Statistics
AuraScore 81/100

Public Health Survey Sampling Frame and Calibration Weighting Verification Checklist

Audit complex sampling frames, stratifications, non-response adjustments, and survey calibration weights for public agencies.

Use this checklist when validating probability sample designs and weighting workflows for regional or national public health surveys. It ensures survey statisticians thoroughly check frame coverage, strata variances, rake weighting convergence, and design effect inflation prior to public data release.

Template

Role: Principal Survey Statistician for Municipal and Public Health Agencies

Context

  • Population Universe: {{target_population_definition}}
  • Registry Frame: {{sampling_frame_source}}
  • Strata & PSU Definitions: {{stratification_variables}}
  • Non-Response Profile: {{expected_non_response_rate}}
  • Calibration Targets: {{post_stratification_benchmarks}}
  • Variance Engine: {{variance_estimation_method}}

Task

Deliver an end-to-end technical audit checklist to verify sample design integrity, non-response bias adjustments, and post-stratification weighting before producing official public health statistics.

Method

  1. Inspect coverage errors and under-coverage bias between {{sampling_frame_source}} and {{target_population_definition}}.
  2. Verify selection probabilities and base design weights across {{stratification_variables}}.
  3. Audit unit non-response adjustments against expected patterns in {{expected_non_response_rate}}.
  4. Check convergence tolerances and extreme weight trimming rules during raking against {{post_stratification_benchmarks}}.
  5. Evaluate Design Effect (DEFF) and Effective Sample Size (n_eff) to detect excessive weight variance.
  6. Validate replicate weights or linearization formulas configured for {{variance_estimation_method}}.
  7. Establish cell suppression and coefficient of variation (CV) flagging thresholds for reporting.
  8. Verify compliance with public sector statistical standards for data privacy and disclosure control.

Constraints

  • Checklist MUST contain specific threshold guidance (e.g., max weight ratios, CV cutoff limits).
  • MUST NOT approve final survey weights without checking trimming thresholds and raking convergence.
  • All items MUST reflect complex survey design principles rather than simple random sampling assumptions.
  • Use unambiguous pass/warning/fail criteria for each verification step.

Output format

  • Markdown checklist categorized into distinct audit stages.
  • Part 1: Frame Coverage & PSU Verification (4-5 checklist items)
  • Part 2: Non-Response & Attrition Adjustments (4-5 checklist items)
  • Part 3: Calibration, Raking & Weight Trimming (5-6 checklist items)
  • Part 4: Variance Estimation & Reporting Safety (4-5 checklist items)
  • Part 5: Sign-Off Governance Table (Task, Metric Checked, Passing Criteria, Analyst Sign-off)

Self-review

  1. Does the checklist specifically reference {{variance_estimation_method}} in the standard error validation steps?
  2. Are post-stratification adjustments against {{post_stratification_benchmarks}} explicit?
  3. Are design effect (DEFF) formulas and weight dispersion checks included?
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

data-analytics
data-statistics
public-sector-nonprofit
survey statistics
sampling weights
public health