|Page tools: Print Page Print All RSS Search this Product|
CHAPTER 22. MULTI-PURPOSE HOUSEHOLD SURVEY
22.0.2 This chapter describes the broad survey methodology of the MPHS. For information on the three labour related MPHS topics that have been conducted to date, see Chapter 22.1 (Barriers and Incentives to Labour Force Participation), Chapter 22.2 (Retirement and Retirement Intentions) and Chapter 21.15 (Work-Related Injuries).
OBJECTIVES OF THE MULTI PURPOSE HOUSEHOLD SURVEY
22.0.3 The MPHS topics are an important part of the ABS's household surveys program, which aims to:
22.0.4 The information requirements of MPHS topics are determined on the basis of submissions from users on their needs for, and uses of, household survey data. They also reflect ABS deliberations on what is required of a national statistics program in the various subject fields, based on user contact and consultation.
22.0.5 The MPHS is a flexible multi-topic survey vehicle which is used to collect and output data in a timely fashion. The MPHS includes a number of topics that require personal interview (rather than using the any responsible adult (ARA) method) and are therefore unsuited to the monthly supplementary program. The MPHS has a shorter development and output time than the special social surveys to achieve flexibility in responding to user demands as they arise and to allow the ABS to respond to emerging demand and contemporary priorities in a timely way (publications are usually available within six months of the completion of data collection).
22.0.6 In addition to the labour-related topics outlined in paragraph 22.0.1, the MPHS program also includes other social topics not related to labour statistics, such as household use of information technology, patient experience, family transitions and people's attitudes to the environment.
22.0.7 Estimates from each MPHS topic are released in separate publications. Expanded Confidentialised Unit Record Files (CURFs) are made available from each MPHS via the Remote Access Data Laboratory (RADL). Information on CURFs and the RADL can be found on the ABS web site, under Services – Microdata - CURFs and TableBuilder.
22.0.8 As with the Labour Force Survey (LFS) the scope of the MPHS is restricted to people aged 15 years and over and excludes:
22.0.9 In addition, the following people are excluded from most MPHS topics (see paragraphs 18.2 to 18.4 of Chapter 18 for further information):
22.0.10 Depending on the topic, there may be further exclusions from scope. For example, some MPHS topics collect information only from a certain population or interest group (e.g. information on retirement is collected from people aged 45 years or over).
22.0.11 The MPHS is conducted in conjunction with the monthly Labour Force Survey (LFS). Approximately one third of the dwellings in the LFS out-going rotation group (one eighth of the sample is rotated out each month) are selected for enumeration in the MPHS. For 2008–09, half of the dwellings in the LFS out-going rotation group were selected for enumeration to compensate for the decreased sample in the LFS (see Chapter 20). From each dwelling selected for the MPHS, one person is randomly selected for interview.
22.0.12 Data are collected using Computer Assisted Interviewing (CAI), whereby responses are recorded directly onto an electronic questionnaire in a notebook during a telephone or face-to-face interview.
22.0.13 Data for MPHS topics are collected each month over a financial year. This reduces the impact of any seasonal effects on the data. For the first MPHS, conducted in respect of 2004–05, interviewing commenced in August 2004 rather than July. For the 2006–07 MPHS, no interviews were conducted in August or September 2006.
22.0.14 The MPHS uses the same sample design as the LFS. The sample used in the MPHS is a subset of the LFS sample. Each month a sample of approximately 1,250 dwellings are selected.
22.0.15 The size of the annual sample selected is approximately 15,000, but the response rates for the MPHS are slightly lower than for the LFS (averaging around 88%). The number of fully responding households is therefore usually around 13,000. The size of the MPHS sample can be increased with user funding.
22.0.16 The estimation technique used for household surveys is the generalised regression estimator.
22.0.17 The survey is benchmarked to the estimated civilian population aged 15 years and over living in private dwellings in each state and territory in non-sparsely settled areas. The process of weighting ensures that the survey estimates conform to person benchmarks by state, part of state, sex, age and household composition.
RELIABILITY OF THE ESTIMATES
22.0.18 Estimates from the MPHS are subject to both sampling and non-sampling error (see Chapter 17 for more detail). The relative standard errors (RSEs) for each of the survey estimates are published in shadow tables. For further information on calculating RSEs from the MPHS, see the technical notes of each publication.
22.0.19 A 'group jack-knife' variance estimator is used to calculate estimates of variance (see Chapter 17 for more detail).
22.0.20 For further details contact the Labour Market Statistics Section in Canberra on (02) 6252 7206 or at <email@example.com.>.