4. LINKAGE RESULTS

Of the 177,380 death records, 159,657 (90.0%) records were linked to one of 22,485,854 eligible Census records. Of the 3,246 Aboriginal and Torres Strait Islander death records, 2,315 (71.3%) were linked.

Examination of the characteristics of the links identified and the unlinked records can be found in 4.3 Characteristics of linked and unlinked Death registrations.


4.1 LINKAGE ACCURACY

The following quality measures were calculated for the linkage and indicate a good level of overall quality:

• The linkage rate, 90%, being the proportion of Death registrations linked to a 2016 Census record; and
• The estimated proportion of correctly linked records, otherwise referred to as 'linkage precision'.

4.2 LINKAGE PRECISION

Not all record pairs assigned as links in a data linkage process are a true match, that is, a record pair belonging to the same individual. While the methodology is designed to ensure that the vast majority of links are true, some are actually false, i.e. the records in the link belong to different people rather than the same person. The linkage strategy used for the project was designed to ensure a high level of accuracy. Accordingly, the strategy was restrictive and conservative.

One of the key measures of linkage quality is the proportion of links in the dataset that are false. The number of false links is able to be estimated through the use of methods such as clerically reviewing a sample of links, or by using modelling techniques. Once an estimate of the number of false links is obtained, a 'precision' can be calculated. The precision is an estimate of the proportion of links that are matches (i.e. belonging to the same entity).

Once the precision of the dataset is estimated, the false link rate is easily calculated.

The estimated link precision of the 2016 Death Registrations to Census linkage dataset is 100% as the decision model did not allow for any false links. As previously discussed, the upper cut-off was set such that it was estimated there were no false links above the cut-off while the clerical review process only accepted links for which there was sufficient evidence to support them being accurate matches. In reality, there will be a small number of false links due to a slight degree of inconsistent decisions between clerical reviewers. While the number of false links is not able to be quantified precisely, the proportion is expected to be very small.


4.3 CHARACTERISTICS OF LINKED AND UNLINKED DEATH REGISTRATIONS

TABLE 3 - CENSUS AND DEATH REGISTRATIONS, Australia

Description		Records
Number
	Census records eligible for linking(a)	22 485 854
	Aboriginal and Torres Strait Islander Census records	649 171
	Records on death file(b)	177 380
	Death records linked	159 657
	Death records not linked	17 723
	Aboriginal and Torres Strait Islander records on death file(c)	3 246
	Aboriginal and Torres Strait Islander records linked(c)	2 315
Percent
	All death records linked	90.0
	Aboriginal and Torres Strait Islander death records linked	71.3

(a) Excludes residents temporarily overseas on Census night, imputed records and Census net undercount adjustment.
(b) Deaths which occurred between 09 Aug 2016 and 28 Sep 2017.
(c) According to Indigenous status reported on death registration form.

The number and percentage of death records linked to Census records by selected characteristics of deceased persons are presented in Table 4. A slightly higher linkage was achieved for females (91.4%) compared with males (88.6%). The linkage rate varied considerably by age, being lowest for 0-14 year old deceased persons (63.4%). This may be due to the comparatively high Census undercount rate in this age group. The linkage rate was highest for 75 years and older deceased persons (92.9%).

