Table 6. Results of the Cox regression estimating the impact of the relative size of the lineages on the mortality of reproductive women.
Hazard ratios are presented together with indicators of statistical significance (* p<0.05, + p<0.1). Robust standard errors are given in parentheses. All models control for ID of interest’s birth cohort, birth order, marital status (husband alive), postpartum period, number of births, number of living offspring, and socio-economic status of the current marriage. Full models are presented in S10 Table.
| Spatial1 |
||||
|---|---|---|---|---|
| Unweighted (simple count) |
Weighted (numbers are weighted with coefficient of relatedness) |
|||
| Clustered2 | Fixed-effect3 | Clustered2 | Fixed-effect3 | |
| Relative sizes of lineages (Ref. natal lineage only) | ||||
| Natal lineage was larger | 1.078 | 1.274 | 1.118 | 1.263 |
| (0.128) | (0.370) | (0.128) | (0.354) | |
| Lineages were equal in size | 1.074 | 1.117 | 1.503* | 2.201 |
| (0.181) | (0.418) | (0.297) | (1.101) | |
| In-law lineage was larger | 0.902 | 0.822 | 0.781+ | 0.703 |
| (0.122) | (0.258) | (0.103) | (0.206) | |
| In-law lineage only | 0.907 | 1.196 | 0.906 | 1.237 |
| (0.080) | (0.264) | (0.080) | (0.275) | |
| N women | 4,914 | 2,908 | 4,914 | 2,908 |
| Dead | 833 | 494 | 833 | 494 |
| N families (cluster and strata, respectively) | 3,201 | 1,195 | 3,201 | 1,195 |
| Observations | 167,699 | 106,980 | 167,699 | 106,980 |
| Log pseudolikelihood | -6345.03 | -247.62 | -6341.04 | -245.72 |
1 –dummies only consider living kin who were residing in the same parish as the ID
2 –each individual is compared to all other reproductive women in the sample
3 –each individual is compared to her reproductive sisters