科研 Skill

政策议题框架分析

beebe-policy-framing

Runs anthropological framing and metaphor analysis on contested-policy documents, producing actor-attributed CSVs for NVivo import. Use for policy-framing or metaphor-analysis coding.

发布者 Codar安全审核通过v0.1.0更新于 2026-11-18

SKILL.md

Beebe Policy-Framing Analysis

This skill operationalizes the research methodology of Dr. James Beebe (Rapid Assessment Process, 2001; Rapid Qualitative Inquiry, 2014) as designed in his study of contested policy debates: an anthropology of public policy (Wedel, Shore, Feldman & Lathrop 2005) combining policy-metaphor analysis (Schlesinger & Lau 2000), conceptual metaphor theory (Lakoff & Johnson 1980), and a social constructionist stance. Its premise: in policy struggles, language is the weapon — the analysis tracks who frames what, with which metaphors, to invite which policy conclusions.

Requirements

A study codebook is required before coding. It defines the actors, activities, venues, frame seeds, and timeline anchors (see codebook-template.md in this skill's folder).

If the user has no codebook, do not draft one from intuition — surface it from the data: first run the qualigpt skill inductively on the corpus (or a pilot sample), then work with the researcher to convert the resulting themes into frame families, the attributed speakers into actor codes, and the contested programs into activity codes, using the mapping worksheet in the template. Get the researcher's explicit approval of the resulting codebook before framing analysis begins. Never code against an unapproved codebook.

This inductive→deductive sequence is part of the methodology, and it continues across runs: [PROPOSED] codes emitted by each framing pass are the inductive channel feeding codebook revisions.

Core principles

  1. The field is multi-sited (Wedel et al.): treat internal documents, press, hearings, investigation reports, and interviews as distinct venues; code the venue of every item. Actors have varying institutional leverage across sites.
  2. Metaphors are conceptual, not decorative (Lakoff): beneath each surface phrase, code the underlying mapping in CAPS (e.g., DEBATE IS WAR, PROJECT IS JOURNEY, INSTITUTION IS SICK BODY).
  3. Metaphors are analyzed in stages (Schlesinger & Lau): stage 1 — what does the metaphor assert about its target? stage 2 — what policy conclusion does it invite the audience to accept?
  4. Every statement is attributed: named speaker, role, date, venue. The unit of analysis is the framing act, not the aggregate theme.
  5. Positionality is data: if the researcher (or their close associates) authored corpus documents, those passages are flagged, collected for reflexive review, and — per the researcher's decision — either analyzed like any actor's or bracketed. Watch for the researcher's own analytic vocabulary; reflexivity extends to it.
  6. Social construction means both sides: code opposing frames of the same events symmetrically. The analysis succeeds only if each side would recognize itself as fairly rendered.
  7. Frame contestation is a first-class finding: note when actors fight about a metaphor (quoting, negating, or qualifying an opponent's term — "so-called X," scare quotes) rather than with one.

Outputs

Produce five linked CSVs plus a report, in an output folder next to the corpus:

  1. statements.csv — statement_id | speaker (actor code) | role | date | venue code | verbatim statement | activity code | stance | positionality flag | source_doc
  2. metaphors.csv — metaphor_verbatim | conceptual_metaphor | speaker | date | venue | activity | stage1_meaning | stage2_policy_inference | stance | positionality
  3. frames.csv — frame code | frame | supporting actors | opposing actors | representative quotes | first/last observed | change over time
  4. positionality.csv — every researcher-authored/attributed passage, verbatim, with a reflexive note
  5. timeline.csv — date | event/document | actors | frames active | anchor aligned to the codebook's anchor events
  6. run report (.md) — principal findings, frame contests, corrections to prior runs, and a gap report: what the corpus cannot show and which acquisitions would fill it

Rules

  • Quotes verbatim, including errors (mark [sic] only in the report, never inside the quote).
  • New frames, actors, or activities not in the codebook are emitted flagged [PROPOSED] — never silently extend the codebook; the researcher (or study steward) approves.
  • Stance is coded relative to the contested policy; statements about adjacent conflicts are AMBIG unless the speaker links them to the policy fight (and whether they do so is itself a datum).
  • On repeat runs over a growing corpus, reconcile: dedupe documents, correct earlier coding, and note corrections explicitly in the run report.
  • All outputs are first-pass coding for the researcher to verify against sources; researcher judgment carries the analysis.

Citations

If this skill contributes to published research, cite: Beebe (2001, 2014); Wedel, Shore, Feldman & Lathrop (2005); Schlesinger & Lau (2000); Lakoff & Johnson (1980). See the repository README for full references.

示例提示词

帮我对这批政策听证记录做框架与隐喻分析,找出行动者的论述立场。

所属插件

社会学、政治学与城市研究面向社会科学研究:问卷与访谈设计、质性编码、文本即数据、社会网络、ABM、比较政治与选举数据、RCT 与 DID、城市与区域分析。

安装该插件会一并获得同领域的 58 个 Skill / MCP。

同插件的其他科研 Skill

常见问题

政策议题框架分析能做什么?

Runs anthropological framing and metaphor analysis on contested-policy documents, producing actor-attributed CSVs for NVivo import. Use for policy-framing or metaphor-analysis coding。

如何安装政策议题框架分析?

在科搭 Codar 左侧点击「插件」,在「组件」页签搜索「政策议题框架分析」并安装;在 Claude Code 或 Codex 中可以发送本页的安装提示词,或下载 Skill 包后解压到 ~/.claude/skills 或 ~/.codex/skills。

政策议题框架分析需要哪些权限?

该科研 Skill没有声明额外权限;执行具体操作时仍由 Agent 按自身规则请求确认。

政策议题框架分析经过官方审核吗?

是。SpeedSkills 收录的每个版本都会记录内容摘要与权限声明,当前版本已通过安全审核。