Mortality Data Description
The data include all records transcribed from the mortality census schedules from 1850 to 1880. Each record in the data represents a person who died within the previous year.
Each person's cause of death is recorded along with other characteristics for which availability varies across census years. The data are described in detail in this working paper, where their limitations and potential applications are discussed. See the variables page for a detailed listing of the available data fields.
Some key data limitations:
- Only people whose household persisted after their death are recorded in the mortality censuses. The data are therefore not fully representative of the population.
- Mortality data are missing for some localities and counties. Some states in each year have excess records compared to the published counts, likely due to duplicate entries that were not readily identifiable.
- The occupation of the deceased person is missing for most cases in 1860-1880.
- When possible, IPUMS links the mortality records to the full count population datasets in 1870 and 1880 (see the variable HISTID). It was not possible to create those links in 1850 and 1860, because family numbers were not recorded on the mortality schedules in those years. Cases derived from health board records in 1880 also cannot be linked.
Supported By
