The Twitter research released by Queensland University of Technology's Social Media Research Group today brings some much-needed science to discussions that have until now been dominated by Silicon Valley hype merchants and the self-important PowerPoint-waving of SMEGs, or "social media expert gurus". But watch out -- this good science will be ruined by them regardless. The team, led by QUT's Professor Axel Bruns, has identified 2.8 million Australian Twitter accounts as at September 2013, out of 750 million globally. The growth chart shows a typical pattern for internet startups that manage to turn into something that real people use -- as opposed to the vast majority, which either fail to ignite their hype rocket, or do so only to crash and burn soon after.

Bruns said in a press release:
"The early hype that saw substantial numbers of early adopters flock to the service may be gone,  but as an established social medium Twitter now enjoys a relatively steady influx of new Australian accounts. What we cannot see from our data, however, is the number of existing accounts that were deleted each month."
Assuming that each account represents one person -- a dangerous assumption, since some people have multiple accounts and some accounts are operated by multiple people -- that translates into a national Twitter adoption rate of 12%. According to Leslie Nassar (Crikey's own Department of Australia), who developed the Twitter integration for TV programs such as Q&A and set up and managed Victoria Police's Twitter presence, the numbers are in line with other estimates. "It's nice to finally see some rigour applied to calculating the number, rather than the voodoo of Global Users x Australian Global Population Percentage = Number of Australian Users," he told Crikey. The SMRG researchers note that the burst of new user registrations in August 2013 took place in the final weeks before the last federal election, and that the ACT has a significantly higher Twitter adoption rate of 30%. While Canberra does have a massive concentration of political, media and public relations operatives, a key demographic in Twitterland, it still seems remarkable. Even Bruns describes it as "astonishing" in The Converstion today:
"So far we have not seen any indications that an unusually large number of false positives have slipped through our net; there are, however, unusually many accounts whose only identifying feature is their ACT timezone setting, and it is possible that people from other UTC+10 timezones (for example in the northern hemisphere) might have chosen the ACT timezone rather than searching for their own options in the pull-down menu."
Bruns said in the press release:
"If this concentration is confirmed by further analysis, it would point strongly to the growing importance of Twitter in the public debate of political and social matters."
Or, it could be a bunch of political robots and other sock puppet accounts, ready to re-bleat the day's talking points but contributing nothing of substance. Or it could be something else entirely. Twitter has clearly become "important" to the political process. Witness the effort put into Twitter by many politicians. According to NPR's On the Media, the Israeli government has around 40 people working on social media propaganda right now. But is it "significant"? That is, does it actually result in anyone changing their opinion, behaviour or vote? In the commercial world, does the buzz of Twitter really translate into more products sold? We simply don't know. That would be a whole new scale of research. SMEGs simply crow about the success of campaigns that generate "buzz" and "engagement", and the failures go down the Memory Hole. Their castle is built on the sands of survivorship bias. We see the same sort of thing in their talk of influencers, which is little more than internet astrology, proven to be rubbish. Today's SMRG research is a fine foundation for further real, validated research on Australia's use of social media. More will appear in coming months, and I'm sure it will be as much a joy as their mapping of Twitter's political connections. But while Bruns and his team will know its limitations, SMEGs will not. Beware of their dodgy factoids.