The Australian Bureau of Statistics (ABS) is the country’s official statistical agency. They are responsible for running the five-yearly Census, but also for producing a range of other national data, including the monthly labour force. But, what do these figures mean and where do they come from? The focus of this piece is on the source of the data, the ABS Labour Force Survey. Future statistics FlagPosts will aim to explain some of the key numbers.
Where do the figures come from?
Each month, the ABS approaches a selection of Australians to ask them a set of questions from the Labour Force Survey. This process results in about 50,000 people answering the questions, about themselves, and on behalf of people they live with (where relevant to the survey). The responding group of people are known as the ‘sample’. They are a sub-set of the population used to provide information about the whole population. These same people are asked survey questions each month, for eight months, until their participation is complete.
How are people selected for the survey?
The ABS randomly selects people living across (almost all of) Australia to complete the survey. People are predominantly selected from private households. To contact sufficient people for the survey, the ABS selects a number of households within specific geographical areas of each state or territory. Households are then approached to participate in the survey. A small number of people, from discrete Aboriginal and Torres Strait Islander communities and non-private dwellings (e.g. motels, hospitals), are also selected to participate.
Who is included in the survey?
The survey includes usual Australian residents, aged 15 years or older, who form part of the civilian population. The civilian population differs from the total population, as it excludes some people, such as permanent defence force members, some diplomatic personnel, and overseas residents (intending to stay for fewer than 12 months).
What types of questions are asked?
A standard set of introductory questions is asked to determine whether anyone should be interviewed for the Labour Force Survey. Once eligibility is confirmed, a series of labour force questions are asked. Sometimes information may be unknown, sometimes it may not be provided. Overall however, the survey experiences a very high rate of response (above 92%).
The questions are broadly the same as those asked since the survey began in the 1960s, but with some changes to the way in which the information is collected, the sample size and processing techniques. As an example, the questionnaire asks: ‘I would like to ask about last week, that is, the week starting Sunday (date) and ending Saturday (date). Did you do any work at all in a job, business or farm?’ People are also asked about unpaid work in a family business, temporary absences from work, whether they were looking for work (in the last four weeks), whether they were changing jobs or waiting to start a new job, and the characteristics of their job. When not in work, people are asked about their job seeking activities or out of work experiences.
A copy of the survey questionnaire is available from the ABS, Information paper: Questionnaires used in the Labour Force Survey, July 2014 (cat. no. 6232.0). Copies of two earlier questionnaires are included in the ‘Past & future releases’ tab. A summary of changes to the survey is available from ‘Labour force comparability over time’ in Labour statistics: concepts, sources and methods, Feb 2018 (cat. no. 6102.0.55.001).
What happens after the information is collected?
Once the survey has been conducted, the responses form a data set, which is the basis for producing a range of labour statistics. However, before the data is released, it undergoes a series of checks and adjustments. The data set is also subject to upsizing, so that the resulting figures, or ‘estimates’, are representative of the population. For example, in Tasmania, one person in the survey represents about 99 people. This means that the characteristics of that person are used to infer the characteristics of a larger group of people. This process is repeated across the data set (to varying degrees) and then checked for consistency. To do this 'upsizing', the ABS uses the Estimated Resident Population (ERP). The ERP provides a snapshot of the Australian population at a particular point in time, which takes into account changes since the Census due to people arriving in or leaving the country, being born or dying, or moving interstate. Comparisons between the data set and the ERP are made and variations identified where possible. Note there may still be unidentified differences, which should be considered when using the data.
The resulting figures are the ABS’ best approximation of what the population’s labour force looks like, based on the information collected and the estimation techniques used.
What are the key figures from the survey?
The key figures are the unemployment rate, participation rate, employment-to-population ratio (also described as the ‘employment rate’), the unemployed (jobless), the employed (workers) and the labour force (active population. Relevant concepts and terminology will be explored in subsequent posts.
For more information
Clients of the Parliamentary Library can request assistance to interpret ABS statistics, navigate the ABS website or find relevant ABS surveys and products by contacting the Statistics and Mapping Section via Library Enquiries.