New study suggests there are 2.8m Australian Twitter accounts
Click to enlarge
There are more than 2.8m Australian accounts registered on social media platform Twitter a new study suggests, with Canberra and the ACT by far the territory with the highest per capita concentration of users.
The new study released today by the Queensland University of Technology, follows up a 2012 study which mapped the Australian Twittersphere, and gives new insight into the growth, development and uptake of the platform down under.
“When we did that (study) we found about 1m accounts,” said Axel Bruns, project leader who is a researcher at QUT’s Creative Industries Faculty.
“What we have done now is to go through the entire Twitter user base and because all users have an ID basically started with ID zero and then worked up to where we stop finding users, which is upwards of two billion IDs.
“So we have gone through the entire user base to get the publicly profile information on all Twitter user IDs. That is a lengthy process because of the limitations Twitter puts on its API (a funnel of information which Twitter makes available to allow software providers to interact with the platform).
“These are the accounts that we believe are Australian,” said Bruns, while warning that the number does not represent the number of people on the platform.
“I want to make the distinction between users and account really clear. Because of course some people have multiple accounts or some accounts might be run by multiple users and therefore it is difficult to say how many users there are.
“We believe however there are 2.8m registered accounts as of when we finished this study in September last year.”
Twitter Australia does not publicly release country by country breakdowns of its user base however, agency sources told Mumbrella earlier this year that the number of domestic users being claimed by the Australian Twitter sales team is now as high as four million accounts.
In an interview Australian managing director Karen Stocks, at the time of the one year anniversary of the opening of the Australian office, said: “We don’t breakdown numbers at the country level… But I am very, very happy with the growth we are seeing.”
Click to enlarge.
The QUT study also provides insight into this uptake with a graph showing a surge in uptake to more than 110,000 in early 2009 (when Twitter really took off), then falling back to between 30,000-60,000 until January 2013, when Twitter began building its local office.
“It is hard to identify why people signed up but the factors point to local activities by Twitter or the events that drive people to Twitter have an impact,” said Bruns. “It is interesting to see how long it has taken for Twitter to take off.”
He also noted that the uptake and development was very different to rival Facebook, which does release local numbers and claims some 12m active monthly users in Australia.
“Partly that’s because it’s a very different kind of social media network to Facebook.
“One is built around strong ties and friendship networks, Twitter is a much more open and flat space and you get more people who use it to follow current events and also promote themselves. Thats why it doesn’t surprise me that it is not as big as Facebook.”
Bruns also warns that the uptake at the end of the survey which concludes in September may also have been driven by the federal election or that Twitter’s anti-spam bot programs may not have removed yet. Twitter is constantly filters for fake accounts designed to spam users with questionable advertising, for scams and pornography.
“There is a big jump that we saw towards the end of our gathering period, which coincided with the election period last year,” he said. “But it may also be that Twitter’s anti spam bots have not gone through all of them yet.”
The study also provides the first window into the geographical distribution of Australian Twitter accounts, with Sydney the biggest s with 394,000, Melbourne with 329,000, then Brisbane and Perth with 138,000 and 119,000 accounts respectively.
Adelaide has 79,000 accounts however many Twitter users do not provide a city or state for their registration and there are also other anomalies, including the exceptionally high per capita number of users in Canberra and the ACT.
“If you look at (Australia’s) population distribution it probably looks quite similar,” said Bruns. “In terms of average per state I think it is reasonably even (in per capita terms) across all states, except the ACT.
“The ACT has far more sign up per head of population and that is strange where you have a concentration of a particular group of people. It was 30,000 to Canberra itself and 115,000 to the ACT overall,” he said, referring to the Territory, which according to the last census had a population of 357,000.
“In that case of the ACT we have a much larger proportion of people who selected an ACT timezone and don’t give much more information about where they are.”
Bruns said the number was generated by those choosing to select the ACT timezone in their registration and on their profile. He noted that more than 400,000 could not be allocated to a state or territory.
He added: “We have looked at Europe where we see an over representation of people in Amsterdam and Athens… but it doesn’t explain Canberra.
“The question there is why choose the ACT timezone. We have found that in some cases people will just choose the first city that comes up in their timezone. Having said that Canberra doesn’t come up first, Brisbane does, which therefore doesn’t explain why people choose the Canberra timezone.
“However, we are a bit surprised by it and we are treating it with some suspicion. I fully accept that the numbers (for Canberra/ACT) may be overinflated however, having said that the demographics of the ACT may also make the demographics of who uses Twitter in Australia. You have an unusually well educated, unusually urbanised population in the ACT just because of what it is as a territory.”
The next stage of the QUT study is to delve the new 2.8m account database, mapping their interactions and attempting to draw out new insights like the number of active users and how often they use the social platform.
“Now we have the 2.8m accounts we are about to get mapping the full network for those accounts, he said. “This will give us a more comprehensive understanding. We will see how it has changed from early 2012, particular since Twitter has done that outreach with sporting codes etc.
“We will also look at the patterns within the data that we have got, patterns like how many of these accounts are private or verified or what the level of activity is by account.
“We will be very interested to see of these 2.8m how many have actually tweeted in the last month.”
Nic Christensen
So 14.8% of the population follows Kyle and Jackie O.
Unless of course some twitter followers are fake.
User ID not verified.
“Unless of course some twitter followers are fake.”
Massive amounts are fake. I personally run 30 or so accounts and you develop a knack for spotting them. It’s hilarious how many big name brands have 90%+ fakes.
(Note to industry, audit all your social media “superstars”. Chances are they’re just good at buying those results you laud them for.)
User ID not verified.
Irrelevant information. Imagine if TV networks did this, x number of TVs sold therefore X number must be watching. The reality is most twitter accounts are dormant, it wouldn’t be more than 250 000 users posting more than once a week.
User ID not verified.
Im from Canberra. Not everyone here picks ‘Canberra’ when selecting where we are from because Canberra is thought of as the inner city. We use our suburb names or ACT as there are only suburbs and bush in the ACT, it’s not like the mix in other states/territories.
Plus we are sick of the crap we get whenever Canberra is mentioned in the media because of the politicians we have the misfortune of having to lodge from the rest of the country…
User ID not verified.