COMPONENTS OF NET UNDERCOUNT

NOTE: A coding error was identified in the 2011 PES, relating to how records were categorised into the contact and non-contact sectors. Table 13 was impacted by this error. See the I-Note on the Explanatory Notes tab for further details.

COMPONENTS OF NET UNDERCOUNT

While net undercount estimates measure Census coverage, and adjustment factors are a means of adjusting Census counts for coverage, it is also important to understand the contributing components of net undercount.

The components of net undercount for a category are:

• undercount (the number of people in that category who were missed in the Census);
• overcount (the number of people in that category counted in the Census when they should not have been);
• differences in classification between the PES and Census (for example, where age, sex or Indigenous status information does not match); and
• imputation error (either under or over-imputation) in the Census.

This section provides additional information on these components to assist in the interpretation of estimates of net undercount presented in this publication.


Contact and non-contact sector

Table 13 provides estimates of the key components of net undercount, the definitions of which follow the table. For the purposes of PES estimation, Census dwellings deemed occupied on Census night are divided into a 'contact' sector (dwellings for which a Census form was received before the commencement of PES enumeration) and a 'non-contact' sector (where no Census form was received or a Census form was received after the commencement of PES enumeration - that is, a 'late return').

13 COMPONENTS OF NET UNDERCOUNT(a)(b), Contact and non-contact sector - 2011
		UNDERCOUNT IN THE CONTACT SECTOR
		Persons missed in the Census	Net difference in classification	Persons with Census category not-stated	Total undercount in the contact sector		Undercount in non-contact sector		Net undercount
		no.	no.	no.	no.	SE	no.	SE	no.	SE
Australia		634 579	-	-	634 579	25 380	-260 039	27 611	374 540	38 315
State/territory of usual residence
	New South Wales	217 639	4 651	-	222 290	18 507	-85 644	18 012	136 647	26 865
	Victoria	120 917	-5 098	-	115 819	11 283	-58 912	13 630	56 906	17 864
	Queensland	139 820	-3 269	-	136 550	11 612	-59 336	12 466	77 215	17 431
	South Australia	32 446	-803	-	31 644	4 727	-14 361	4 344	17 283	6 045
	Western Australia	87 034	2 211	-	89 246	7 760	-31 327	8 800	57 918	12 004
	Tasmania	13 839	1 041	-	14 880	2 369	-4 619	1 709	10 261	3 076
	Northern Territory	16 609	1 952	-	18 561	2 642	-2 845	1 402	15 716	3 074
	Australian Capital Territory	6 274	-685	-	5 589	2 209	-2 994	2 027	2 595	3 055
Sex
	Male	393 555	1 664	-	395 219	17 603	-154 423	16 281	240 796	24 534
	Female	241 023	-1 664	-	239 360	14 827	-105 615	14 508	133 744	20 772
Age groups (years)
	0-4	49 140	-22 080	-	27 060	6 344	-9 797	4 772	17 263	8 017
	5-9	29 582	-586	-	28 996	6 680	-9 069	4 118	19 926	7 781
	10-14	19 312	729	-	20 042	6 511	-14 127	3 761	5 914	7 664
	15-19	46 229	-2 473	-	43 756	7 534	-7 861	4 865	35 895	9 044
	20-24	108 284	7 686	-	115 970	10 635	-7 821	6 638	108 148	12 812
	25-29	95 330	-188	-	95 142	10 487	-2 505	7 812	92 636	13 444
	30-34	59 418	3 548	-	62 966	8 415	-18 019	5 450	44 946	10 407
	35-39	49 768	-15 574	-	34 194	7 650	-16 987	4 897	17 207	9 424
	40-44	37 122	5 794	-	42 915	7 644	-21 892	4 100	21 023	8 856
	45-49	37 607	-6 756	-	30 851	7 013	-25 360	3 687	5 491	8 063
	50-54	30 646	7 866	-	38 513	7 131	-24 765	4 242	13 747	8 411
	55+	72 142	22 032	-	94 174	9 610	-101 834	7 172	-7 659	11 500
Registered marital status
	Never married(c)	425 154	34 811	-	459 966	22 591	-74 371	18 606	385 595	29 990
	Widowed, divorced or separated	74 578	-66 701	-	7 877	12 757	-32 148	6 714	-24 271	14 630
	Married	134 847	31 889	-	166 736	15 225	-153 520	10 761	13 216	18 737
Indigenous status(d)
	Indigenous	56 650	5 128	7 009	68 787	11 982	45 402	5 969	114 188	14 274
	Non-Indigenous	577 928	-5 128	280 012	852 812	25 388	465 987	26 476	1 318 799	37 272
Country of birth(d)
	Australia	405 053	14 448	307 088	726 589	20 883	335 809	20 578	1 062 398	30 156
	England	21 653	-15 325	22 106	28 434	6 270	15 761	3 082	44 195	7 049
	New Zealand	23 881	-3 398	8 368	28 851	4 894	17 685	3 766	46 536	6 344
	China	33 905	-4 261	6 002	35 646	6 782	20 319	5 062	55 965	8 934
	India	12 117	-2 246	5 336	15 207	3 829	16 487	4 811	31 694	6 075
	Italy	2 904	-2 248	6 775	7 431	2 169	1 453	1 067	8 884	2 513
	Vietnam	9 389	-1 765	4 458	12 082	4 075	4 013	1 984	16 095	4 313
	Philippines	6 090	3 171	3 318	12 579	3 053	4 527	1 884	17 106	3 603
	South Africa	3 496	-895	2 289	4 890	2 013	6 822	2 755	11 712	3 383
	Scotland	1 268	-7 096	3 506	-2 322	1 733	3 904	1 374	1 582	2 208
	Other overseas	114 821	19 616	53 599	188 036	12 411	85 768	9 553	273 804	15 894
- nil or rounded to zero (including null cells)
(a) Estimates are based on a sample survey, and hence are subject to sampling error indicated by the SEs provided. For further information see the Sampling Error Technical Note (in Explanatory Notes).
(b) A negative value indicates a net overcount.
(c) Includes those who are living with a de facto partner and have never been in a registered marriage.
(d) Net undercount is based on Census counts for a category. In the Census, Indigenous status and Country of birth were set to not-stated where the response was blank and where imputed person records were created for non-responding dwellings. Hence components of undercount for Indigenous status and Country of birth do not sum to the Australia total.

