Receive up-to-the-minute news updates on the hottest topics with NewsHub. Install now.

Census fail: Doubts remain over accuracy of 2016 data despite being declared ‘fit for purpose’

27 June 2017 2:07 AM
120 0
Census fail: Doubts remain over accuracy of 2016 data despite being declared ‘fit for purpose’

AFTER much embarrassment, the 2016 Census results have been released and Australians don’t have to worry about it again for another five years — or do we?

It’s easy to get swept away in the tide of new Census data that tells us we’re turning our backs to religion as we get older and are welcoming more migrants from Asia than Europe, but we should back up for a second.

Let’s not forget the great #CensusFail which made a mockery of the Australian Bureau of Statistics and spawned a string of hilarious memes when the Census website went down in a blaze of glory.

The online survey crashed when one of the routers experienced a clumsy cyber attack from overseas and mistakenly distributed information suggesting data had been compromised, causing Australians to fret over the safety of their personal information.

In October, computer giant and Census contractor IBM admitted to a Senate inquiry the website outage could have been avoided by simply turning one of its routers off and on again.

Prime Minister Malcolm Turnbull said the debacle was “utterly predictable” and said IBM had paid a “very substantial” multi-million dollar settlement to cover the fallout.

Chief statistician David Kalisch says yes. The ABS established an Independent Assurance Panel to provide an independent view of the Census data quality. In a report released today, the panel found the data was “fit-for-purpose”.

However, despite the data concluding that Melbourne was growing faster than Sydney, members of the panel were “not in a position to make a judgment below the state and/or territory level”.

It was also estimated in a Post Enumeration Survey that 786,689 Aboriginal and Torres Strait Islander people should have been counted in the Census, compared to the 648,939 people who were actually counted.

“This is equivalent to a net undercount of 137,750 people, or a rate of 17.5 per cent of the estimated Post Enumeration Survey population count,” panellists wrote.

The SBS also reported an overestimation by the ABS of which dwellings actually had people in them has dropped the response rate for the 2016 Census close to the minimum required for it to be considered a success.

According to results released on Tuesday, the response rate dropped from the preliminary figure of 96 per cent announced in October to 95.1 per cent. The ABS has previously said the “minimum required response rate” was 93.3 per cent.

Former Australian Bureau of Statistics chief Bill McLennan previously said he was “certain there is an impact” by the backlash on the Census data.

“One thing I am also sure of is that it is not going to be the same right across Australia, it is not going to be the same across socio-economic groups.

“You can’t have people complaining all the time about privacy and being forced by the compulsory nature (of the Census) and it not have some impact on what they say.

“It’s a psychological thing, it’s not a statistical thing. You tell people you’re going to prosecute the buggers and they’ll tell you something.

“What I can categorically say from my experiences, is that there is going to be a certain proportion that does not answer properly or deliberately give you a wrong answer.”


Share in social networks:

Comments - 0