The problem: The death toll for soldiers killed in the Civil War has long been cited as 620,000 — a number also considered to be inaccurate because neither the Union nor the Confederacy kept standardized personnel records.
The researcher: Binghamton University historian J. David Hacker.
The research: Hacker analyzed census data to arrive at what he says is a more accurate number of deaths, about 750,000. His findings will be published in December in the journal Civil War History.
The strategy: There are problems using census data to estimate mortality, Hacker explains. “You can track the number of people of certain ages from one census to the next, and you can see how many are missing. But the potential problem with that is that each census undercounted people by some unknown amount, and an unknown number of people moved in and out of the country between censuses.”
However, new data sets produced in the past 10 years or so, instead of giving the aggregate number of people in certain age groups, identify each person and his or her age, race and birthplace. Hacker realized that civilian deaths were so low relative to soldiers’ deaths that he could compare the number of native-born men missing in the 1870 Census relative to the number of native-born women missing and produce an estimate from that.
Hacker looked at the ratio of male survival relative to female survival for each age group. He established a “normal” pattern in survival rates for men and women by looking at the numbers for 1850–1860 and 1870–1880. Then he compared the war decade, 1860–1870, relative to the pattern.
His new estimate of Civil War deaths contains a wide margin: 650,000 to 850,000, with 750,000 as the central figure.
Hacker discusses his analysis of the Civil War death toll.