Persons missed in the Census in the contact sector

For a given category of person as reported in the PES (e.g. 'Males' recorded in the PES), 'Persons not counted in the contact sector' is the difference between the number of people in that category that should have been counted in the Census and the number of people in that (PES) category that were counted on Census forms (irrespective of their Census category).

Net difference in classification in the contact sector

Occasionally, the responses obtained for a person in the PES interview are not consistent with those obtained for the same questions in the Census. In addition, where a value has been imputed in the Census for a missing response (for example, Age, Sex, or State/territory of usual residence), it may differ from the PES response.

The net difference in classification for persons actually counted in the Census contact sector is equal to the estimated number of people in the given category as reported in PES, minus the number counted in that same category in the Census. For more information see the Differences in classification Technical Note (in Explanatory Notes).

Persons (in the contact sector) with Census category not-stated

The Census contact sector contains some dwellings which were responding in the Census but returned only a partially completed Census form. For Census purposes, values for Age, Sex, State/territory of usual residence and Marital status are imputed during Census processing in cases where these items have been left blank. Missing values for other items remain 'not-stated' in the final version of Census counts (including Indigenous status and Country of birth). Persons in the contact sector with a Census category value of 'not-stated' contribute to net undercount estimates for the category in which they should have been counted as reported in the PES (e.g. Indigenous status).

Undercount in the non-contact sector

Non-responding dwellings in the Census are dwellings where the Census never obtained a return, and the dwelling could not be established as having been unoccupied on Census night. During Census processing, a 'hot-deck' imputation method (see Glossary) was utilised to impute people and their Age, Sex, Marital status and Place of usual residence into these dwellings. These values were, in many cases, based on information provided by the Census collector about the dwelling and its residents. Values for all other variables (e.g. Indigenous status) were set to 'not-stated' or 'not applicable', depending on the imputed value for Age.

Inevitably, the imputed values differ from the true but unknown values.

Imputed records made up the majority of the 2011 Census non-contact sector. Late returns made up only a small proportion of this sector.

An estimate of the undercount in the non-contact sector for a category of person is obtained by calculating the difference between the PES estimate of the number of people who should have been counted in the non-contact sector in that category (as reported in the PES) and the Census count of people in the non-contact sector for the category (including imputed person records within the category). For a category of persons classified by Age, Sex, Marital status and State/territory of usual residence, this component of net undercount largely represents an estimate of imputation error in the Census for this category of imputed persons. For the Indigenous status and Country of birth categories, this component largely represents an estimate of persons who should have been counted in this category in non-responding dwellings.