Dwelling type (information about this variable and its quality)


Dwelling type

Dwelling type classifies all dwellings (ie occupied dwellings, unoccupied dwellings, and dwellings under construction) according to their structure and function.


A dwelling is any building or structure - or its parts - that is used, or intended to be used, for human habitation. It can be of a permanent or temporary nature and include structures such as houses, motels, hotels, prisons, motor homes, huts, and tents.

There can be more than one dwelling within a building. For example, each apartment in an apartment building is a dwelling.

Private dwelling

A private dwelling accommodates a person or a group of people and is not generally available for public use. The main purpose of a private dwelling is as a place of habitation; it is usually built (or converted) to function as a self-contained housing unit.

Private dwellings may be considered part of housing stock, or not part of housing stock.

Dwellings that are considered part of housing stock include:

  • houses, flats, units, townhouses and apartments (these may be stand alone or joined together). Generally, they will be fully self-contained but there may be exceptions, for instance where several flats share a toilet, laundry, or kitchen.
  • independent self-care units in retirement complexes
  • private dwellings within a non-private dwelling structure or complex
  • residences attached to a business or institution.

Dwellings that are not considered part of housing stock include:

  • dwellings in a motor camp. These include any caravan, campervan, house bus, cabin, unit, tent, or improvised dwelling in a motor camp that has permanent residents and is therefore not generally available for public use.
  • mobile dwellings. These include any mobile dwelling, on water or land, that is not in a motor camp, such as houseboats, campervans, mobile homes, house buses, house trucks, caravans, and tents. They are intended to be transportable and movable but may be fixed in one location.
  • improvised dwellings. These include dwellings or shelters not necessarily erected for human habitation, but which are occupied. The structure will support a roof of some kind, no matter how roughly fashioned or makeshift, and will lack some or all of the usual household amenities such as electric lighting, piped water, bathroom, toilet, and kitchen/cooking facilities. For example, shacks, garages, and private vehicles other than those designed as, or converted into, dwellings.
  • places of habitation with no dwelling. These include public or outdoor areas, not intended for human habitation but which are occupied: public parks, bus shelters, under bridges, on beaches, in caves, train stations, doorways, and private property such as car parks, and farm land are all included in the roofless or rough sleeper category.
  • vehicles lived in
  • vessels lived in

People may offer board or lodging to paying guests in their own homes (such as bed-and-breakfast, farm stay, home stay, or families hosting foreign students or boarders). Such homes are counted as private dwellings unless their main intent is to house boarders or paying guests.

Non-private dwelling

Non-private dwellings provide short or long-term communal or transitory type accommodation. They are generally available to the public for reasons of employment, study, special need, legal requirement, or recreation.

Non-private dwellings include:

  • backpackers, guest accommodation, hotels, motels, youth hostels
  • camps, communal staff quarters, hospitals, and institutional complexes
  • bed-and-breakfasts, farm stays, and home stays that are mainly intended to be used as facilities for paying guests.

Number of storeys

For separate dwellings, number of storeys is the number that the dwelling has.

For joined dwellings, number of storeys is the number in the entire building that the dwelling is part of.



Variable Details

Other Variable Information

Priority level

Priority level 2

We assign a priority level to all census variables: Priority 1, 2, or 3 (with 1 being highest and 3 being the lowest priority).

Dwelling type is a priority 2 variable. Priority 2 variables cover key subject populations that are important for policy development, evaluation, or monitoring. These variables are given second priority in terms of quality, time, and resources across all phases of a census.

The census priority level for dwelling type remains the same as 2013.

Quality Management Strategy has more information on the priority rating.

Overall quality rating for 2018 Census

Moderate quality

Data quality processes section below has more detail on the rating for this variable.

The External Data Quality Panel has provided an independent assessment of the quality of this variable and has rated it as poor quality. 2018 Census External Data Quality Panel: Assessment of Variables has more information.

