心理学数据分析(统计检验、结果解读)

v1.0.1

心理学数据分析决策与执行助手。当用户需要分析心理学研究数据、决定使用什么统计检验、处理数据、运行分析、解读结果时使用。触发场景包括:"帮我分析数据"、"做什么检验"、"SPSS怎么做"、"Python分析"、"结果怎么读"、"验证我的假设"、"数据分析"、"处理数据"等。

0· 30·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for wangjinhongmy-pixel/psychology-data-analysis.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "心理学数据分析(统计检验、结果解读)" (wangjinhongmy-pixel/psychology-data-analysis) from ClawHub.
Skill page: https://clawhub.ai/wangjinhongmy-pixel/psychology-data-analysis
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install psychology-data-analysis

ClawHub CLI

Package manager switcher

npx clawhub@latest install psychology-data-analysis
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description ask for recommending and running statistical tests. The skill includes a Python analysis script and reference docs that match that purpose. There are no unrelated required binaries, environment variables, or credentials.
Instruction Scope
SKILL.md limits actions to collecting study design info, recommending tests, guiding SPSS, and running the included Python script on user-provided data. That is in-scope. Minor mismatch: SKILL.md lists references/apa_format.md as an included reference but the file manifest does not contain this file — a documentation/inventory inconsistency. Also the Python script asks for filenames/variable names interactively in some flows (input()) and prints results; confirm expected UX for automated runs.
Install Mechanism
No install spec; skill is instruction-only plus a script. No downloads or external install URLs are present. The Python script relies on common scientific Python packages (pandas/scipy/statsmodels/etc.), which is appropriate for the stated functionality.
Credentials
The skill requests no environment variables, credentials, or config paths. The resources it accesses are local data files supplied by the user (CSV/Excel/SAV). There are no network endpoints, HTTP calls, or tokens present in the code or docs.
Persistence & Privilege
always is false and the skill does not request system-wide configuration changes or persistent privileges. It does not attempt to modify other skills or system settings.
Assessment
This skill appears to do what it claims: recommend tests, guide SPSS, and run local analyses with the included Python script. Before installing/using: 1) Inspect the full python_analysis.py yourself (or run it in a sandbox) and test it on non-sensitive dummy data to confirm behavior and outputs. 2) Confirm and/or add any missing reference files (SKILL.md mentions references/apa_format.md but it is not in the manifest). 3) Ensure required Python packages are installed in a controlled environment (virtualenv/conda) to avoid unexpectedly installing packages globally. 4) Because the script reads files you provide, avoid running it on data that contains secrets (identifiable personal data) unless you trust the execution environment. 5) If you plan to let an agent invoke the skill autonomously, be aware it can run the script on any data path you provide — that is normal but double-check which data the agent will pass.

Like a lobster shell, security has layers — review code before you run it.

latestvk97cb0hhdcp36jmgmcdyar70x185pym8
30downloads
0stars
2versions
Updated 8h ago
v1.0.1
MIT-0

Psychology Data Analysis Assistant

心理学数据分析决策与执行助手。输入研究设计 → 推荐检验 → 执行分析 → 解读结果。

工作流程

Step 1: 收集研究信息

向用户询问以下关键信息:

  1. 研究设计类型

    • 被试间设计(不同组之间的差异比较)
    • 被试内设计(同一组被试在不同条件下的测量)
    • 混合设计(同时包含被试间和被试内因素)
    • 问卷/相关研究(变量间关系)
  2. 变量信息

    • 自变量(X):名称、类型(连续/分类)、水平数
    • 因变量(Y):名称、类型(连续/分类)
    • 协变量(如果有)
  3. 样本信息

    • 总样本量 N
    • 每组样本量(如果是被试间设计)
    • 是否满足正态性假设(可做 Shapiro-Wilk 检验)
  4. 研究假设

    • 差异假设(组间差异)
    • 相关假设(变量间相关)
    • 预测假设(回归/预测)
    • 中介/调节假设
  5. 软件偏好

    • SPSS(倾向菜单操作还是语法)
    • Python(倾向 Jupyter Notebook 环境还是脚本)
  6. 数据格式

    • Excel (.xlsx)
    • CSV
    • SPSS (.sav)

Step 2: 推荐检验方案

根据收集的信息,推荐合适的统计检验:

变量类型组数/设计推荐检验
连续因变量,两组被试间独立样本 t 检验
连续因变量,两组被试内配对样本 t 检验
连续因变量,多组被试间单因素 ANOVA
连续因变量,多组被试内重复测量 ANOVA
连续因变量,≥2自变量被试间/混合多因素 ANOVA/混合 ANOVA
分类因变量任意卡方检验
连续变量相关研究Pearson/Spearman 相关
连续变量预测研究多元线性回归
连续因变量,有中介路径分析中介效应检验(Bootstrap)
连续因变量,有调节调节分析调节效应检验(层次回归)

检验前提检查:

  • 正态性:Shapiro-Wilk 检验(N>50 时看偏度/峰度)
  • 方差齐性:Levene 检验
  • 球形性:Mauchly 检验(重复测量 ANOVA)
  • 样本量评估

Step 3: 执行分析

根据用户选择的软件,执行相应分析。

Python 执行

使用 scripts/python_analysis.py,支持:

  • 数据加载与清洗
  • 描述性统计
  • 假设检验(t 检验、ANOVA、相关、回归等)
  • 中介/调节效应
  • 结果可视化

执行方式:

python scripts/python_analysis.py --data <数据路径> --analysis <分析类型> --options <选项>

SPSS 指导

提供详细操作步骤:

  1. 菜单路径
  2. 对话框设置
  3. 语法命令(可选)
  4. 输出结果解读

Step 4: 结果解读

输出内容包括:

  • 统计量:检验统计值(t/F/r 等)、自由度、p 值
  • 效应量:Cohen's d、η²、r、β 等
  • 置信区间:95% CI
  • APA 格式报告:可直接粘贴到论文的结果部分

Step 5: 假设验证

根据分析结果,判定:

  • 假设是否得到支持
  • 结果的实际意义解读
  • 局限性说明
  • 后续分析建议

脚本

  • scripts/python_analysis.py — 主要 Python 分析脚本

参考资料

  • references/stat_decision_tree.md — 统计检验决策树
  • references/python_guide.md — Python (scipy/scikit-learn/statsmodels) 使用指南
  • references/spss_guide.md — SPSS 操作指南(常用分析)
  • references/effect_size.md — 效应量参考表
  • references/apa_format.md — APA 结果报告格式

注意事项

  • 始终先检查数据是否满足检验的前提假设
  • 大样本(N>50)时,t/F 检验对正态性不敏感,但中位数比较仍需非参数检验
  • 报告效应量与置信区间,而非仅报告 p 值
  • Python 输出结果后,主动提供 APA 格式的报告文本

Comments

Loading comments...