Starter Writing skill
Write a data methodology note
Write a data methodology note that explains sources, population, definitions, collection, transformation, calculation, quality, uncertainty, revision, and limitations.
Make the path from source to published result traceable.
When to use
- Use for public statistics, dashboards, research, reports, metrics, models, or analytical releases.
- Do not claim representativeness, accuracy, causality, or comparability beyond the evidence.
Procedure
- Identify publication, audience, version, owner, purpose, outputs, and decision uses.
- Describe sources, collection authority, population, sampling frame, period, geography, coverage, and exclusions.
- Define units, fields, events, numerator, denominator, dimensions, time, currency, and key terms.
- Explain cleaning, joins, deduplication, weighting, imputation, aggregation, adjustments, suppression, and confidentiality.
- Report missingness, quality checks, error, uncertainty, revisions, breaks, comparability, and known bias.
- Link data dictionary, code or query where publishable, source versions, licenses, and contact.
- Reproduce representative values and obtain statistical or domain review.
--- name: write-a-data-methodology-note category: write description: Write a data methodology note that explains sources, population, definitions, collection, transformation, calculation, quality, uncertainty, revision, and limitations. Use when readers need to reproduce or correctly interpret a dataset, metric, chart, or report. --- # write-a-data-methodology-note Make the path from source to published result traceable. ## When to use - Use for public statistics, dashboards, research, reports, metrics, models, or analytical releases. - Do not claim representativeness, accuracy, causality, or comparability beyond the evidence. ## Procedure 1. Identify publication, audience, version, owner, purpose, outputs, and decision uses. 2. Describe sources, collection authority, population, sampling frame, period, geography, coverage, and exclusions. 3. Define units, fields, events, numerator, denominator, dimensions, time, currency, and key terms. 4. Explain cleaning, joins, deduplication, weighting, imputation, aggregation, adjustments, suppression, and confidentiality. 5. Report missingness, quality checks, error, uncertainty, revisions, breaks, comparability, and known bias. 6. Link data dictionary, code or query where publishable, source versions, licenses, and contact. 7. Reproduce representative values and obtain statistical or domain review. ## Done - A methodology note documents sources, population, definitions, transformations, calculations, quality, uncertainty, revisions, limits, and links - Reproduction, denominator, source, period, coverage, privacy, uncertainty, comparability, and owner-review checks verify interpretation