Subject population

Occupied dwellings (private and non-private)

‘Subject population’ means the people, families, households, or dwellings to whom the variable applies.

How this data is classified

Census Dwelling Type Classification 2018 V1.0.0

Dwelling type is a hierarchical classification with three levels. Level one contains two categories:

  1. Private dwelling

  2. Non-private dwelling

    • Level two contains seven categories.
    • Level three contains 33 categories.

Census dwelling type data can be output at the lowest level of the classification, subject to meeting confidentiality requirements.

No residual categories (ie categories such as not stated and response unidentifiable) are used for dwelling type. All dwellings are classified as private or non-private during processing. If no further information is available about what type of private or non-private dwelling it is, then the dwelling is classified as ‘private dwelling not further defined’ or ‘non-private dwelling not further defined’, as appropriate.

Although there have been no conceptual changes to this variable, there have been minor changes to the classification of this variable since the 2013 Census. These are:

  • more detailed information on number of storeys for joined dwellings. The top category has been raised from ‘four or more’ to ‘ten or more’ and categories for ‘four to six storeys’ and ‘seven to nine storeys’ have been added. These changes were made to provide better information on apartments.
  • dwellings joined to businesses or shops are now classified as joined dwellings instead of being included in ‘occupied private dwelling not further defined’ as previously.

Each independent self-care unit, villa, or house within a retirement village is classified as a private dwelling and is included in the appropriate private dwelling category according to whether it is separate or joined and the number of storeys.

The homes of people identified during collection as living in a motor camp are classified as private dwellings in a motor camp. However, it is possible that some people living in a motor camp were not identified during collection and are included in the non-private motor camp/camping ground category.

Residential care for older people includes rest homes and complexes providing different levels of care such as rest home and hospital-level care.

Educational institution includes school hostels, seminaries, theological colleges, university halls of residence, and apartment-style student accommodation.

The night shelter category is solely for night shelters. Other forms of accommodation for people who are housing-deprived such as transitional housing and Salvation Army hostels are classified as welfare institutions.

Hotel, motel, guest accommodation includes some dwellings that provide long-term accommodation. Usual residents of these dwellings may include students.

A boarding house is defined as a dwelling that is mainly intended for boarders, has lockable bedrooms that are rented by the room, communal facilities, and can accommodate six or more boarders.

The Standards and Classifications page provides background information on classifications and standards.

Question format

Dwelling type data is derived from dwelling description, dwelling joined or separate and number of storeys on the online dwelling form (questions 2, 3, and 4 on the paper form) and information used in our operational phase which is compiled before the census. Stats NZ Store House has samples for both the individual and dwelling paper forms.

There were changes to the method of collection for this variable since the 2013 Census. The questionnaire design was changed to improve data quality. Information on non-private dwellings is no longer collected on the dwelling form but from information compiled in preparation for our operational phase.

There were differences between the modes of collection (paper and online form). There were no differences between the wording or question format in the online and paper versions of these questions but there were differences in the way a person could respond.

On the online dwelling form:

  • only one response could be selected for each question
  • the free text field for ‘other’ type of dwelling only appeared when a respondent selected ‘other’
  • if the respondent selected ‘mobile dwelling’ the number of storeys question did not appear.

On the paper dwelling form:

  • multiple responses to these questions were possible
  • it was possible to mark a tick-box and complete the free text field for ‘other’ type of dwelling.

These responses were resolved by edits.

How this data is used

Outside Stats NZ

  • To monitor trends and changes in housing.
  • To measure dwelling density patterns.
  • To identify substandard housing.
  • For helping to measure severe housing deprivation (homelessness).
  • For planning services to institutional type dwellings.

Within Stats NZ

  • As an administrative aid for census collectors and processing staff to help identify dwellings in the scope of census.

2018 data sources

We used alternative data sources for missing census responses and responses that could not be classified or did not provide the type of information asked for. Where possible, we used responses from the 2013 Census, administrative data from the Integrated Data Infrastructure (IDI), or imputation.

