data-analysis

name	data-analysis
description	Structured data analysis with statistical reasoning
version	1.0.0
tags	["data","analysis","statistics","reasoning"]

When to Use

Use this skill when analyzing datasets, interpreting numbers, identifying trends, comparing metrics, or drawing conclusions from quantitative information.

Steps

1. Understand the data: Identify what each field/column represents, the data types, units, and time ranges. Note any metadata.
2. Check quality: Look for missing values, outliers, duplicates, and inconsistencies. Note data limitations before drawing conclusions.
3. Explore patterns: Calculate summary statistics (mean, median, range, distribution shape). Identify trends, clusters, and anomalies.
4. Compare: When comparing groups or time periods, use appropriate baselines and normalize for relevant factors (size, time, population).
5. Interpret: Explain what the patterns mean in context. Distinguish between correlation and causation.
6. Present: Choose the clearest way to communicate findings (tables, ranked lists, percentages, comparisons).

Rules

• Always state the sample size and time period when presenting statistics.
• Use percentages AND absolute numbers together — "15% (150 out of 1,000)" is clearer than either alone.
• Do not imply causation from correlation without explicit justification.
• Acknowledge confounding variables and alternative explanations.
• Round numbers appropriately — false precision (34.7823%) implies more accuracy than the data supports.
• When comparing, always specify the baseline: "20% higher than last quarter" not just "20% higher."

Output Format

1. Summary: Key finding in 1-2 sentences.
2. Data Overview: What data was analyzed, its scope, and any quality issues.
3. Findings: Organized by theme, with specific numbers and comparisons.
4. Interpretation: What the findings mean in context, including limitations.
5. Recommendations: Actions suggested by the data (if applicable).