TABLE 4 - DEATH REGISTRATIONS LINKED TO CENSUS RECORDS BY SELECTED CHARACTERISTICS, Australia

		Total death records	Linked records	Linked records
Reported characteristics in death registration		no.	no.	%
Sex
	Males	91 143	80 796	88.6
	Females	86 237	78 861	91.4
Age (years)
	0-14	686	435	63.4
	15-24	1 219	868	71.2
	25-44	5 704	3 974	69.7
	45-64	22 543	18 582	82.4
	65-74	27 979	25 066	89.6
	75 and over	119 247	110 730	92.9
Indigenous Status
	Aboriginal and Torres Strait Islander	3 246	2 315	71.3
	Non-Indigenous	173 186	156 546	90.4
	Not stated	948	796	84.0
State of usual residence
	New South Wales	59 887	54 077	90.3
	Victoria	43 130	38 915	90.2
	Queensland	34 017	30 463	89.6
	South Australia	15 349	14 045	91.5
	Western Australia	16 269	14 439	88.8
	Tasmania	5 315	4 832	90.9
	Northern Territory	1 077	785	72.9
	Australian Capital Territory	2 294	2 066	90.1
Marital status
	Never married	18 547	14 774	79.7
	Married	70 298	64 811	92.2
	Widow	64 004	59 249	92.6
	Divorced	18 268	15 807	86.5
	Separated	152	123	80.9
	Not applicable (<15 years)	6 111	4 893	80.1
Elapsed time between Census and death
	Within 6 months of Census	80 234	71 238	88.8
	Beyond 6 months of Census	97 146	88 419	91.0

The linkage success varied by state of usual residence as reported on the death registration form. Rates were highest for South Australia (91.5%) and lowest for the Northern Territory (72.9%). All other states and territories had linkage rates between 88.8% and 90.9%. The low linkage rate for the Northern Territory reflects comparatively low linkage rates for both the Aboriginal and Torres Strait Islander and non-Indigenous populations. The linkage rate was similar for married and widowed persons (92.2 and 92.6% respectively). The linkage rate was lower for deaths which occurred within six months of the Census (88.8%) than those which occurred beyond six months after the Census (91.0%).

The linkage success also varied by Indigenous status recorded on the death registration form. People of non-Indigenous origin on the death registration form had a considerably higher linkage success (90.4%) compared with people of Aboriginal and Torres Strait Islander origin (71.3%). A more strict approach to implementing the 2016 linkage clerical review resulted in a lower, but more accurate linkage rate than in 2010-2012.

TABLE 5 - DEATH REGISTRATIONS LINKED TO CENSUS RECORDS BY STATE OF USUAL RESIDENCE AND INDIGENOUS STATUS, Australia

	Indigenous Status
State of Usual Residence	Aboriginal and Torres Strait Islander	Non-Indigenous	Not Stated (a)
New South Wales	676	52 415	474
Victoria	145	38 689	115
Queensland	663	29 997	66
South Australia	143	13 887	15
Western Australia	354	13 987	98
Tasmania	46	4 787	-
Northern Territory	272	501	-
Australian Capital Territory	16	2 283	24
Total	2 315	156 546	796

(a) Small cell counts have been suppressed to preserve confidentiality.

4.4 REASONS FOR UNLINKED DEATH REGISTRATIONS

There were two main reasons why death registrations were not linked to a Census record:

1. Records belonging to the same individual were present in the Death registration and Census datasets but these records failed to be linked because they contained missing or inconsistent information; or
2. A link was not possible because there was no Census record corresponding to the death registration as the person was missed from the Census. Proximity of death to Census night is a significant factor in the ability for a link to be achieved.

MISSING AND/OR INCONSISTENT INFORMATION

The quality of a data linkage project is significantly dependent on the quality of three key sources of information, these being name, address and date of birth. When all three sources of information are of very high quality on the linking datasets, identifying true links becomes less complicated, resulting in a high quality outcome for the linkage.

In some cases, the true match was present in the pool of all record pairs but it was not identified because there was a high level of inconsistency between information on the Death registration and the 2016 Census record, or key linking fields were missing from one or both datasets. The reasons for the match being missed can be categorised into the following groups:

• the missing or inconsistent information did not allow the record pair to be compared in the same blocking categories and could not be linked;
• the record pair did not contain enough unique common information to distinguish the match from other potential record pairs;
• the record pair was linked, but was attributed a low link weight as it contained substantial missing or inconsistent information and was positioned below the cut-off identified in sample clerical review; or
• the record pair was subjected to clerical review, but the high level of inconsistency prevented it from being deemed a true link.

