Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› arsoc-literature-synthesis
arsoc-literature-synthesis
GitHub用于系统性地搜集、阅读和综合社会学文献，确保覆盖各理论传统与方法，避免引用遗漏。通过种子集、滚雪球及多数据库检索构建证据库，并利用矩阵进行批判性综合，为年度综述提供全面且平衡的文献基础。
Trigger Scenarios
需要系统梳理子领域文献时
担心遗漏重要贡献或特定流派时
拥有大量研究但无法有效比较或调和发现时
需应对审稿人关于文献缺失的质疑时
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill arsoc-literature-synthesis -g -y
SKILL.md
Frontmatter
{
"name": "arsoc-literature-synthesis",
"description": "Use when systematically gathering, reading, and synthesizing a large body of sociological research for an Annual Review of Sociology (ARSoc) review — coverage discipline and avoiding citation gaps across theoretical traditions. Builds the evidence corpus and synthesis notes; it does not impose the analytical spine (arsoc-organizing-framework) or write prose (arsoc-writing-style)."
}
Literature Synthesis (arsoc-literature-synthesis)
When to trigger
- The topic was commissioned and it is time to read the subfield thoroughly
- Reading feels unsystematic; you fear missing important contributions or a whole tradition
- You have many studies but no way to weigh, compare, or reconcile their findings
- A Committee/referee at ARSoc is likely to ask "why did you omit X / the Y literature / the qualitative work?"
Coverage discipline: read the subfield, not a convenience sample
An ARSoc review's credibility rests on the reader's belief that you read everything that matters across methods and schools. Sociology spans quantitative, qualitative, computational, and theoretical work, and reviewers notice when one mode is slighted. Build coverage systematically rather than from memory:
- Seed set. Start from the field's canonical pieces and the references behind the commissioned topic.
- Forward + backward snowball. Backward: every study's reference list. Forward: who cites the seeds (Google Scholar / Web of Science / Scopus). Iterate until new searches stop turning up unseen important work (saturation).
- Database + keyword sweep. Search Sociological Abstracts, Web of Science, Scopus, JSTOR, and Google Scholar by topic keywords and by the names of the rival theoretical traditions, to catch work indexed under adjacent labels.
- Mind-the-modes coverage. Deliberately sweep for qualitative and ethnographic work, computational / big-data studies, demographic analyses, and the theoretical statements, not only the quantitative articles you know best — siloing by method is a recurring ARSoc referee complaint.
- Recent + adjacent. Include recent papers and forthcoming work (the frontier) and the bordering literatures the subfield draws on, or referees will flag staleness and siloing.
- Saturation log. Record when each search stops yielding new must-cite work — this is your documented evidence of comprehensiveness for
arsoc-comprehensiveness-and-balanceandarsoc-transparency-and-reproducibility.
From reading to synthesis (not summary)
Summarizing is restating each study; synthesizing is making the studies talk to each other. Maintain an evidence matrix as you read:
| Column | What to capture |
|---|---|
| Study | author–year, the result you will cite it for |
| Question/claim | exactly what it argues or measures (so non-comparable studies are not pooled) |
| Method/data | design, sample, mode (quant / qual / computational / theoretical) — so you can weight credibility |
| Finding | direction + magnitude, or the central claim (for who-found-what tables) |
| Tradition | which theoretical school / school it speaks for |
| Credibility | identification strength, sample, analytic transparency — your appraisal, since ARSoc runs no new analysis |
| Tension | which other studies it agrees/conflicts with, and why |
This matrix is the raw material for the organizing framework, the summary tables, and the even-handed treatment of debates. Crucially, you appraise primary studies (you are the subfield's referee-of-record for the wider discipline); you do not re-analyze them.
Checklist
- Seed set drawn from canonical pieces + the commissioned topic's references
- Forward and backward snowball iterated to saturation
- Sociological Abstracts / Web of Science / Scopus / JSTOR / Scholar swept by keyword and by tradition
- Qualitative, computational, demographic, and theoretical modes all deliberately covered (no method silo)
- Recent/forthcoming work and adjacent literatures included (no staleness, no silo)
- Saturation log records where searches stopped yielding new must-cites
- Evidence matrix captures claim + method/mode + finding + tradition + credibility appraisal + tensions
- Non-comparable studies flagged as such (not pooled into a false consensus)
- No major author/school/method whose omission a referee could name
Anti-patterns
- Citing from memory or from a personal reading list (predictable coverage gaps)
- Summarizing study-by-study instead of synthesizing across studies
- Covering only the quantitative literature and ignoring the qualitative/theoretical work (method silo)
- Pooling studies that measure different objects into one "the literature finds…" claim
- Ignoring forthcoming work (looks stale) or ignoring adjacent subfields (looks siloed)
- Re-analyzing or "correcting" primary studies — an ARSoc review appraises, it does not re-estimate
Output format
【Seed set】<canonical + commissioned-topic references>
【Snowball status】backward/forward iterated to saturation? Y/N
【Databases swept】Sociological Abstracts / WoS / Scopus / JSTOR / Scholar — by keyword + tradition? Y/N
【Mode coverage】quant + qual + computational + demographic + theoretical all swept? Y/N
【Frontier + adjacency】recent/forthcoming work and bordering literatures included? Y/N
【Saturation evidence】<where searches stopped yielding new must-cites>
【Evidence matrix】rows ready with claim / method / finding / tradition / appraisal / tension? Y/N
【Coverage risks】<any author/school/method an omission referee could name>
【Next step】→ arsoc-organizing-framework (impose the analytical spine on the matrix)
Version History
- 1839142 Current 2026-07-05 12:25