The table below shows the breakdown of the various data sources used for this variable.

2018 dwelling type – occupied dwellings (private and non-private)
Source Percent
Response from 2018 Census 91.7 percent
2013 Census data 2.4 percent
Administrative data 3.7 percent
Statistical imputation 2.2 percent
No information 0.0 percent
Total 100 percent
Due to rounding, individual figures may not always sum to the stated total(s)  

The ‘no information’ percentage is where we were not able to source dwelling type data for a dwelling in the subject population. In 2018 this was zero as dwelling type data was available either from 2013 Census data, administrative data or via statistical imputation for all remaining records for which a response had not been given.

Administrative data sources

Data from the following administrative source was used:

  • Tenancy Bonds, Ministry of Business, Innovation and Employment.

Please note that when examining dwelling type data for specific population groups within the subject population, the percentage that is from 2013 Census data, administrative data, and statistical imputation may differ from that for the overall subject population.

Missing and residual responses

This variable does not have residual categories that are grouped together as ‘not elsewhere included’ for output purposes. However, it does have categories for ‘private dwelling not further defined’ and ‘non-private dwelling not further defined’.

Responses that could not be classified beyond the private/non-private distinction (ie those in ‘private dwelling not further defined’ or ‘non-private dwelling not further defined’) remain in the data where we have been unable to find information from another source. In the data sources table, these responses are included in the ‘Response from 2018 Census’ percentage.

In 2018 no non-private dwellings were classified as not further defined and the percentage of private dwellings classified as not further defined was lower than in previous censuses due to the use of the additional data sources described above.

Private dwelling not further defined:

  • 2018: <0.1 percent
  • 2013: 5.8 percent.

Non-private dwelling not further defined:

  • 2018: 0.0 percent
  • 2013: 1.7 percent.

2013 Census data user guide provides more information about non-response in the 2013 Census.

Data quality processes

Overall quality rating: Moderate quality

Data was evaluated to assess whether it meets quality standards and is suitable for use.

Three quality metrics contributed to the overall quality rating:

  • data sources and coverage
  • consistency and coherence
  • data quality.

The lowest rated metric determines the overall quality rating.

Data quality assurance for 2018 Census provides more information on the quality rating scale.

Data sources and coverage: Very high quality

We have assessed the quality of all the data sources that contribute to the output for the variable. To calculate a data sources and coverage quality score for a variable, each data source is rated and multiplied by the proportion it contributes to the total output.

The rating for a valid census response is defined as 1.00. Ratings for other sources are the best estimates available of their quality relative to a census response. Each source that contributes to the output for that variable is then multiplied by the proportion it contributes to the total output. The total score then determines the metric rating according to the following range:

  • 98–100 = very high
  • 95–<98 = high
  • 90–<95 = moderate
  • 75–<90 = poor
  • <75 = very poor.

2013 Census data was highly comparable to 2018 Census responses, while admin data was mostly comparable, and data sourced through statistical imputation was moderately comparable to census forms. The high proportion of data from received forms in comparison to the low proportion sourced from 2013 Census, admin data and statistical imputation contributed to the score of 0.98, determining the very high quality rating.

Quality rating calculation table for the sources of dwelling type data –
2018 occupied dwellings (private and non-private)
Source Rating Percent of total Score contribution
2018 Census form 1.000 91.72 0.92
2013 Census 0.940 2.41 0.02
Admin data 0.800 3.69 0.03
Donor’s 2018 Census form 0.600 2.02 0.01
Donor’s response sourced from 2013 Census 0.564 0.07 0.00
Donor’s response sourced from admin data 0.480 0.09 0.00
No Information 0.00 0.00 0.00
Total 100.00 0.98
Due to rounding, individual figures may not always sum to the stated total(s) or score contributions.      

Data sources, editing, and imputation in the 2018 Census has more information on the Canadian census edit and imputation system (CANCEIS) that was used to derive donor responses.

