Nine things we wouldn’t know without a Census
Last week the ABS confirmed that a decision regarding the fate of the 2016 Census and the future of the five yearly Census cycle was still on hold. One of the reasons the ABS wants to move the Census to a ten year time frame is that they claim that most of the data can be obtained through other sources, such as administrative data, and that this can be released on a more frequent basis. But how true is this? The Australian Statistician, David Kalisch, has been quoted as saying that the Census isn’t as useful as many people think, because most of the data can be obtained elsewhere. Really? He used births data as an example – and it’s true that the ABS sources this data from the State Registries of Births, Deaths and Marriages – but it’s not telling the full story. I believe one of the great benefits of Census data is what it tells us about communities and how they’re changing. Glenn and Lailani have already blogged about the impact of the loss of small area data, but I think it’s also worthwhile considering what we wouldn’t know if the Census wasn’t held. Here’s my not exhaustive list –
1. How migrant communities are ageing
It’s a common belief that Greek and Italians are the largest migrant communities in Australia, but it does depend on how you measure them – birthplace, language or ancestry. As a birthplace, their numbers are declining as their mortality catches up with them. In fact, Greece is no longer one of the top ten birthplaces in Australia (Italy is still number five). But as ancestries, the community is larger as many descendants may self identify on the Census form as being of Italian or Greek origin, and in fact the numbers are increasing.
2. The types of households we live in
Most administrative data sources collect data on individuals rather than households. Data on types of households eg families, couples and lone persons is critical in determining drivers of population change in a community. The way households change over time is critical to understanding how and why populations change, as well as the demand for new dwellings.
3. The shift to larger dwellings
There is very limited data on the size of dwellings in Australia – new dwellings yes – but total dwellings no. A useful proxy is the number of bedrooms. Did you know that over the last ten years there has been a decline in the number of houses with two bedrooms, but an increase in the number with three and four? This is particularly evident in established parts of our cities. Presumably what were two bedroom houses are being renovated or knocked down to become three and four bedroom houses. It’s a great but little known example of how our urban areas are changing and how the population is changing the way it consumes housing.
4. We aren’t as religious as we once were
The Census is the only source of religion data in Australia. Though the question is optional, more than 90% of people still answered it in 2011. The data provides great examples of how we are changing as a society. So-called shock jocks may claim Australia is being overrun by Muslims, but did you know there are more Buddhists? Furthermore, almost a quarter of the Australian population now claim no religion and the number increased by more than 1 million between 2006 and 2011. Surely that tells us more about Australia in the 21st century.
5. People don’t necessarily live in private dwellings
Non-private dwellings (NPDs) are a communal type of living arrangement, and are measured differently to private dwellings. Examples include aged care facilities and university campus accommodation. They often have a strong relationship with age and sex eg there are more older women living in nursing homes.
6. The diversity of our Indigenous population
Even the ABS website says that the Census is the only comprehensive source of data on Indigenous population. On many indicators Indigenous Australians display higher levels of social disadvantage. There are marked differences in the distribution of the Indigenous population across Australia which impact of service delivery at the State and Local Government level. Policies such as Closing the Gap are also informed by the Census results and the progress on the various measures would suffer under a ten year cycle.
7. The changing nature of the Australian economy
OK so the ABS runs a lot of economic surveys, but they are really only useful at macro levels of geography. Progressive cut backs to major survey vehicles such as the Labour Force Survey mean that the standard errors for detailed data are increasing in frequency. It’s also a bit of a worry when even the Australian Statistician queries the reliability of the data….a great example of why the five yearly benchmarking process is necessary to help validate the surveys!
8. It’s squishy on the tram today
State governments often have transport policies which aim to increase the proportion of people travelling to work by public transport, but clearly this needs to be placed in the context of the total amount of travel to work, not just how many people “touch on” with their Myki or similar. But there are also clear spatial patterns associated with the mode of transport to work – funnily enough if you live within walking distance of the tram or train you are more likely to use it to get to work.
9. The “myth busting” nature of the data
There is such a wealth of data released through the Census that if you have access to the right tools, you can use this as the evidence base in your work, or even to answer those difficult questions at a trivia night. You only have to browse through our blog to see the vast range of topics that we’ve written about using Census data. They tell fantastic stories about the characteristics of our communities and how they’re changing.
The use of any data source needs to be balanced against it’s quality, timeliness and the purpose for which it is being used. I don’t doubt that administrative data has potential, but on the whole it is not collected with a view to output. An investment needs to be made in making the data “clean” enough so that it can be used as an evidence base. In dollar terms, is this really a cost saving compared to the value of the Census output? The ABS already does use administrative data for the purpose of statistical output and the aforementioned births data is a good example. But if you think about it, births data is not a substitute for the number of 0 year olds in the population, because you still need to account for overseas migration and infant mortality.
We hear from our clients ALL THE TIME about how the Census data informs their work. From housing strategies to economic development policy, the Census data provides this important evidence base. We also hear from our clients about how they need reliable updates because things change (or don’t change – just as important information) even within a five year period. Moving to a ten year Census time frame would make analysis of local, state and national population characteristics extremely difficult.