pith. sign in

arxiv: 1906.12000 · v1 · pith:FMSWDPJRnew · submitted 2019-06-28 · 📊 stat.AP

Estimating adult death rates from sibling histories: A network approach

Dennis M. Feehan , Gabriel M. Borges This is my paper

Pith reviewed 2026-05-25 13:55 UTC · model grok-4.3

classification 📊 stat.AP
keywords sibling survival methodadult mortality estimationnetwork samplingdemographic surveysstatistical estimatorsdeath registrationmortality data
0
0 comments X

The pith

Re-framing sibling survival data as a network sampling problem enables formal derivation of statistical estimators for adult mortality.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

In countries without complete death registration, surveys ask respondents to list their siblings and report each one's survival status to estimate adult death rates. The paper shows that treating these reports as draws from an underlying network allows formal statistical estimators to be derived. This approach makes the required conditions for valid estimates explicit. It also produces internal consistency checks for assessing data quality and identifies quantities that could be measured to relax assumptions in future work.

Core claim

By modeling sibling histories as a network sampling problem, statistical estimators for adult death rates can be formally derived from sibling survival data; the derivation clarifies the precise conditions under which the estimates hold, supplies consistency checks, and reveals measurable quantities that could relax current assumptions.

What carries the argument

The re-framing of the sibling survival method as a network sampling problem, which applies network sampling theory to derive estimators from reports of siblings' survival status.

If this is right

  • Adult mortality estimates gain formal statistical grounding with explicit assumptions.
  • Data quality can be assessed through internal consistency checks derived from the network model.
  • Future data collection can target quantities that relax the current assumptions.
  • The method applies only when sibling reports satisfy the network sampling conditions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar network approaches might apply to other kinship-based demographic estimates.
  • Integration with administrative data could validate or adjust the network-derived estimates.
  • Survey designs could be optimized to collect the additional quantities suggested by the model.

Load-bearing premise

The sibling history data can be validly modeled as a network sampling problem under the precise conditions required for the derived estimators to apply.

What would settle it

If the network-derived estimators applied to real sibling data produce mortality rates that diverge systematically from independent sources, or if the proposed consistency checks fail without identifiable reporting errors, the modeling assumption would be challenged.

Figures

Figures reproduced from arXiv: 1906.12000 by Dennis M. Feehan, Gabriel M. Borges.

Figure 1
Figure 1. Figure 1: Framing sibling survival as a network reporting estimator. (a) A population connected [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Aggregate and individual visibility estimates for log-adult death rates from the Malawi [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Sensitivity to invisible deaths. (a) Estimated fraction of respondents to the Malawi 2000 DHS who would not be visible in sibling history data. (b) Illustration of the relative error in using the visible death rate MV as an estimate for the total death rate M. The proportion of exposure that is invisible, p I Nα , varies along the x axis; the relationship between the visible and invisible death rates, capt… view at source ↗
Figure 4
Figure 4. Figure 4: Internal consistency checks using sibling reports from the 2000 Malawi DHS sibling [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Network reporting when respondents are not included in their reports. (a) A single sibship. [PITH_FULL_IMAGE:figures/full_fig_p029_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Illustration of the relative error in using the visible death rate [PITH_FULL_IMAGE:figures/full_fig_p039_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparing three variants of adjusted aggregate visibility estimates to the unadjusted [PITH_FULL_IMAGE:figures/full_fig_p069_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Overview of the method used to construct a pseudo-universe that forms the basis for the [PITH_FULL_IMAGE:figures/full_fig_p075_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Age-sex distribution of the simulated population. [PITH_FULL_IMAGE:figures/full_fig_p076_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Age- and sex-specific death rates among members of the simulated population. [PITH_FULL_IMAGE:figures/full_fig_p076_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: True age-specific death rates (x axis) against adjusted and unadjusted death rate [PITH_FULL_IMAGE:figures/full_fig_p078_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: True age-specific death rates (x axis) against adjusted and unadjusted death rate [PITH_FULL_IMAGE:figures/full_fig_p079_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Mean square error, squared bias, and variance for the individual and aggregate visibility [PITH_FULL_IMAGE:figures/full_fig_p082_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Mean square error, squared bias, and variance for the individual and aggregate visibility [PITH_FULL_IMAGE:figures/full_fig_p083_14.png] view at source ↗
read the original abstract

Hundreds of millions of people live in countries that do not have complete death registration systems, meaning that most deaths are not recorded and critical quantities like life expectancy cannot be directly measured. The sibling survival method is a leading approach to estimating adult mortality in the absence of death registration. The idea is to ask a survey respondent to enumerate her siblings and to report about their survival status. In many countries and time periods, sibling survival data are the only nationally-representative source of information about adult mortality. Although a huge amount of sibling survival data has been collected, important methodological questions about the method remain unresolved. To help make progress on this issue, we propose re-framing the sibling survival method as a network sampling problem. This approach enables us to formally derive statistical estimators for sibling survival data. Our derivation clarifies the precise conditions that sibling history estimates rely upon; it leads to internal consistency checks that can help assess data and reporting quality; and it reveals important quantities that could potentially be measured to relax assumptions in the future. We introduce the R package siblingsurvival, which implements the methods we describe.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The paper re-frames the sibling survival method for estimating adult mortality as a network sampling problem. This re-framing is used to formally derive statistical estimators, clarify the precise conditions required for validity (such as accurate reporting of sibling ties and known sampling probabilities), provide internal consistency checks for data quality, and introduce the R package siblingsurvival that implements the estimators.