Consistency and coherence: Moderate quality

Dwelling type in the 2018 Census is moderately comparable with the 2013 and 2006 Census data.

Variable data is not consistent with some expectations across one or more consistency checks. There are some differences in the data compared with expectations and benchmarks. Where these differences occur, this cannot be explained through likely real-world change or a change in how the variable has been collected.

Quality issues to note:

  • the proportion of separate houses and the proportion of joined dwellings did not change as expected
  • there is a large increase in data in ‘no storey information’ categories compared with previous censuses
  • the data shows an unexpected decrease in joined dwellings in buildings of four or more storeys since 2013
  • the number of private dwellings classified as ‘not further defined’ has decreased substantially. This is an improvement in data quality but may affect comparability over time.

Data quality: Moderate quality

The data quality checks for dwelling type included edits for consistency within the dataset and cross-tabulations to the regional council level

Dwelling type data has various data quality issues including:

  • an undercount of joined dwellings compared with separate dwellings which means the data has some bias toward separate dwellings. This is partly due to lower participation by those in joined dwellings and difficulties enumerating some of these dwellings, including dwellings in secure access buildings.
  • missing storeys information – a significant number of dwellings are in the ‘no storey information’ categories and the numbers in these categories are substantially higher than previously. This reflects lower overall response to census and non-response to the number of storeys question than expected response to the census overall, higher than desirable non-response to the number of storeys question. Alternative data sources did not provide information on storeys.
  • a potential undercount of boarding houses – although the 2018 Census data is likely to provide better representation of these dwellings than previous census data, the 2018 figure may still be an undercount. These dwellings can be difficult to identify.
  • other private dwellings such as dwellings in a motor camp, mobile dwellings not in a motor camp, improvised dwellings or shelter, roofless or rough sleeper, have been undercounted so the 2018 Census data does not fully represent the use of housing of these types or the number of people living in these housing situations.

Recommendations for use and further information

While new imputation methods, administrative data, and 2013 Census data have been used to produce the 2018 Census data, the overall quality of the data is moderate.

When using this data you should be aware that:

  • data has been checked to regional council level. Some variation is possible at geographies below this level.
  • at regional council level, the amount of data from other sources varies due to differences in non-response rates.
  • the significant decrease in data in the ‘private dwelling not further defined’ category may affect comparability over time depending on the particular analysis being done.
  • the patterns and trends this data shows for separate versus joined private dwellings may not always fully represent real-world changes due to some bias toward separate dwellings.
  • the significant number of dwellings in the ‘no storey information’ categories means that the data that is available on number of storeys may not always fully represent real-world trends. Caution should be used when interpreting trends over time.
  • the changes in the counts for residential care for older people and for residential and community care facilities are likely to be due to real-world change and improvements in identifying these dwelling types.
  • the increased number of boarding houses for 2018 is likely to be due to better identification of these dwellings.

As well as providing information on the numbers of different types of dwellings, this data can also be used to provide information on the numbers and characteristics of people living in different types of dwellings.

Although the standard subject population for this variable is occupied dwellings, dwelling type information for unoccupied dwellings and dwellings under construction is also available. Caution is needed when using dwelling type data for unoccupied dwellings and dwellings under construction due to the high level of imputation of dwelling type for these dwellings.

Contact our Information Centre for further information about using this variable.

Revision Information

Currently viewing revision 12 by on 11/03/2020 4:10:14 a.m.

Revision 12 *
11/03/2020 11:39:04 p.m.
Revision 11
19/02/2020 2:59:11 a.m.
Revision 10
4/11/2019 5:34:59 a.m.
Revision 9
3/10/2019 2:16:37 a.m.
Revision 6
25/09/2019 4:39:47 a.m.
Revision 5
22/09/2019 9:53:26 p.m.

Show / Hide more...


DDI Agency
DDI Version


DDI 3 Download

Select the languages to display