Working Paper: NBER ID: w16134
Authors: J. David Hacker; Michael R. Haines
Abstract: This paper constructs new life tables for the American Indian population in the late nineteenth and early nineteenth centuries, thus pushing back the availability of age-specific mortality and life expectancy estimates nearly half a century. Because of the lack of reliable vital registration data for the American Indian population in this period, the life tables are constructed using indirect census-based estimation methods. Infant and child mortality rates are estimated from the number of children ever born and children surviving reported by women in the 1900 and 1910 Indian censuses. Adult mortality rates are inferred from the infant and child mortality estimates using model life tables. Adult mortality rates are also estimated by applying the Preston-Bennett two-census method (1983) to the 1900-1910 intercensal period.
Keywords: American Indian mortality; life tables; census data; demographic estimation
JEL Codes: I1; I3; N11
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| infant and child mortality rates (J13) | adult mortality rates (J11) |
| American Indian life expectancy at birth (I14) | child mortality rates (J13) |
| child mortality rates (J13) | American Indian life expectancy at birth (I14) |
| American Indian life expectancy at birth in 1940 (N32) | life expectancy estimates in the early 1900s (J17) |
| census data (C80) | mortality estimates (J17) |
| age misstatement and differential enumeration (C83) | mortality estimates (J17) |