Skip to main content

Table 4 Administrative dataset characteristics

From: Estimating parameters for probabilistic linkage of privacy-preserved datasets

  NSW(13,534,177 records) SA(2,509,914 records) WA(6,772,949 records)
Field Unique Values Missing % Discriminating Power Unique Values Missing % Discriminating Power Unique Values Missing % Discriminating Power
First Name 168,766 2.9% 8.61 124,849 5.5% 9.18 78,992 0.3% 8.54
Middle Name 114,686 54.2% 6.96 22,180 75.4% 7.19 61,241 40.8% 7.13
Last Name 291,595 0% 10.92 81,431 5.3% 10.81 123,481 0% 10.73
Dob Year 123 0% 6.47 115 0% 6.45 118 0% 6.39
Dob Month 12 0% 3.58 12 0% 3.58 12 0% 3.58
Dob Day 31 0% 4.94 31 0% 4.94 31 0% 4.94
Sex 2 0% 1.00 2 0% 1.00 2 0% 0.99
Address 3,084,889 1.5% 16.96 690,615 8.1% 14.92 1,350,796 0.2% 16.05
Suburb 49,843 0.5% 9.30 10,729 6.9% 7.85 5542 0.1% 7.73
Postcode 3947 0.8% 8.17 2238 8.5% 6.90 2319 0.2% 6.58