letter, revision
  plan, review framework (for first-time reviewers).

## Guardrails

- **Constructive tone always.** Critiques must be actionable, never dismissive.
  "The argument is unclear" is not actionable; "The connection between the
  theoretical framework in section 2 and the ethnographic analysis in section
  4 could be strengthened by explicitly stating how concept X illuminates
  finding Y" is actionable. Every critique should suggest a path forward.
- **Distinguish fixable issues from fundamental conceptual problems.** A paper
  with a strong core argument but unclear structure is a different case from a
  paper with a fundamental evidentiary gap. Reviews should make this
  distinction explicit because it determines the recommendation.
- **Flag methodological concerns without overstepping reviewer expertise.** If
  the reviewer is not a methods specialist, concerns should be framed as
  questions ("Could the authors clarify how they addressed X?") rather than
  assertions ("The methodology is flawed"). Epistemic humility about the
  limits of reviewer expertise is a professional obligation.
- **Rebuttal letters must address every reviewer point.** Even when
  disagreeing, every point raised by every reviewer must receive a response.
  Skipping uncomfortable points is the most common rebuttal failure mode and
  editors notice immediately.
- **Do not write reviews that reveal knowledge of author identity in blind
  review.** Avoid references to the author's previous work by name, conference
  presentations, institutional affiliation, or other identifying information.
  If the reviewer knows who the author is, the review must still be written
  as if they do not.
- **Reviews should evaluate the manuscript, not prescribe a different project.**
  The reviewer's job is to assess whether the paper achieves what it sets out
  to do, not whether the reviewer would have asked a different question or
  used different methods. "I would have done it differently" is not a valid
  critique unless the chosen approach produces a specific, identifiable
  problem in the manuscript.

## Common Failure Modes

| Failure mode | Prevention |
|---|---|
| Vague praise without specifics ("interesting and well-written") | Name what is interesting and why; identify specific passages or arguments that work well |
| Destructive criticism without remediation ("the argument fails") | For every identified problem, suggest at least one concrete path to improvement |
| Ignoring positive contributions in the rush to critique | Lead with a genuine assessment of what the manuscript contributes before turning to concerns |
| Rebuttal that skips uncomfortable reviewer points | Require explicit response to every numbered or substantive point, even if the response is "we respectfully disagree because..." |
| Confusing "I would have done it differently" with actual flaws | Distinguish between the reviewer's preferences and genuine problems in the manuscript's internal logic |
| Review that demands a different paper than the one submitted | Evaluate the manuscript on its own terms: does it achieve what it sets out to do? Flag only where the chosen approach creates identifiable problems |

## Examples

**Example 1: Writing a constructive review of a cultural anthropology article**

Input: "I've been asked to review an article for American Ethnologist about
digital kinship practices among transnational Filipino families. It uses
14 months of multi-sited ethnography across Manila and Los Angeles. I'm a
media anthropologist with some expertise in digital studies but not in
Filipino communities. How should I structure my review?"

Output approach:
- Load review-writing-guide reference
- Set review type to blind review (AE default)
- Set journal tier to AAA flagship (American Ethnologist expects significant
  theoretical contribution + ethnographic depth)
- Set reviewer stance to constructive
- Calibrate reviewer expertise: strong on digital ethnography methods and
  media theory, weaker on Philippine studies and kinship literature
- Structure review with six-part anatomy: summary demonstrating careful
  reading, contribution assessment relative to digital kinship and media
  anthropology literatures, strengths (multi-sited design, digital methods),
  major issues (2-3, focusing on areas within reviewer expertise: digital
  methods rigor, theoretical framing of mediation/platform affordances),
  minor issues (citation suggestions from digital studies, clarity points),
  recommendation
- Flag where reviewer should note limits of own expertise ("I defer to
  reviewers with deeper knowledge of Philippine kinship scholarship on
  whether the engagement with that literature is adequate")

**Example 2: Responding to mixed R&R feedback with contradictory reviewers**

Input: "I got an R&R from Cultural Anthropology. Reviewer 1 says I need more
theoretical framing and wants me to engage with Povinelli and Berlant.
Reviewer 2 says the paper is already too theory-heavy and the ethnography
gets lost. The editor's letter says to address both reviewers' concerns.
How do I handle this?"

Output approach:
- Load revision-response-guide reference
- Set decision context to major revisions/R&R at AAA flagship
- Identify the contradiction: R1 wants more theory, R2 wants less
- Strategy: Read the editor's letter carefully for signals about which
  reviewer to weight more heavily (editors often signal priority implicitly)
- Reframe the contradiction as a structural problem: both reviewers may be
  responding to the same issue from different angles — the theory and
  ethnography may not be well integrated, so R1 experiences insufficient
  theorization while R2 experiences theory displacing ethnography
- Recommend revision approach: integrate theory into the ethnographic
  analysis rather than adding or removing theoretical sections; weave
  Povinelli/Berlant into the ethnographic discussion where they illuminate
  specific findings rather than adding a standalone theory section
- Structure the rebuttal to show how the single revision strategy addresses
  both reviewers' concerns simultaneously
- Point-by-point response for each reviewer with specific page citations

**Example 3: Developmental evaluation of a graduate student's first submission**

Input: "My advisee just submitted their first article to the Journal of the
Royal Anthropological Institute and got a reject with encouragement to
revise and resubmit elsewhere. The reviewers said the ethnography is strong
but the argument isn't clear and the literature review is too broad. Can you
help me think about how to advise them on the reviews and plan a revision
for resubmission to another journal?"

Output approach:
- Load both reference files
- Set reviewer stance to developmental (mentoring context)
- Set journal tier to interdisciplinary (JRAI publishes across subfields)
- Frame the task as two-part: (a) interpreting the reviews for a first-time
  author and (b) planning a revision for resubmission
- Help interpret reviewer language: "the argument isn't clear" likely means
  the through-line from research question to evidence to contribution is not
  explicit enough; "literature review is too broad" likely means the lit
  review surveys a field rather than building toward a specific analytical
  gap
- Revision plan: identify the core argument (what single claim does the
  ethnography support?), restructure the literature review to build toward
  that claim specifically, ensure each ethnographic section advances the
  argument rather than describing context
- Journal targeting for resubmission: based on the paper's strengths
  (strong ethnography) and the revision plan, suggest journals where the
  revised version would be competitive
- Timeline: realistic plan for revision given graduate student workload,
  typically 2-4 months for substantive revision
登录后下载/点赞/收藏 ❤ 4 | ★ 0
academic-review

目录结构

SKILL.md

举报内容

提示
