How good is probabilistic record linkage to reconstruct reproductive histories? Results from the Aberdeen children of the 1950s study

Table 4 Numbers of births^a (and corresponding median linkage scores) by whether or not there was a corresponding match between details (mother, sex, year of birth) in SMR 2 and questionnaire, classified according to differences in total number of births/mother between both sources.

		Total number of births to mother
Source of individual birth details		Same in SMR 2 and questionnaire	More in SMR 2	More in questionnaire	Total Number of births
In SMR 2 and questionnaire (exact match)	Number of births	4166	160	604	4930
	Median linkage score (25 ^th , 75 ^th centile)	35 (31, 40)	35 (32, 40)	32 (26, 36)
In SMR 2 only (exact match not found in questionnaire)	Number of births	113	131^b	39	283
	Median linkage score (25 ^th , 75 ^th centile)	36 (32, 41)	24 (23, 30)	27 (24, 41)
In questionnaire only (exact match not found in SMR 2)	Number of births	113	4	633	750
	Median linkage score (25 ^th , 75 ^th centile)	n.a.	n.a.	n.a.
	Total number of mothers	2061	94	449	2604

^a restricted to data on mothers for whom information was available from both linkage to SMR 2 and questionnaires.
^bOut of 135 additional births found above the diagonal in table 3, 4 births were in fact matches according to the overall information given by the mother in the questionnaire (they specified to have children, but omitted detailed information)
n.a.: not applicable

ISSN: 1471-2288