Back to Details

rw-research-design

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

RW Research Design

RW Research Design

把研究问题转成构念、设计、样本、测量、分析、证伪和执行计划。
Transform research questions into constructs, design, samples, measurement, analysis, falsification, and execution plans.

启动

Startup

  1. 读取 
    references/standalone.md
    、
    references/method.md
    和 
    references/standards.md
    。
  2. 读取用户本轮提供的材料；没有材料时，只完成当前证据允许的部分。
  3. 需要规则判断时检索 
    references/atoms.jsonl
    ；遇到相似任务时读取 
    references/cases.md
    。
  4. 需要选择方法或工具时读取 
    references/domain-guide.md
    ，并使用 
    assets/worksheet.md
    组织交付。
  5. 读取 
    references/acceptance.md
    。
    references/behavior-tests.json
    只用于测试，不作为用户任务事实。
  6. 当前文献、API、报告规范和期刊要求可能变化时，打开 
    references/source-map.md
    中的官方链接核验并记录日期。
  1. Read 
    references/standalone.md
    , 
    references/method.md
    , and 
    references/standards.md
    .
  2. Read the materials provided by the user in this round; if no materials are provided, only complete the parts allowed by current evidence.
  3. Retrieve 
    references/atoms.jsonl
    when rule judgment is needed; read 
    references/cases.md
    when encountering similar tasks.
  4. Read 
    references/domain-guide.md
    when selecting methods or tools, and use 
    assets/worksheet.md
    to organize deliverables.
  5. Read 
    references/acceptance.md
    . 
    references/behavior-tests.json
    is only used for testing and not as factual information for user tasks.
  6. When current literature, APIs, report specifications, and journal requirements may change, open the official links in 
    references/source-map.md
    for verification and record the date.

工作阶段

Work Stages

  1. 写出主要问题、目标估计量或需要解释的现象。
  2. 定义构念、变量、测量时点、数据来源和目标人群。
  3. 比较实验、观察、诊断、预测、定性、混合方法和经济设计的适配度。
  4. 识别对照、混杂、选择、测量、缺失、时间和实施偏倚。
  5. 预设主要分析、敏感性分析、样本依据和失败判据。
  6. 输出首选设计、替代设计、资源限制、伦理事项和报告规范。
  1. Write down the main questions, target estimators, or phenomena that need explanation.
  2. Define constructs, variables, measurement time points, data sources, and target populations.
  3. Compare the suitability of experimental, observational, diagnostic, predictive, qualitative, mixed-method, and economic designs.
  4. Identify control, confounding, selection, measurement, missing, time, and implementation biases.
  5. Preset primary analysis, sensitivity analysis, sample justification, and failure criteria.
  6. Output the preferred design, alternative designs, resource constraints, ethical considerations, and report specifications.

运行规则

Operating Rules

  • 研究设计从问题和目标估计量出发，不从软件功能出发。
  • 报告规范用于检查透明度，不能替代设计合理性。
  • 随机化能处理已知和未知混杂，但不能自动解决流失、测量和执行偏差。
  • 观察研究要明确时间顺序、混杂结构和目标人群。
  • 诊断研究要明确 index test、reference standard、阈值和使用情境。
  • 预测研究要区分开发、内部验证、外部验证、更新和影响评价。
  • 定性研究要让取样、情境、研究者位置和分析方法与问题一致。
  • 混合方法设计要说明两类数据在何时、为何以及如何整合。
  • 样本量依据要对应主要问题、设计和分析，不只给一个人数。
  • 主要和次要分析、亚组和探索分析要预先区分。
  • 可行性限制应触发替代设计，不能静默降低结论强度。
  • 设计必须写明什么结果会削弱主要解释。
  • Research design starts from the problem and target estimator, not from software functions.
  • Report specifications are used to check transparency and cannot replace design rationality.
  • Randomization can handle known and unknown confounding, but cannot automatically solve attrition, measurement, and execution biases.
  • Observational studies must clarify temporal order, confounding structure, and target populations.
  • Diagnostic studies must clarify the index test, reference standard, threshold, and usage scenario.
  • Predictive studies must distinguish between development, internal validation, external validation, update, and impact evaluation.
  • Qualitative studies must align sampling, context, researcher position, and analysis methods with the problem.
  • Mixed-method designs must explain when, why, and how the two types of data are integrated.
  • Sample size justification must correspond to the main question, design, and analysis, not just provide a number.
  • Primary and secondary analyses, subgroup and exploratory analyses must be distinguished in advance.
  • Feasibility constraints should trigger alternative designs, not silently reduce the strength of conclusions.
  • The design must specify what results would weaken the main interpretation.

输出

Output

  • 研究设计和目标估计量。
  • 变量、样本、偏倚和分析计划。
  • 替代设计、失败判据和执行前缺口。
  • Research design and target estimator.
  • Variables, samples, biases, and analysis plan.
  • Alternative designs, failure criteria, and pre-execution gaps.

停止条件

Stop Conditions

  • 现有数据不能回答问题时，明确标记不可回答。
  • 缺少伦理、许可或数据治理信息时，不把方案写成可立即执行。
  • 不把观察性关联包装成随机试验证据。
  • 不提供替代统计师、伦理委员会或领域专家审批的结论。
  • When existing data cannot answer the question, clearly mark it as unanswerable.
  • When ethical, licensing, or data governance information is missing, do not write the plan as immediately executable.
  • Do not package observational associations as evidence from randomized trials.
  • Do not provide conclusions that replace the approval of statisticians, ethics committees, or domain experts.

独立运行

Independent Operation

  • 默认不读取私人工作区、个人语料目录或预设 research-lab。
  • 用户提供的文件、文本、链接和数据是当前任务输入，不是安装依赖。
  • 网络不可用时，使用 Skill 内的稳定方法继续；需要当前事实的部分标记为待核验。
  • 本包内其他科研 Skill 存在时可以接续；单独安装时直接返回下一步说明，不停止当前任务。
  • By default, do not read private workspaces, personal corpus directories, or preset research-lab.
  • Files, text, links, and data provided by the user are inputs for the current task, not installation dependencies.
  • When the network is unavailable, continue using stable methods within the Skill; mark parts requiring current facts as pending verification.
  • Can connect with other research Skills in this package if they exist; when installed alone, directly return instructions for the next step without stopping the current task.

来源纪律

Source Discipline

  • 把用户材料、公开来源、当前推断和未知事项分开。
  • 报告规范只检查报告透明度，不自动证明设计质量。
  • 公开来源摘要保存在 Skill 内；需要版本、费用、政策、API 或期刊现状时回到官方页面。
  • 不生成不存在的论文、数据、DOI、工具运行结果或期刊要求。
  • Separate user materials, public sources, current inferences, and unknown matters.
  • Report specifications only check report transparency and do not automatically prove design quality.
  • Public source abstracts are stored within the Skill; return to official pages when version, cost, policy, API, or journal status is needed.
  • Do not generate non-existent papers, data, DOIs, tool operation results, or journal requirements.