Significance. If the derivation holds under the stated assumptions, the work supplies a transparent statistical foundation for a method that is often the only source of nationally representative adult mortality data in countries lacking vital registration. The network perspective yields explicit conditions, consistency diagnostics, and an open R package, all of which support reproducibility and future relaxation of assumptions via additional measurements.

minor comments (3)
  1. The abstract states that the approach 'clarifies the precise conditions' and 'reveals important quantities that could potentially be measured'; the main text should include an explicit enumerated list or table of these conditions and quantities, with cross-references to the derivation steps.
  2. Figure or table captions should state the exact sample sizes, time periods, and countries used in any empirical illustration so that readers can assess generalizability without searching the text.
  3. The R package is mentioned; the manuscript should include a short code snippet or vignette reference demonstrating the core estimator call and the consistency-check output.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive and accurate summary of our manuscript, the assessment of its significance for adult mortality estimation in data-scarce settings, and the recommendation for minor revision. No specific major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained

full rationale

The paper reframes the sibling survival method as a network sampling problem and derives statistical estimators under explicitly stated assumptions (accurate reporting of ties, known sampling probabilities). It supplies the derivation, internal consistency checks, and an R package implementation. No load-bearing self-citation chains, self-definitional steps, or fitted inputs renamed as predictions appear in the provided text or abstract. The central claim rests on the network model and its consequences rather than reducing to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on treating sibling histories as network samples; no free parameters, invented entities, or additional axioms are identifiable from the abstract alone.

axioms (1)
  • domain assumption Sibling survival data can be represented and analyzed as a network sampling problem
    This is the foundational re-framing that enables the derivation of estimators.

pith-pipeline@v0.9.0 · 5718 in / 1069 out tokens · 33568 ms · 2026-05-25T13:55:50.071301+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages

  1. [1]

    Quantity Versus Quality: A Survey Experiment to Improve the Network Scale-up Method

    “Quantity Versus Quality: A Survey Experiment to Improve the Network Scale-up Method.” American Journal of Epidemiology, March, kwv287. Gakidou, E., and G. King. 2006. “Death by Survey: Estimating Adult Mortality Without Selection Bias from Sibling Survival Data.”Demography 43 (3): 569–85. http://www.springerlink.com/index/ W2Q1X41501666JL0.pdf. Garenne, ...

    work page doi:10.2307/2137971 2006

  2. [2]

    Reporting Errors in Siblings’ Survival Histories and Their Impact on Adult Mortality Estimates: Results from a Record Linkage Study in Senegal

    “Reporting Errors in Siblings’ Survival Histories and Their Impact on Adult Mortality Estimates: Results from a Record Linkage Study in Senegal.” Demography 51 (2): 387–411. http://link.springer.com/article/10.1007/s13524-013-0268-3. Helleringer, Stéphane, Gilles Pison, Bruno Masquelier, Almamy Malick Kanté, Laetitia Douillot, Géraldine Duthé, Cheikh Sokh...

    work page doi:10.1007/s13524-013-0268-3 2014

  3. [3]

    Adult Mortality from Sibling Survival Data: A Reappraisal of Selection Biases

    Zomba, Malawi: National Statistical Office. Masquelier, Bruno. 2013. “Adult Mortality from Sibling Survival Data: A Reappraisal of Selection Biases.” Demography 50 (1): 207–28. http://link.springer.com/article/10.1007/s13524-012-0149-1. Masquelier, Bruno, and Catriona Dutreuilh. 2014. “Sibship Sizes and Family Sizes in Survey Data Used to Estimate Mortality...

    work page doi:10.1007/s13524-012-0149-1 2013

  4. [4]

    We sample with replacement, so some sibships are sampled multiple times, and each sibship is sampled with probability proportional to its visibility to the frame population, i.e

    Starting from the 13,161 sibships in the dataset, sampleMsibships to form the pseudo-population of sibships. We sample with replacement, so some sibships are sampled multiple times, and each sibship is sampled with probability proportional to its visibility to the frame population, i.e. the number of sibship members, including the respondent, who were eli...

    work page

  5. [5]

    (This accounts for the fact that only females were interviewed in Malawi; without this step, we would end up with an unrealistic gender distribution.)

    For each of theMsibships resampled sibships, with probability 1 2, we flip the sexes of the reported siblings. (This accounts for the fact that only females were interviewed in Malawi; without this step, we would end up with an unrealistic gender distribution.)

    work page

  6. [6]

    The result is a universe of siblings who are assigned to sibships that is approximately representative of the 2000 Malawi sibship population

    We then form a universe of siblings from the individual siblings corresponding to the (possibly gender-flipped) resampled sibships. The result is a universe of siblings who are assigned to sibships that is approximately representative of the 2000 Malawi sibship population. Figure 9 and Figure 10 show the age-sex distribution and death rates in the simulate...

    work page 2000