OpenClaw R Stats#

When to Use#

User asks for any statistical analysis, hypothesis testing, group comparison, prediction, association, survival analysis, meta-analysis, causal inference, power/sample size, or mentions R statistical packages.

What This Skill Does NOT Do#

• Claim causality from observational data (use "associated with")
• Run large exploratory fishing without clear user intent
• Silently ignore assumption violations
• Report only p-values (always include effect sizes and CIs)

Pre-Flight (Mandatory)#

1. Confirm dataset exists and is readable
2. Run schema inspection: bash {baseDir}/scripts/run-rstats.sh schema --data <path>
3. Report: rows, columns, types, missing values
4. If missing > 5%, warn. If n < 30, warn small sample.

Environment Setup#

First time or errors: bash {baseDir}/scripts/run-rstats.sh doctor

Install by profile (only when needed):

ProfileScriptMethodsCoreinstall-core.Rt-test, regression, ANOVA, chi-sqSurvivalinstall-survival.RKM, Cox, competing risks, RMSTMissinginstall-missing.RMICE, MCAR testMixedinstall-mixed.RLMM, GLMM, GEE, ICCBayesinstall-bayes.Rbrms, Bayes factorsCausalinstall-causal.RPSM, IPTW, IV, DiD, RDD, TMLEMetainstall-meta.Rmeta-analysis, NMASEMinstall-sem.RSEM, CFA, lavaanDiagnosticinstall-diagnostic.RROC, kappa, alphaAdvancedinstall-advanced.RGAM, quantile, zero-inflatedPowerinstall-power.Rpower/sample size

Workflow#

1. Determine analysis type (see references/METHOD_TABLE.md)
2. Inspect dataset schema
3. Build JSON spec:

json

{
  "dataset_path": "<path>",
  "analysis_type": "<type>",
  "outcome": "<column>",
  "predictors": ["<col1>"],
  "hypothesis": "<plain language>",
  "alpha": 0.05,
  "seed": 42,
  "output_dir": "<path>"
}

1. Save as .json, run: bash {baseDir}/scripts/run-rstats.sh analyze --spec <path>
2. Read summary.json + report.md
3. Present: Summary → Statistics → Interpretation → Plots → Assumptions → Caveats

Analysis Selection#

For the complete 82-method table with user intent mapping, see references/METHOD_TABLE.md.

Quick lookup — most common:

Intentanalysis_typeCompare 2 groupsttest or wilcoxonCompare 3+ groupsanova or kruskalCategorical associationchisq or fisherPredict continuouslinear_regressionPredict binarylogistic_regressionSurvival curveskaplan_meierSurvival regressioncox_regressionMeta-analysismeta_analysisCausal effectpropensity_match or didPower/sample sizepower_analysis

Automatic Method Switching#

• Non-normal + n < 30 → wilcoxon over ttest
• Unequal variance → Welch t-test ( equal_var: false)
• Expected cells < 5 → fisher over chisq
• Overdispersion in Poisson → suggest negative binomial
• Heteroscedastic residuals → robust SE warning

Reporting Rules (Non-Negotiable)#

Every analysis MUST include:

• Sample size (n) and missing data handling
• Method name and rationale
• Point estimates with confidence intervals
• Effect sizes (Cohen's d, η², R², OR, HR, etc.)
• Assumption check results
• Limitations

Language: "associated with" / "evidence suggests" — NEVER "proves" / "causes"

Spec Field Reference#

See references/SPEC_REFERENCE.md for required/optional fields per analysis_type.