Statistical Analysis

Statistical Analysis Skill

Biomechanical data statistical analysis with Excel output

Overview

This skill provides a complete workflow for biomechanical research:

• Paired t-test: Condition and task comparisons with Bonferroni correction
• Repeated-measures ANOVA: N×M within-subject design analysis
• Excel output: Formatted reports with standard color scheme and templates
• Excel formula automation: Auto-updating analysis with hybrid Python + formula approach

When to Use

• Analyzing EMG, force, or kinematic data
• Comparing conditions (e.g., new vs old cart)
• Comparing tasks (e.g., lift, pull, push)
• Running repeated-measures experiments
• Generating publication-ready statistical reports in Excel
• Creating auto-updating Excel analysis sheets

Files

Key Principle

⚠️ These are FUNCTION LIBRARIES, not standalone scripts.

The AI must:

1. Inspect the data to determine column structure
2. Identify dependent variable and condition factors
3. Call functions with explicit parameters
4. Always output results to Excel

Usage Example

T-test Analysis

from ttest_statistical_analysis import (
    load_and_preprocess_data,
    aggregate_trials,
    paired_ttest_condition,
    export_to_excel
)

# AI determines these by inspecting data
dependent_var = "rvc_norm_rms"
condition_col = "cart_categories"
condition_values = ["new", "old"]

# Run analysis
df = load_and_preprocess_data(data_path, dependent_var, condition_col)
df_agg = aggregate_trials(df, dependent_var, condition_col)
results = paired_ttest_condition(df_agg, condition_col, condition_values)

ANOVA Analysis

from anova_statistical_analysis import (
    compute_cell_means,
    find_valid_subjects,
    run_rm_anova,
    export_to_excel
)

# Run analysis
cell_means = compute_cell_means(df, dependent_var, condition_col)
valid_subjects, _ = find_valid_subjects(cell_means, condition_col, dependent_var)
anova_results = run_rm_anova(cell_means_valid, dependent_var, condition_col)

Excel Output Format

All analyses produce Excel files with standard sheets:

1. methods: Analysis methodology (dynamically generated)
2. descriptives: Mean, SD, SEM, N per condition
3. statistical_tests: Test results with significance highlighting
4. cell_means: Subject-level data (optional)

Standard Color Scheme

See excel-format.md for detailed formatting specifications. See excel-formula-automation.md for formula automation workflow.