docs(methodology): Binder (1983) survey TSL variance paper review

[igerber]

Summary

• Add docs/methodology/papers/binder-1983-review.md — a paper review of Binder (1983), "On the Variances of Asymptotically Normal Estimators from Complex Surveys" (International Statistical Review 51(3):279-292, DOI 10.2307/1402588), the canonical source for the design-based Taylor-series-linearization (TSL) variance used on the survey path.
• Captures the central sandwich V(θ̂) = J⁻¹ Σ_U J⁻ᵀ (Eq. 3.3/3.4), its OLS specialization S_XX⁻¹ Σ̂(B̂) S_XX⁻¹ (Eq. 4.7), the GLM / logit / log-linear cases (Eq. 4.4-4.15), the Appendix Conditions 1-11 / Lemma 1 / Corollaries 1-2, the Canada Health Survey design (Section 5), and a Gaps section noting what Binder does not prescribe (degrees of freedom, lonely-PSU handling, explicit FPC).
• Point the Survey Data Support tracker row at the new review and add Binder to the Related Documents list. Status stays In Progress — this is PR-A of the 2-PR methodology promotion; source validation, the dedicated methodology test file, R-parity table, deviations, and the flip to Complete land in PR-B.

Methodology references

• Method name(s): Survey Data Support — Binder (1983) design-based TSL variance.
• Paper / source link(s): Binder, D. A. (1983), International Statistical Review 51(3):279-292, DOI 10.2307/1402588.
• Any intentional deviations from the source (and why): None — paper-review-only; the doc records what the paper does and does not prescribe, deferring downstream conventions (df, lonely-PSU, FPC) to their own sources.

Validation

• Tests added/updated: No test changes — paper-review documentation only. Source validation and the dedicated tests/test_methodology_survey.py land in the PR-B follow-up.
• Backtest / simulation / notebook evidence (if applicable): N/A.

Security / privacy

• Confirm no secrets/PII in this PR: Yes

[github-actions]

Overall Assessment: ✅ Looks Good

Executive Summary

• No P0/P1 methodology, inference, variance/SE, or default-behavior issues found in the PR diff.
• Affected method: Survey Data Support, specifically Binder (1983) design-based TSL variance documentation.
• The added review’s core sandwich/OLS mapping aligns with the existing registry and survey-theory contract.
• Binder gaps for df, lonely PSU handling, FPC, and weight-type semantics are called out as downstream conventions, not hidden deviations.
• Remaining methodology promotion work is already tracked in METHODOLOGY_REVIEW.md.

Methodology

• Severity: P3 informational
  Impact: No action required. The review’s central TSL claims at docs/methodology/papers/binder-1983-review.md:L54-L76 match the registered survey TSL formula at docs/methodology/REGISTRY.md:L4111-L4117 and the survey-theory implementation description at docs/methodology/survey-theory.md:L452-L479.
  Concrete fix: None required.

• Severity: P3 informational
  Impact: The review correctly treats df rules, lonely-PSU handling, FPC, and weight-type semantics as not prescribed by Binder and as downstream/library conventions. Those conventions are already documented in docs/methodology/REGISTRY.md:L4121-L4137 and docs/methodology/REGISTRY.md:L4174-L4188.
  Concrete fix: None required.

Code Quality

• Severity: N/A
  Impact: No code changed.
  Concrete fix: None.

Performance

• Severity: N/A
  Impact: Documentation-only PR; no runtime path affected.
  Concrete fix: None.

Maintainability

• Severity: P3 informational
  Impact: METHODOLOGY_REVIEW.md links the new Binder review, but the Survey Data Support reference in docs/methodology/REGISTRY.md:L4113-L4115 still only names Binder/Lumley without a direct paper-review link. This is discoverability-only, not a methodology defect.
  Concrete fix: Optionally add “paper review on file: docs/methodology/papers/binder-1983-review.md” to the registry’s TSL reference line.

Tech Debt

• Severity: P3 informational
  Impact: The remaining promotion work is explicitly tracked: dedicated methodology tests, R parity table, deviation documentation, and cross-estimator gap enumeration are listed in METHODOLOGY_REVIEW.md:L1439-L1443 and reiterated at METHODOLOGY_REVIEW.md:L1488-L1490. This is acceptable deferred work.
  Concrete fix: None for this PR; complete those items in the planned follow-up.

Security

• Severity: N/A
  Impact: No secrets, credentials, or PII observed in the changed documentation.
  Concrete fix: None.

Documentation/Tests

• Severity: P3 informational
  Impact: No tests are added, but this PR is paper-review documentation only and does not alter estimator behavior. The missing dedicated methodology test file is already tracked in METHODOLOGY_REVIEW.md:L1440.
  Concrete fix: None required for approval; add the Binder-equation-numbered survey methodology tests in PR-B as already planned.