Skip to main content

Table 2 Characteristics of global similarity (gS) and atomic similarities (aS) in the identified pairs without exact concordance

From: Medical record linkage in health information systems by approximate string matching and clustering

Atomic similarities

 

gS

BN*

MN*

BN/MN*

First name

Date of birth

Pairs

28,517

12,093

988

16,013

12,066

11,578

Mean

0.92

0.79

0.78

0.36

0.77

0.82

Stand. error

0.48

0.20

0.23

0.25

0.17

0.07

Minimum

0.85

0.00

0.00

0.00

0.00

0.65

Percentiles

      

   25th

0.87

0.76

0.62

0.22

0.63

0.77

   50th

0.92

0.88

0.90

0.41

0.82

0.85

   75th

0.97

0.92

0.94

0.52

0.92

0.88

   90th

0.99

0.95

0.96

0.63

0.95

0.88

   95th

0.99

0.96

0.96

0.75

0.96

0.88

  1. * BN = birth name, MN = married name