Skip to main content
. Author manuscript; available in PMC: 2024 Aug 17.
Published in final edited form as: Hist Methods. 2023 Aug 17;56(3):138–159. doi: 10.1080/01615440.2023.2239699

Table 1.

Example of Sets Displayed to Trainer for Birth Certificate to Census Linking

A. Information Sets May Decrease Trainer Certainty about a Match
Record Name First/Last Name
Commonality Scores
Age Birthplace
4 Candidate Jason O’Sullvan (0.58/0.32) 33 Ohio
3 Candidate Jason O’Sullivon (0.58/0.32) 35 Ohio
2 Candidate Jason O. Sullivan (0.58/0.84) 33 Ohio
1 Candidate Jason O’Sullivan (0.58/0.84) 35 Ohio
Primary Jason O’Sullivan (0.58/0.84) 35 Ohio
B. Information Sets May Increase Trainer Certainty about a Match
Record Name First/Last Name
Commonality Scores
Age Birthplace
7 Candidate Susan H. Ovie (0.76/0.078) 31 North Carolina
6 Candidate Shelly H. Olive (0.53/0.58) 35 North Carolina
5 Candidate Sheilagh H. Oglvie (0.091/0.098) 31 North Carolina
4 Candidate Sally H. Ogivie (0.76/0.050) 31 North Carolina
3 Candidate Shelly O’Neill (0.53/0.71) 33 North Carolina
2 Candidate Sylvie H. Ogilbie (0.40/0.28) 29 North Carolina
1 Candidate Cecela F. Ogilvie (0.35/0.53) 34 North Carolina
Primary Shelagh Harris Ogilvie (0.20/0.53) 31 North Carolina

Notes: Name commonality scores are computed as a ratio, a/b, where a is the log count of the first or last name under consideration in the 1940 Census and b is the log count of the most common first or last name in the 1940 Census. These examples are truncated to a handful of candidate links for ease of presentation. Training sets typically included up to 30 individuals.