| name | methodology-explainer |
| description | Create a clear, transparent explanation of analytical methodology for any audience level. Activate when you deliver findings that require the audience to trust the method — A/B tests, attribution models, forecasts, statistical analyses, or anything where "how did you get that?" is a likely question. |
When to use
- Delivering A/B test results where the statistical method needs justification
- Presenting a forecast or model where assumptions must be communicated
- Documenting methodology for reproducibility and audit
- Onboarding a stakeholder who needs to understand and trust a recurring analysis
- Any analysis where "how did you get that?" is a likely question
Process
- Determine audience tier — classify as executive (decision-maker, no stats background), analyst (technically literate peer), or technical peer (will scrutinise the method). Tier determines depth and format.
- Select explanation pattern — choose: narrative (for executives), layered technical (for analysts — plain first, then detail on request), or Q&A format (for peer review). See
references/explanation_patterns.md.
- Draft the core explanation — cover: what question the analysis answers, what data was used, what method was applied, what assumptions were made, and what the key limitations are
- Simplify statistical language — replace p-values with confidence statements, model metrics with error margins, technical method names with purpose descriptions. See
references/stats_plain_language.md.
- Add a limitations section — every methodology explanation must include what the analysis cannot conclude and what would change the findings
- Produce deliverables — fill
assets/methodology_writeup_template.md or assets/methodology_slide_template.md depending on format
Inputs the skill needs
- Required: the method to explain — choose from the built-in list:
wilson_score, iqr_outlier, z_score, linear_regression, kmeans
- Required: target audience —
field_staff | manager | council | public
- Optional: known concerns or questions the audience is likely to raise
- Optional: format requirements (slide, written doc, email)
Quick start (script — no external data required):
python3 scripts/explain_method.py --method wilson_score --audience manager
python3 scripts/explain_method.py --method z_score --audience council
The script prints the formula, plain-language explanation, an NYC DOT example,
and the audience-tailored phrasing. For custom methods not in the list, use the
templates in assets/ directly.
Output
- Plain-language methodology write-up
- Limitations paragraph
assets/methodology_writeup_template.md or assets/methodology_slide_template.md (filled)