Opinion

The Australian Census: will it crash the internet and can we trust the data?

If the internet crashes when one million people hit a site like Click Frenzy on sale day, how does the Federal Government expect its site to hold up with 15 million of us due to fill out the Census tonight, asks Alice Almeida?

It was November, 2012. I was at home on the couch in my pj’s, laptop ready and active, credit card details memorized, action purchase plan plotted out, I had told mum not to call – I was ready to bag a bargain!

Click Frenzy was launching in Australia that night and the promised deals were incredible! Nothing was going to stop me from buying designer goods for 20% of the normal price. And then…. The internet broke.

alice almeida- hitwise

We have all experienced it at some point. Whether it’s online sales, buying concert or airline tickets – it seems the internet breaking is not a rarity. Think Justin Bieber and Kim Kardashian nude (not together, god help us if that ever happens!) and how the internet practically melted.

Each time the internet broke, it was due to too many people accessing a particular website at once. It couldn’t handle the overload of visitors, so it throws a tantrum and shuts down, throwing millions of people into fits of anger and frustration.

kim kardashian app stor emoji

So answer me this…. How exactly is the internet going to cope when the entire Australian population logs on to complete the Census forms tonight?

If it crashes with one million people on one website how will it cope with an estimated 15 million* people accessing at one time? What is the back-up plan? And given people will be fined if they don’t complete the Census, how will they manage this?australian census homepage

Problem is, access is only one of the major problems facing the Census this year. The amount of negative press around it has been huge, mainly around privacy issues.

This year sees the first time that it’s compulsory for you to give out PII (Personal Identifiable Information) to the ABS. Privacy advocates are up in arms, and rightly so. But in speaking to a few people who don’t really know, I was concerned that not more people are being made aware of what’s happening to the PII data.

To gain a broader understanding of how Australians are feeling about this year’s Census, I turned to our AudienceView platform which tracks online search terms and profiles audiences (VERY important to note that we don’t collect PII data) which can be then segmented by a particular topic. So let me break it down for you.

In just the past week, there were 314,000 people seeking out information on the Census across 68 search terms using the word ‘Census’. Below is a word chart showcasing the different search terms Australians are using.

census graphic 1

In order to distinguish between negative and positive terms, I clumped them into themes – no surprises, there was one positive search term from a job seeker, and that’s it. The four themes are; General, Fines/Penalty for not completing, Privacy issues and my favourite ‘How to get out of it’.

census graphic - search terms - 2

For 40% of all search terms around ‘Census’ to be negative in nature should be a huge concern for the ABS. People are actively seeking out how to avoid the Census.

As a researcher, the results from this Census are a concern. Surveying an angry audience has many risks involved mainly ‘F**k you answers. How many ‘FU’ answers will there be? Sure, you’ll be fined if you fudge your name, but household income? Number of children? How many of those responses will be ‘top box ticked’ in order to complete it and avoid the fine?

So many government agencies rely on Census data to plan such as transportation, health services, ageing support, etc. Can this data really be used to decide such significant things?

As I was analysing the data I assumed that the people who had negative feelings towards the Census would be the digitally savvy, data conscious, privacy aware generation but in profiling the ‘Negatives’ I was proven wrong. 25% of those who are concerned about privacy and are seeking out ways to get out of doing the Census are 55+. Now I could be stereotyping, but I feel there’s a much higher fear of the online world held by this demographic, especially around privacy and data so it’s not that much of a surprise. See the below chart.

census graphic - age of negatives - 3.png

When looking at our Mosaic segments, the below groups are the most concerned around privacy.

B: Knowledgeable Success –   Well educated, wealthy family and couple households in suburban areas of major suburbs.  (17%)

F: New Homes & Hopes –  Young families in new homes in some of the fastest growing suburbs in Australia. (14%)

C: Independence & Careers – Apartment dwelling young and maturing professionals and students in city central locations.  (12%)

My concern when looking at the mosaic segments is the socio-economic split. Highly educated, wealthy professionals and families are the largest audience who do not want to participate in the Census, and who are possibly more able to tolerate the fine.

This means that the lower socio-economic groups who cannot afford the fines, and may not hold the knowledge around concerns of data privacy, will be the ones to complete the Census (and most likely in an honest way). This will of course skew the results substantially.

I understand the importance of the Census (I do work in research after all) but the concerns around privacy should not and cannot be ignored. Time will tell, though.

As I mentioned at the start of this blog, we might not even get the option to complete the forms if the internet makes the decision for us, and has a Kim Kardashian-style meltdown…

For more information on AudienceView, contact the Hitwise team here.

Alice Almeida is APAC manager of innovation and insights at Hitwise 

*ABS prediction

ADVERTISEMENT

Get the latest media and marketing industry news (and views) direct to your inbox.

Sign up to the free Mumbrella newsletter now.

 

SUBSCRIBE

Sign up to our free daily update to get the latest in media and marketing.