statistical-analysis

Guided statistical analysis with test selection and reporting. Use when you need help choosing appropriate tests for your data, assumption checking, power analysis, and APA-formatted results. Best for academic research reporting, test selection guidance. For implementing specific models programmatically use statsmodels.

1.13x

Quality

75%

Does it follow best practices?

Impact

91%

1.13x

Average score across 6 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/statistical-analysis/SKILL.md

Evaluation results

90%

15%

Comparing Response Times Across Two Product Variants

Assumption checking and adaptive test selection

Criteria

Without context

With context

assumption_checks module

100%

normality check present

100%

homogeneity check present

100%

pingouin for t-test

100%

Welch vs standard selection

100%

Cohen's d reported

100%

Effect size CI reported

100%

APA t-test format

100%

Descriptive stats in report

100%

Assumption results in report

100%

93%

31%

Evaluating Training Program Effectiveness Across Three Delivery Formats

ANOVA with post-hoc tests and power analysis

Criteria

Without context

With context

pingouin for ANOVA

100%

partial eta-squared reported

25%

100%

Tukey HSD post-hoc

80%

100%

multiple comparison correction noted

100%

APA ANOVA format

70%

100%

assumption checks present

100%

sensitivity or a priori power

86%

100%

no post-hoc power

41%

descriptive stats in report

100%

post-hoc results in report

100%

14%

Predicting Student Exam Performance from Study Behaviors

Multiple regression with VIF diagnostics and APA reporting

Criteria

Without context

With context

statsmodels for regression

100%

sm.add_constant used

100%

VIF computed

100%

VIF thresholds applied

100%

regression assumptions checked

100%

R² and adjusted R² reported

100%

APA regression format

66%

100%

CIs for predictors

100%

multicollinearity statement in report

100%

standardized coefficients

100%

Comparing Recovery Outcomes: Bayesian Analysis for a Clinical Pilot Study

Bayesian analysis with pymc/arviz and convergence diagnostics

Criteria

Without context

With context

pymc used for model

100%

arviz used for diagnostics

100%

Priors explicitly specified

100%

Difference as Deterministic

100%

pm.sample with tune

100%

R-hat convergence check

100%

ESS convergence check

80%

100%

Credible interval reported

80%

100%

Posterior probability statement

100%

Posterior plot saved

100%

86%

12%

Understanding Wellness Predictors: Correlation Analysis for a Health Research Study

Correlation analysis with test selection and multiple comparisons correction

Criteria

Without context

With context

Normality checked before correlation

100%

Spearman for non-normal variables

100%

pingouin for correlations

CIs for correlations reported

80%

Multiple comparisons correction applied

100%

Correction method named

100%

APA correlation format

70%

100%

Effect size r reported

62%

75%

Correlation heatmap saved

100%

Practical significance discussed

100%

79%

-9%

Analyzing Training Program Outcomes: Categorical and Non-Parametric Analysis

Categorical data analysis with chi-square, Fisher's exact, and Cramér's V effect size

Criteria

Without context

With context

Expected counts checked

100%

Fisher's exact when expected counts < 5

100%

Cramér's V for larger tables

100%

Phi coefficient for 2x2

100%

scipy association function used

APA chi-square format

100%

Exact p reported for Fisher's

100%

Effect size benchmarks applied

50%

62%

All three analyses reported

100%

Observed/expected frequencies described

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: 086de41

Evaluated: about 1 month ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Comparing Response Times Across Two Product Variants Evaluating Training Program Effectiveness Across Three Delivery Formats Predicting Student Exam Performance from Study Behaviors Comparing Recovery Outcomes: Bayesian Analysis for a Clinical Pilot Study Understanding Wellness Predictors: Correlation Analysis for a Health Research Study Analyzing Training Program Outcomes: Categorical and Non-Parametric Analysis

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.