← Back to Skills Marketplace
zhan599

Psychology Statistical Analysis

by zhan599 · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
35
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install psychology-stats-analysis
Description
Provides comprehensive statistical analysis for psychological research, including t-tests, ANOVA, effect sizes, power analysis, and assumption checks.
README (SKILL.md)

心理学统计分析助手

一个专为心理学研究设计的全面统计分析工具,特别关注社会认知和实验心理学研究。

功能介绍

提供从描述统计到假设检验再到效应量计算的完整数据分析工具包。适用于实验心理学、社会认知研究、隐性学习研究等各类定量心理学工作。

主要功能

假设检验

  • t检验:独立样本t检验、配对t检验、单样本t检验
  • 方差分析(ANOVA):单因素方差分析、多因素设计
  • 相关与回归:皮尔逊相关、线性回归
  • 卡方检验:分类数据分析
  • 正态性与方差齐性检验:数据验证

描述统计

  • 均值、标准差、标准误
  • 中位数、四分位数、四分位距
  • 置信区间
  • 分布特征摘要

效应量计算

  • Cohen's d(t检验的效应量)
  • η²(方差分析的效应量)
  • 相关强度解释
  • 效应量大小分类

统计功效分析

  • 样本量估计
  • 统计功效计算
  • 研究设计优化与规划

使用方法

基础用法

from psychology_stats import PsychologyAnalyzer

analyzer = PsychologyAnalyzer()

# 描述统计
stats = analyzer.descriptive_stats(数据)

# 独立样本t检验
result = analyzer.independent_t_test(组1, 组2)

# 单因素方差分析
anova_result = analyzer.one_way_anova(组1, 组2, 组3)

凝视线索实验(Gaze-Cueing)

# 分析同余和不同余条件下的反应时差异
同余凝视_rt = [450, 460, 455, ...]  # 凝视线索同向条件
不同余凝视_rt = [480, 495, 490, ...]  # 凝视线索反向条件

result = analyzer.independent_t_test(
    np.array(同余凝视_rt), 
    np.array(不同余凝视_rt)
)
# 返回: t值、p值、Cohen's d、效应量解释

隐性学习与启动效应研究

# 比较不同条件下的启动效应
害怕面孔_正确率 = [0.65, 0.72, 0.68, ...]
中性面孔_正确率 = [0.58, 0.61, 0.59, ...]

result = analyzer.paired_t_test(害怕面孔_正确率, 中性面孔_正确率)
# 获得效应量和统计显著性

功效分析示例

# 规划所需样本量
所需样本量 = analyzer.power_analysis_ttest(
    effect_size=0.5,  # 预期的Cohen's d
    alpha=0.05,       # 第I类错误率
    power=0.8         # 期望的统计功效
)
# 返回每组推荐样本量和总样本量

应用场景

  • 凝视线索与注意研究:分析线索化效应和反应时模式
  • 社会认知研究:检验心理理论、视角采择、归因等假设
  • 隐性学习与启动:量化隐性记忆和启动效应
  • 实验心理学:分析阶乘设计和复杂交互作用
  • 快速数据检查:在进行推论统计前验证假设前提

系统要求

  • Python 3.8 或更高版本
  • numpy、scipy、pandas、matplotlib

输出格式

所有函数返回包含以下内容的字典:

  • 检验统计量(t、F、r、χ²)
  • p值和显著性阈值
  • 效应量(Cohen's d、η²、r²)
  • 人性化的结果解释
  • 样本量和自由度

输出示例

{
  "t_statistic": 2.45,
  "p_value": 0.024,
  "significant": true,
  "cohens_d": 0.95,
  "effect_size_interpretation": "large",
  "mean_difference": 28.5,
  "note": "发现显著差异,效应量为大"
}

使用建议

  • 进行参数检验前务必检查正态性和方差齐性假设
  • 同时报告原始统计量和效应量
  • 在研究设计阶段使用功效分析
  • 如可能,在结果中包含置信区间
  • 记录数据预处理和排斥标准

关于本工具

为社会认知、实验心理学和行为科学研究人员创建。旨在使统计分析更加易于理解和应用。

作者: @zhan599
所属机构: 华南师范大学 应用心理学系
研究方向: 社会认知、凝视线索、隐性学习

Usage Guidance
This appears low risk to install from a security standpoint. Users should still verify the statistical methods for their research needs and note that the documentation describes some capabilities that the included Python module does not currently implement.
Capability Assessment
Purpose & Capability
The stated purpose is psychology-focused statistical analysis, and the Python module implements descriptive statistics, t-tests, ANOVA, correlation, effect-size interpretation, power analysis, and assumption checks. The README advertises a few broader features, such as two-way ANOVA and visualization, that are not present in the included code, but this is a completeness issue rather than a security concern.
Instruction Scope
Instructions are ordinary usage examples for importing and calling the analyzer; no role changes, prompt overrides, hidden directives, or instructions to access unrelated user data were found.
Install Mechanism
The manifest declares common scientific Python dependencies and no install hooks, setup scripts, background services, or automatic commands.
Credentials
Runtime behavior is proportionate to local statistical computation; inspected code imports numpy, pandas, scipy, typing, and json, and does not use network access, shell execution, broad file reads, credential stores, or local indexing.
Persistence & Privilege
No persistence mechanism, privilege escalation, background worker, account mutation authority, or stored external state was found; the only state is an in-memory results dictionary.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install psychology-stats-analysis
  3. After installation, invoke the skill by name or use /psychology-stats-analysis
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- SKILL.md fully translated from English to Chinese for improved accessibility. - All feature descriptions, usage examples, and output formats now available in Chinese. - No code or functional changes; documentation update only. - Target audience expanded to Chinese-speaking users in psychology and behavioral sciences.
v1.0.0
Initial release – comprehensive toolkit for statistical analysis in psychological research. - Supports t-tests (independent, paired, one-sample), ANOVA (one-way, factorial), correlation/regression, chi-square, and assumption checks. - Calculates descriptive statistics, effect sizes (Cohen’s d, eta-squared), and offers effect size interpretation. - Includes power analysis for sample size calculations and study design. - Designed workflows for gaze-cueing and implicit memory/priming experiments. - Returns results with test statistics, p-values, effect sizes, and human-readable summaries.
Metadata
Slug psychology-stats-analysis
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Psychology Statistical Analysis?

Provides comprehensive statistical analysis for psychological research, including t-tests, ANOVA, effect sizes, power analysis, and assumption checks. It is an AI Agent Skill for Claude Code / OpenClaw, with 35 downloads so far.

How do I install Psychology Statistical Analysis?

Run "/install psychology-stats-analysis" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Psychology Statistical Analysis free?

Yes, Psychology Statistical Analysis is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Psychology Statistical Analysis support?

Psychology Statistical Analysis is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Psychology Statistical Analysis?

It is built and maintained by zhan599 (@zhan599); the current version is v1.0.1.

💬 Comments