Page tools: Print Page Print All | |||||||||||||||||
|
This document was added or updated on 26/05/2020. METHODS USED IN ABS HOUSEHOLD SURVEYS Table 17.1: Household surveys and collection methods
Intensive follow up procedures for non-response are in place for household surveys. Interviewers make a number of attempts to contact households at different times of the day and on different days during the week. For households unable to be contacted by telephone, a face-to-face visit is attempted. If the household can still not be contacted within the survey period after repeated attempts, and the dwelling has been verified as not vacant, the household is listed as a non-contact. SAMPLE DESIGN With the exception of the Census of Population and Housing, most ABS household surveys use probability sample designs, drawing their sample from the Monthly Population Survey (MPS) and the SSS samples, which are drawn from a ‘Master Sample’. These household surveys all use a multi-stage, stratified sample design. Typically three stages are used; the first stage units (FSUs) are randomly selected areas the size of Statistical Area Level 1’s (SA1s) - about 200 dwellings. The Master Sample consists of these FSUs. The Master Sample is drawn from the Population Survey Framework, which is composed of three components: the private dwelling framework, the special dwelling framework, and the Aboriginal and Torres Strait Islander Communities framework. These three frames are generally non-overlapping, and therefore enable the selection of samples that represent the Australian population. The overlap occurs as there are some special dwellings within the Aboriginal and Torres Strait Islander Communities framework. For more information about sample design and method of estimation, see articles 'Sample Design' and 'Method of estimation' in Information Paper: Labour Force Survey Sample Design (cat. no. 6269.0). Private dwelling framework In general, private dwellings are structures built specifically for living purposes, such as houses, flats, home units, and any other structures used as private places of residence. A private dwelling can also be a caravan, a houseboat, a house attached to an office, or rooms above a shop. In practice, some dwellings such as caravan parks and marinas are listed on the special dwelling list. In most areas of Australia, private dwelling sample selection is structured around the selection of fine geographic regions defined by the aggregation of mesh blocks. Mesh blocks are the finest unit in the Australian Statistical Geography Standard (ASGS), the ABS Geography Standard which replaced the previous standard in 2012. For more information about mesh blocks, see Australian Statistical Geography Standard (ASGS): Volume 1 - Main Structure and Greater Capital City Statistical Areas (cat. no. 1270.0.55.001). The key geographic sampling unit in the new framework is called the Base Frame Unit. These Base Frame Units were created by combining contiguous mesh blocks in nearly all regions of Australia, and were created solely for the purpose of household survey sampling. Their intended role is to define the geographic area within which dwellings are organised into groups which are selected in a sample together. These selected dwellings within the selected Base Frame Units are termed the “cluster”. The clusters vary in size from 5-15, reflecting the cost of enumeration. If an area is remote and costly to enumerate, it will have a cluster size at the upper-end of this range of cluster sizes. Three special strata are adopted: Secure Apartment Buildings, Pre-Determined Growth, and Indigenous geography strata. There is a single special stratum of each type within a State/Territory (at most), so the sample in these strata can cut across the area unit boundaries Each area selection unit in the master sample is assigned an "area type" class based on the geography of Australia. A variety of geographic classifications defined by different sources are combined to derive the area type classes:
Special dwelling framework The special dwelling household framework is a list of 'special' dwellings, from which samples of special dwellings and their residents can be selected. Special dwellings are establishments which provide predominantly short-term accommodation for communal or group living, and often provide common eating facilities. They include hotels, motels, hostels, hospitals, religious institutions providing accommodation, educational institutions providing accommodation, prisons, boarding houses, short-stay caravan parks, and may include some Aboriginal and Torres Strait Islander communities that are not on the Aboriginal and Torres Strait Islander Community Frame. Some special dwellings are designed for a particular purpose (e.g. hospitals) and, as such, provide accommodation for specific groups of persons. Special dwellings each comprise a number of dwelling units. Currently, there are around 26,000 special dwellings on the frame. The framework contains information about the occupancy of each special dwelling as it was on Census night. The special dwelling framework is also stratified geographically, though at a broader level than the private dwelling framework. In many cases the demographic, social and labour force characteristics of the occupants of special dwellings are not typical of the population residing in private dwellings, and therefore it is necessary to sample special dwellings separately by placing them in separate strata within each geographic (sample design) region. This provides for more effective samples of persons within special dwellings and private dwellings, and the flexibility to select some samples which exclude all or some special dwellings, or to select samples in which special attention is paid to persons residing in particular special dwellings. Aboriginal and Torres Strait Islander Community Frame The Aboriginal and Torres Strait Islander Community frame is a tool used to ensure adequate sample selection for this population. It can be thought of as an extension of the private dwelling frame. A Mesh Block is classified as a discrete community mesh block if it is deemed to have an Aboriginal or Torres Strait Islander community population of 75% or more, and lies in the non-metropolitan area of Queensland, South Australia and Western Australia or Northern Territory. This frame is constructed using information from the Census of Population and Housing and other information covering the communities. There are two sample groups included on this frame. Discrete Aboriginal and Torres Strait Islander communities (including any out-stations associated with them) are referred to as the 'community sample'. Dwellings in areas not covered by the community sample are referred to as 'non-community sample'. Information on the Aboriginal and Torres Strait Islander Community frame, community and non-community sample is contained in the ABS publication National Aboriginal and Torres Strait Islander Social Survey: User Guide, 2014–15 (cat. no. 4720.0). The Aboriginal and Torres Strait Islander Communities frame is stratified geographically by State/Territory, with Torres Strait Islander communities in Queensland separately stratified. MPS AND SSS MASTER SAMPLES From July 2018, there will be a single Master Sample covering the sample requirements for both the Monthly Population Survey (MPS) and the Special Social Surveys (SSS)’s. The 2018 Master Sample will be the first to make use of the Address Register, which is now also used to support the enumeration of the Census of Population and Housing. In addition, a new method of selection (known as Conditional Selection) will also operate from 2018 onward, which will support more flexible sampling methods. Conditional selection is a method of selecting survey samples that allows the ABS to effectively manage overlap between different surveys, to prevent any household from being selected for two or more surveys, while also allowing survey samples to be located nearby to each other in order to reduce survey costs. The MPS sample and the SSS samples comprise Base Frame Units taken from the private dwelling framework, special dwellings, and Indigenous communities (IC) from the ICF. Most household surveys conducted by the ABS use samples drawn from the Master Sample. The MPS consists of monthly LFS, the Multipurpose Household Survey (MPHS), and also various supplementary surveys conducted in conjunction with the LFS. Dwellings selected in the LFS sample remain in sample for eight consecutive months. The program of SSSs consists of large-scale periodic surveys covering a wide variety of topics. Most SSSs have similar (though slightly smaller) survey scope to the MPS, so the requirements and structure of the samples are also similar. In terms of the geographic scope of MPS and SSSs, a key difference is that most SSSs exclude very remote areas. Most SSSs do not obtain sample from discrete Indigenous communities, or select persons in special dwellings. To date, the SSS Base Frame Units do not include any Base Frame Units selected in the MPS sample, thereby preventing households selected for the MPS from also being selected for a SSS during the life of a specific sample design. It has traditionally been the practice that the Master Sample is re-selected and redesigned every five years following the Census of Population and Housing. The move from Census-based master samples to Address Register-based designs enables more frequent updates, with the first Address Register-based sample expected to be in use for 3 years, from July 2018 to June 2021. Sample selection From 2018, the ABS is using an Address Register in the sample selection process for all of its household surveys. The Address Register, which is also now used to support the enumeration of the Census of Population and Housing, is a list of all physical addresses (both residential and non-residential) in Australia. The main input to the register is the Geocoded National Address File (G-NAF), with continuing supplementation from other available address sources and from field work undertaken by ABS officers. The ABS has developed this register as the central source of addresses used in the collection of information in response to the need for more efficient and effective household survey designs, including:
The Address Register Common Frame is a trusted and comprehensive data set of Australian address information. It contains current address text details, coordinate reference (or “geocode”), and address use information for addresses in Australia. Stages of selection There are three stages of selection:
The Mesh Block is the finest ASGS 2016 geographical unit, typically containing 30-60 dwellings. First Stage Units are typically a set of contiguous Mesh Blocks. These stages of selection within a stratum are illustrated in Figure 17.1 below. Figure 17.1: Three stages of selection FSU – First Stage Unit MB – Mesh Blocks The degree of clustering in the selections is controlled by defining the number of dwellings that can be selected within each First Stage Unit in any month of the MPS or within a cycle of a SSS. This number is referred to as the cluster size. The cluster size should ideally balance cost and accuracy considerations, within each stratum of the sample. Dwellings in a cluster will be ideally within a Base Frame Unit, and not dispersed beyond this. ESTIMATION METHODS Household survey estimates are generally calculated using calibration estimation techniques. BENCHMARKS Changes to the LFS population benchmarks impact primarily on the magnitude of the LFS estimates (i.e. employment and unemployment) that are directly related to the underlying size of the population. Estimates of the population produced from household surveys are calculated in such a way as to add up to independently estimated counts (benchmarks) of the population. For the LFS, these benchmarks are based on Census of Population and Housing data, adjusted for under-enumeration and updated for births, deaths, interstate migration, and net permanent and long term migration. Benchmarks have been developed for state/territory of usual residence, part of state of usual residence (for example, capital city, rest of state), age and sex. Each cross-classification of these benchmark variables is known as a benchmark cell. Revisions are made to benchmarks after each Census of Population and Housing, and when the bases for estimating the population are reviewed. Other household surveys use various combinations of benchmark variables to produce benchmark cells. Some surveys use supplementary information (such as LFS estimates), referred to in this context as pseudo-benchmarks, to supplement independent demographic benchmarks based on Census of Population and Housing data. Household surveys may use calibration methods to incorporate other auxiliary information on target populations into estimates - for instance, benchmarks for the Indigenous population or the population of private households. NON-RESPONSE Non-response arises when no information is collected from one or more occupants of a selected dwelling. Interviewers make a number of attempts to contact households at different times of the day and on different days during the week. For households and persons unable to be contacted by telephone, face-to-face visits are attempted. If the household still cannot be contacted within the survey period after repeated attempts (and if the dwelling has been verified as not vacant), it is listed as a non-contact. Non-contact is the most common form of non-response. The response rate commonly quoted for ABS household surveys refers to the number of fully responding dwellings expressed as a percentage of the total number of selected dwellings excluding sample loss. Examples of sample loss for the LFS include:
For most household surveys, a non-response adjustment is performed implicitly by the estimation system, which effectively imputes for each non-responding person on the basis of all responding persons in the same post-stratum. This adjustment accounts for both full non-response and non-response for individual questions. RESPONSE RATES The response rate usually quoted for ABS household surveys is defined as the number of fully responding households, divided by the total number of selected households excluding sample loss. Examples of sample loss for household surveys include: households where all persons are out of scope and/or coverage; vacant dwellings; dwellings under construction; dwellings converted to non-dwellings; derelict dwellings; and demolished dwellings. Averaged over the three years from January 2015 to January 2018, the LFS response rate was 93%, which is high by international standards. FOOTNOTES
BACK TO TOP Document Selection These documents will be presented in a new window.
|