The Oldest Old sample contains approximately 1,500 men whose ages at death are confirmed to be over 95. Researchers initially compiled a list of nearly 6,000 potential nonagenarians. The list was based upon many sources, including:
Gravestone databases
Obituaries
Newspapers accounts drawn from the National Tribune
Grand Army of the Republic death rolls
List of attendees at the Last Reunion of the Blue and Gray in 1938
1930 U.S. Census
1940 U.S. Census
To confirm an age of over 95 at death, researchers examined the pension files for these veterans at the National Archives and Records Administration in Washington, D.C. These files usually contained detailed information on birth and death, and researchers collected the entire pension file of any veteran confirmed to be at least 95 at death. These records include detailed surgeons’ examinations and a wealth of health data. Researchers then linked these veterans to the 1850, 1860, 1870, 1880, 1900, 1910, 1920, 1930, and 1940 censuses. Because so much information is available about men who lived to late ages and because electronic indexes make linking efficient, researchers were able to achieve extremely high linkage rates.
Decade
1850
1860
1870
1880
1900
1910
1920
1930
1940
% Found
77.16
78.57
81.51
91.64
94.54
94.94
92.87
94.44
89.39
Even though the Oldest Old sample uses familiar records, the data contain additional information from private physician affidavits and home visits by Veterans Administration representatives. These accounts provide some information on care arrangements, functional limitations (ability to walk, blindness, deafness), and mental acuity that are not generally found for younger soldiers.
A unique 10-digit identification number, stored in the variable recidnum, identifies each recruit throughout the separate data sets of the Early Indicators projects.
Supported by
the National Institute on Aging grants #P01 AG010120, #P01 AG010120-16, #R01 AG027960, #P01AG010120, and #R21AG064460