メニュー
Writes a plain-language explanation of a data journalism methodology — what data was used, where it came from, how it was processed, and what the analysis found — suitable for publication alongside a data story.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | methodology-explainer |
| description | Writes a plain-language explanation of a data journalism methodology — what data was used, where it came from, how it was processed, and what the analysis found — suitable for publication alongside a data story. |
| status | stable |
| category | data-journalism |
| subcategory | publishing |
| version | 1 |
| eval_score | 4.5 |
| tags | ["methodology","transparency","data journalism","publication","editorial"] |
Writes a plain-language explanation of a data journalism methodology — what data was used, where it came from, how it was processed, and what the analysis found — suitable for publication alongside a data story.
Required: A description of the data used (source, time period, scope). The analysis steps you performed (what you calculated, how you grouped or filtered data, any joins or merges with other datasets). The key findings. Any known limitations or caveats in the data or analysis. Optional: How the data was obtained (FOI, open data, licensed); what software or tools were used; whether the data was independently verified; whether any external expert reviewed the methodology.
A methodology statement of 300–500 words structured under four headings: Data Sources, How We Analysed It, What We Found, and Caveats and Limitations. Tone: clear, direct, confident but honest. Written in first-person plural ("we obtained," "we analysed") or third-person ("the analysis used"), according to publication style. No jargon unexplained.
Data used: National hospital waiting time records obtained via Freedom of Information request to the National Health Statistics Agency. Dataset covered Q1 2019 to Q3 2024, all 47 regional health authorities, all medical specialties. Analysis: Calculated median and 90th-percentile wait times per authority per quarter. Compared 2019 Q3 (pre-pandemic baseline) to 2024 Q3 to measure change. Ranked authorities by median wait time in 2024 Q3. Calculated the gap between highest and lowest authority as a ratio. Key finding: The highest-waiting authority (North Coastal) now has a median wait 2.2 times longer than the lowest-waiting authority (Central Metro). In 2019, that ratio was 1.4. Limitations: Data does not capture outcomes — we cannot say whether patients deteriorated while waiting. Duration_on_market field was 40% incomplete and was excluded from analysis. Tools: Python (pandas), verified in Excel.
Data Sources
The analysis is based on hospital waiting time records obtained via a Freedom of Information request to the National Health Statistics Agency. The dataset covers 22 consecutive quarters from January 2019 to September 2024 and includes records for all 47 regional health authorities across all medical specialties. The records were received in February 2026.
How We Analysed It
For each health authority, we calculated median waiting times and 90th-percentile waiting times — the point below which 90% of patients are seen. We used the 90th-percentile figure alongside the median because it captures the experience of the longest-waiting patients, which median figures alone can obscure.
We compared each authority's figures for Q3 2024 against Q3 2019, which we used as the pre-pandemic baseline. To measure inequality between authorities, we calculated the ratio of the highest-waiting authority's median to the lowest-waiting authority's median. Analysis was conducted in Python and independently verified in Excel.
What We Found
The gap between the longest and shortest-waiting authorities has grown substantially since 2019. In Q3 2024, the highest-waiting health authority has a median wait 2.2 times longer than the lowest — compared to a ratio of 1.4 in 2019. North Coastal Authority recorded the longest median wait at 68 days; Central Metro recorded the shortest at 31 days.
Caveats and Limitations
The data records how long patients wait, not what happens to them while they wait. We cannot conclude from this data alone that longer waits caused patient harm — that would require outcome data, which was not available. One field in the dataset (days listed on market) was approximately 40% incomplete; this field was excluded from all analysis.
Waiting time data reflects completed waits — patients still waiting at the time of the data extract are not included, which means current wait times may be underrepresented, particularly for longer waiters.
The full dataset and analysis code are available on request.
Generates the complete set of metadata fields — tags, categories, keywords, slug, and summary — needed to publish a piece of content in a content management system, ready to copy and paste into each field.
Takes a finished piece of content and produces a complete, archive-ready metadata record — including descriptive tags, subject classifications, rights notes, and an archival summary — for entry into a content management or digital archive system.
Drafts a GDPR compliance note for a specific piece of journalistic content or data collection activity — documenting the lawful basis for processing personal data, what data is held, how long it is retained, and who has access.
Reviews the details of a specific image and flags potential rights clearance issues before publication, identifying which permissions you likely need and which questions to bring to your legal or picture desk.
Reviews a story brief, draft article, or broadcast script and flags potential legal risks — including defamation, privacy, contempt of court, and data protection — so you know what to check before publication.
Drafts a formal, professional request letter to a music rights holder asking for permission to use a track in a specific media production under defined terms.