Inconsistent Census information may be recorded due to a range of factors, including:

• transcription errors in the Census, where the wrong category is selected or the information is transposed, such as the day the person was born being reported in the month field instead of in the day field;
• data capture errors, where the Census form is scanned using Optical Character Recognition (OCR) software and certain characters may be misclassified, such as a 1 captured as a 7 or a 3 as an 8;
• reporting errors, where information is given for the wrong member of the household (e.g. person 1's information is reported for person 3) or where the person completing the Census form for a household guesses or estimates information about a fellow household member; or
• information that was not stated by the respondent and has been imputed as part of Census processing (such as age or sex), while set to missing for linking, the imputed values are included in the analytical dataset.

Accurate address coding was crucial in narrowing the search and differentiating between true and false links. However the nature of the data did not always allow for linkage on address or geography to be possible, as people may have changed address after Census night and prior to their death (e.g. moving to a nursing home); therefore the address recorded on the death record may not have been captured on the Census

NO CENSUS RECORD

A person may have had no 2016 Census record because they were not in scope of the Census due to absence from Australia, or died in the period around Census night, or they may have been missed in the 2016 Census.

Due to the size and complexity of the Census, it is inevitable that some people are missed and some are counted more than once. It is for this reason that the Census Post Enumeration Survey (PES) is run shortly after each Census, to provide an independent measure of Census coverage. The PES determines how many people should have been counted in the Census, how many were missed (undercount), and how many were counted more than once (overcount). It also provides information on the characteristics of those in the population who have been under- or overcounted.

The net undercount rate for the 2016 Census was 1%, with a higher rate for Aboriginal and Torres Strait Islander people (17.5%) than for the non-Indigenous population (6.6%). For more information please refer to Census of Population and Housing - Details of Overcount and Undercount, Australia 2016 (cat. no. 2940.0).

In a small number of cases, the absence of a Census form could be the result of the person being overseas at the time of the Census but subsequently dying in Australia and the death registered during the linkage reference period.

TIMING OF DEATH REGISTRATION

Due to an individual generally having reduced capacity to complete a Census form when near death, deaths occurring closest to Census night are more difficult to link. That is, it is more likely that deaths in the Census enumeration month do not possess a corresponding Census record, as the Census was not completed for the individual with sufficient information for linking (note that Census counts people in institutions such as hospitals, but may not collect all information required for linking). In 2011, 10% of deaths in August were unlinked, compared to a range of 6-8% for other months. Due perhaps to an extended enumeration period in 2016, the rate of unlinked records in the month closest to Census night doubled to nearly 20%.

Rolling enumeration procedures for the 2016 Census in remote Aboriginal and Torres Strait Islander communities may have increased the likelihood of an equivalent Census record not existing for deaths of members of these communities occurring around the time of the Census. Rolling enumeration involves conducting the Census over an extended period of four weeks. In these instances it is possible that a resident who moved during the enumeration period may have been missed and therefore a corresponding Census record would not exist, or they may have passed away after Census Night (9 August 2016) but before Census enumeration was conducted in their residential area.

TABLE 6 - DEATH REGISTRATIONS LINKED TO CENSUS RECORDS BY MONTH AND YEAR OF DEATH, Australia

Year of Death	Month of Death	Total Death Registrations no.	Linked Death Registrations no.	Linked Death Registrations %
2016
	August	4 220	3 250	77.0
	September	14 054	12 160	86.5
	October	12 775	11 508	90.1
	November	13 753	12 437	90.4
	December	11 398	10 285	90.2
2017
	January	12 703	11 522	90.7
	February	11 843	10 718	90.5
	March	13 454	12 240	91.0
	April	10 331	9 370	90.7
	May	14 766	13 407	90.8
	June	13 906	12 636	90.9
	July	13 043	11 879	91.1
	August	16 053	14 579	90.8
	September	15 081	13 666	90.6