Racial and socioeconomic gaps in mortality persisted throughout the twentieth century. We know little, however, about how racial or socioeconomic gaps in mortality were related to each other or how cause-specific mortality evolved over the twentieth century more generally. Demographers have repeatedly documented serious data problems that limit our ability to analyze these issues. In an attempt to overcome these problems, we link a random sample of death certificates taken at five year intervals from 1910 to 1975 to the manuscript federal census files of the deceased's early in life and then to the death certificates of the deceased's parents. To our knowledge, the data we construct is the first of its kind in linking parent and child death certificate information with the additional information from the census files. We show that our research design allows us to construct a panel data set that allows us to look at mortality (both general and cause-specific) over time and for specific cohorts. This paper presents preliminary evidence from our pilot study of death certificates from the Carolinas in the twentieth century, documenting racial and occupational differences in mortality over the twentieth century. We outline several avenues of future research to be investigated with this data.
Country of focus: United States of America.