How to spot a hoax Twitter account – a case study

If you were following the Jan Moir-Stephen Gateley story that was all over Twitter today you may have come across a Twitter account claiming to be Jan Moir herself – @janmoir_uk. It wasn’t her – but it was a convincing attempt, and I thought it might be worth picking out how I and other Twitter users tried to work out the account’s legitimacy.

The too-good-to-be-true test

The first test in these cases is the too-good-to-be-true test, and this works on a number of levels. Jan Moir tweeting in itself was a great story – but not completely unbelievable. Her second tweet said “I have been advised by my editor to create a twitter account and offer my sincere apologies for any upset and distress i have caus” [sic] – a superficially plausible story. Would you buy it?

But there were some other too-good-to-be-true claims in her tweets. One said “My son is gay. I am not homophobic. Please read my article properly.” Does Jan Moir have a son? Is he gay? Would she announce it on Twitter?

And finally, the promise of a formal apology and the tweeted apology itself ticked the too-good-to-be-true box.

Style and personality

Jan is a journalist, and so is unlikely to use a lower case ‘i’ to refer to herself, regardless of the medium. She would also probably capitalise Twitter. And a later tweet uses the phrase “heart felt” which should be one word. Not all journalists have impeccable grammar, but this should raise suspicions.

More suspicious is the fact that the link she gives to her statement takes us to… The Independent newspaper. Would a Mail journalist link to a competitor?

Finally, Jan uses the hashtag #janmoir – unusual on someone’s first day using Twitter – although you might suggest a more experienced user was guiding her.

Who’s following?

One of the first things to check with a potential hoax account is who is following them. If they genuinely work for the Daily Mail, you could expect other staffers to be following them, or official accounts. That wasn’t the case here.

Who’s she following?

Likewise, is this account following the sort of people you would expect, particularly the first few? In this case the first person followed was an American footballer, followed by a US government agency, ReadWriteWeb, Wired magazine, the LA Lakers, 50 Cent, Flaming Lips and various others. That was a real alarm bell. At best they might have been followed on her behalf by a tech support person who was helping her, but more likely is that the hoaxer is either leaving clues or – more likely – following accounts that are likely to get them noticed. The numbers of tech websites in the list suggest that the latter was the case.

Messages to/about them

Who’s talking to this account? What are they saying? Again, if this person genuinely works for the Daily Mail there may be others there talking to her; conversely, if this is a fake account people may be pointing to proof of that fakery.

Suspicious behaviour (bait)

This is similar to the too-good-to-be-true test – the promise of a formal apology and an appearance on Channel 4 News were teases – bait to get people to retweet. And it worked. But ask: why was Channel 4 not talking about the interview they’d supposedly obtained? (After all, this was the top trend on Twitter for a while) Why was Jan Moir issuing apologies on Twitter, and not in a more ‘official’ setting? What was lacking here was…

External verification

This was what really sealed it with @janmoir_uk  - she said she would be appearing on Channel 4 News that night to apologise. A couple of Twitter users asked Channel 4 reporter Krishnan Guru-Murthy whether this was true. “No,” he replied, twice (h/t Todd Nash for pointing me to this). Of course you could have called The Daily Mail to check, too…

Anything else?

What techniques have you used to verify the authenticity of Twitter accounts? It would be great to compile more examples.