Baseline model and interaction effect between grandparental death and geographic proximity to adult child on annual probability of parity progression, pooled across parities: Linear fixed-effects models
| . | Prob(Any Birth)a . | |
|---|---|---|
| . | Baseline . | Locality × Grandparent Dead . |
| Any Grandparent Dead | −0.0190*** | −0.0153*** |
| (0.00249) | (0.00300) | |
| Same Locality | 0.0119* | |
| (0.00544) | ||
| Any Grandparent Dead × Same Locality | −0.00800* | |
| (0.00367) | ||
| Number of Observations | 607,998 | 607,998 |
| R2 (within) | .225 | .225 |
| . | Prob(Any Birth)a . | |
|---|---|---|
| . | Baseline . | Locality × Grandparent Dead . |
| Any Grandparent Dead | −0.0190*** | −0.0153*** |
| (0.00249) | (0.00300) | |
| Same Locality | 0.0119* | |
| (0.00544) | ||
| Any Grandparent Dead × Same Locality | −0.00800* | |
| (0.00367) | ||
| Number of Observations | 607,998 | 607,998 |
| R2 (within) | .225 | .225 |
Notes: Standard errors are shown in parentheses. The table is abbreviated; full model results are available from the authors upon request. Variables not shown are age of each grandparent (ref. = 60–64), age of wife (ref. = 25–29), age of husband (ref. = 25–29), years of exposure since last birth (ref. = Year 1), and calendar year (ref. = 1986).
Source: Israeli linked population register and census files, 1986–2014.
Prob(Any Birth) refers to annual probability of parity progression to birth n conditional on achieving parity n – 1 for any n from 2 to 5.
p < .05; ***p < .001