The Australian Bureau of Statistics (ABS) will reveal 2016 census data today (27thTuesday ).
But this census was dogged by an online debacle that embarrassed the government, inconvenienced millions of Australians .
Response rates were still high in this census
The census website was offline for more than 40 hours from census night but after it returned the ABS reminded the public they had until September 23 to complete the forms.
Around 58 per cent, or 4.9 million households, submitted their forms online and another 3.5 million filled out paper forms.
Alastair McGibbon, the Prime Minister's special adviser on cyber security, noted in his report on the census issues that a response rate of 96.5 per cent would mean the census was "fit-for-purpose".
As of October, the response rate was more than 96 per cent and on Tuesday, the ABS will reveal what rate was achieved.
As of October, 10,531 people had refused to complete their census forms, less than the 13,000 who refused in the 2011 census.
In every census, there are questions that people are not keen to answer, particularly about their religion and how much they earn, and this can affect the quality of the data.
Glenn Capuano, a demographer with ID The Population Experts, said that any census questions where the "not stated" category was large would raise a red flag that the quality of the data had been compromised.
Religion is the only census question that is optional.
The ABS told the Senate inquiry into the census failures that non-responses for these types of questions were lower than for the 2011 census, though not all the forms had come in when the analysis was done.
Some people respond to census questions with false or nonsensical answers.
In the 2011 census, more than 64,000 people reported that their religion was 'Jedi' or some variation of it.
Those answers were coded as "not defined".
Mr Capuano said he did not expect the nonsensical answers to be any higher with this census but an increase in the "not defined" categories would be a warning sign that people had given more false answers.
The census website outage could also have affected time-sensitive data.
ABS survey conducted after the census found following key factors
- 42 per cent agreed, to some extent, that the census had been a failure
- 33 per cent agreed, to some extent, that the data collected from the census was unreliable
Mr Glenn Capuano, (demographer with ID The Population Experts) said the ABS would not be releasing the data if it was not reliable.
"I think we can be confident that this census is at least as good as any previous censuses," he said. "I don't think there's any reason not to trust the dataset."
