Assess the effect of missing maternal-migration observations

Determine how systematic missingness in maternal migration status affects the maternal-migration-related findings and the inferred transition structure in the Agincourt Health and Demographic Surveillance System child-mortality analysis.

Background

The analysis excludes person-periods with unrecorded maternal migration rather than imputing them. Because maternal migration is identified by the Orbit Theory analysis as a major organiser of sequential transitions into child-death states, non-random observation gaps could alter the estimated prevalence and direction of those transitions.

The paper specifically leaves unresolved whether mothers who migrate out are less likely to be observed and, consequently, what directional effect this missingness has on the maternal-migration-related results.

References

If out-migrating mothers are less likely to be observed, this represents a conservative analytical choice whose directional effect on the MM-related findings is unknown.

— State-space representation of early-life child mortality dynamics: linking event history analysis and orbit-based models  (2610.07877 - Lefotlha et al., 6 Oct 2026) in Conclusion, limitations paragraph