Skills statistical-analysis

📊

statistical-analysis

Name: statistical-analysis
Author: K-Dense-AI

Safe ⚡ Contains scripts⚙️ External commands🌐 Network access

Analyze data with statistical tests and APA reporting

Also available from: davila7

Stop guessing which statistical test to use. Get guided test selection, assumption checking, and APA-formatted results for your research data. Perfect for students and researchers who need rigorous statistical analysis.

Supports: Claude Codex Code(CC)

🥈 77 Silver

Download the skill ZIP

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

Toggle on and start using

Test it

Using "statistical-analysis". I want to compare anxiety levels between therapy group and control group. Which test should I use?

Expected outcome:

Recommended test: Independent samples t-test
Assumptions to check:
Independence: Participants in separate groups
Normality: Use Shapiro-Wilk test (n < 50) or visual inspection (n > 50)
Homogeneity of variance: Use Levene's test
Scale type: Continuous dependent variable required
If assumptions violated: Use Welch's t-test (if only homogeneity violated) or Mann-Whitney U test (if normality violated)
Effect size: Calculate Cohen's d with 95% CI

Using "statistical-analysis". How do I report an ANOVA result in APA format?

Expected outcome:

F(df_between, df_within) = F-value, p = p-value, partial eta-squared = value
Example: F(2, 57) = 4.56, p = .014, ηp² = .14
Report means and standard deviations for each group in a table
Include a figure if helpful for visualization
Interpret: There was a significant effect of the independent variable

Security Audit

Safe

v4 • 1/17/2026

All 190 static findings are FALSE POSITIVES. The 'external_commands' detections are Python code examples in markdown documentation blocks (not shell execution). The 'weak cryptographic algorithm' detections are triggered by common words like 'comprehensive' in documentation text at line 3 of markdown files. The 'system reconnaissance' detections flag statistical terms like ACF/PACF (Autocorrelation Function). This is a legitimate statistical analysis documentation skill with no security risks.

Files scanned

3,639

Lines analyzed

findings

Total audits

Risk Factors

⚡ Contains scripts (1)

scripts/assumption_checks.py:1-540

⚙️ External commands

No specific locations recorded

🌐 Network access

No specific locations recorded

Audited by: claude View Audit History →

Quality Score

Architecture

100

Maintainability

Content

Community

100

Security

Spec Compliance

What You Can Build

Select statistical tests for thesis data

Get help choosing the right tests, checking assumptions, and writing APA-style results sections for your dissertation or thesis.

Validate analysis plan before conducting study

Plan your statistical analysis strategy, determine sample sizes with power analysis, and ensure appropriate test selection for grant proposals.

Learn proper statistical testing workflow

Follow guided workflows to understand when to use t-tests vs ANOVA vs non-parametric alternatives with assumption checking.

Try These Prompts

Choose the right test

I want to compare test scores between students who received tutoring vs those who didn't. I have 30 students in each group. Which statistical test should I use and what assumptions do I need to check?

Check assumptions

I'm planning to run a one-way ANOVA with 4 groups. How do I check normality and homogeneity of variance assumptions? Show me the code and how to interpret results.

Report results APA style

I ran an independent t-test and got t(58) = 2.34, p = .022, with means of 78.5 (SD = 8.2) and 72.1 (SD = 9.5). Write this in proper APA format with effect size.

Power analysis

I need to detect a medium effect size (Cohen's d = 0.5) with 80% power for an independent t-test at alpha = .05. How many participants do I need per group?

Best Practices

Always check assumptions before interpreting test results - never skip this step
Report effect sizes with confidence intervals, not just p-values
Use visual inspection for normality with large samples (n > 100) as formal tests become overly sensitive

Avoid

Choosing tests based on desired outcomes rather than data characteristics
Reporting only p-values without effect sizes or confidence intervals
Assuming normality without checking, especially with small samples

Frequently Asked Questions

What is the difference between parametric and non-parametric tests?

Parametric tests assume specific distributions (usually normal) and equal variances. Non-parametric tests make fewer assumptions but may have less power. Use parametric when assumptions are met.

How do I know if my data is normally distributed?

Use both visual (Q-Q plots, histograms) and formal tests (Shapiro-Wilk). With n > 30, mild violations are usually acceptable due to the Central Limit Theorem.

What if my data violates multiple assumptions?

Consider data transformations (log, square root), use robust alternatives (Welch's test), or switch to non-parametric tests. Document all violations and remedies.

How do I report statistical results in APA format?

Include test statistic, degrees of freedom, p-value, and effect size: t(58) = 2.34, p = .022, d = 0.61. Always report confidence intervals for effect sizes.

When should I use Bayesian statistics?

Bayesian methods are useful when you have prior information, want to quantify evidence strength (Bayes Factors), or need to update beliefs as data accumulates.

What is the minimum sample size I need?

Conduct a power analysis before data collection. For medium effects and 80% power, you typically need 25-35 per group for t-tests. Larger samples needed for smaller effects.