What Is Independent Australia Independent Of?

Quality control, consistency, accuracy and editorial skill.

At least, if my own encounters with its opinion-polling coverage are any guide.

According to its "About Us" page, Independent Australia is "a progressive journal focusing on politics, democracy, the environment, Australian politics and Australian identity".

Further on:

"IA supports quality investigative journalism, as well as citizen journalism and a diversity of voices.  It believes Australians are short-changed by the mass media - and so dedicates itself to seeking out the truth and informing the public.

Independent Australia believes in a fully and truly independent Australia, a nation that determines its own future, a nation that protects its citizens, its environment and its future.  A country that is fair and free".

Which all sounds well and good, as troubled as the often pretentious use of the term "progressive" in the left has often been.  IA also claims to support independent candidates and oppose partisanship, though the waters here are slightly muddied by its endorsement of the curious so-called "Australian Independents" party.  But the real problem starts when we get to this:

"IA also features an exclusive weekly column by the Australian literary legend Bob Ellis".



Now, it's great to be featuring columns by literary legends if those columns are about literature, or better still are literature.  But Ellis's (which IA pays for) mostly aren't, although some are certainly fiction.  The vast majority are political; according to a list of his articles on the site, it's been a few years since he namedropped a writer in the headline of one of his IA pieces for a reason other than making political points.

It would be one thing if our hero stuck to his continual attacks on Abbott, Morrison, Hockey etc, political thoughts which clearly interest a lot more people than mine, and are probably no more biased or ridiculous than those of some on the right who do much the same thing for major pay.  But  Ellis also runs his usual side-line of pollster-bagging - typically, calling pollsters (excluding Morgan) crooks who deserve to be imprisoned, and doing so on the basis of comically error-riddled claims about polls.

I gave an example of just how many factual errors can be found in a single Ellis polling article in Bob Ellis: Embarrassment To The Left.   My interest here is not primarily in chopping up another one, but in the publication of this guff by a source that claims to support good reporting and informing the public, but that does not fact-check comments about polls pre-publication.  Even publishing such pieces at all shows that IA betrays its claimed journalistic principles on a regular basis, but I was curious about how they would react if some errors in one of these pieces were pointed out.

In a recent IA piece, "Today's Ipsos and Newspoll" Ellis made the following claims:

"Ipsos showed a tie. To understand how they did this – give Labor 400,000 fewer votes than Newspoll – one must look at their fool methodology. As I understand it, their machines rang 80,000 people, of whom 1,403 took the call. These were people prepared to talk to a machine – the old, the mad, the loveless, the paranoid, the paupered, the resentful, society’s grumps and invalids – on three nights, when half the adult population were out of the house, in cinemas and restaurants, their mobiles with them, unrung by by either pollster."

Ignoring the question of how many of Ellis's eight insults after "machine" could be fairly applied to himself, the problem here is that Ellis has assumed Ipsos is a robopollster that only rings landlines. Both these claims are false.  In the case of the claim that mobiles were "unrung by either pollster", one can just go to Ipsos's Australian website, go to their Press Centre and find the report of their latest poll.  Right there at the bottom, "31% of sample comprised mobile phone numbers".

So I put a post in comments to the article pointing out just some of the errors of this Ellis piece.  When I checked back today, it was gone:

Whether because of my comment or not, the site did add an asterisk to Ellis's "As I understand it" with a note that "In fact, it appears the Ipsos surveyors are human."  But Ellis's false claim that Ipsos do not call mobiles - without which pretty much his whole argument against Ipsos falls over - remains uncorrected on the article.

I think it especially worthwhile that this matter be set straight, because Ipsos is currently Coalition-leaning compared to other pollsters.  It would be easy for someone who has been suckered by Ellis's rants about Newspoll not calling mobiles to deduce from that that Ipsos has a bias caused by not calling mobiles.  But the fact is that Ipsos call a lot of mobiles and have put a lot of energy into trying to get their mobile phone sampling right.    Those looking for reasons why Ipsos produces more Coalition-friendly results than the other five major active pollsters clearly need to look elsewhere. 

So I think this is very poor form by IA to remove a comment that corrects two errors in an article, but not to even clean up both of the errors properly.  

As to why it was deleted (other than the site perhaps being overprotective of its star fiction writer) I can only guess.  Perhaps the "then there wouldn't be much left" was deemed to have stepped on the toes of the IA comments policy, under which they may remove comments that are "inflammatory".  But on that matter, the IA comments policy* is so nebulous that almost anything could be said to disobey it, and so foolish that to breach it in a case like this verges on duty.  

Bob Ellis is permitted to routinely defame opinion pollsters (and both he and IA would go down screaming if the directors sued claiming they'd been personally defamed by implication).  He's permitted to inflame discussion by accusing pollsters of corruption.  He's permitted to "close down debate" by asserting that those he disagrees with are crims and crooks and otherwise contemptible. Yet if you want to be even a little bit snarky in response to him, your comment may be deleted at whim with no guarantee of consistency about what is deleted and what isn't (and no correspondence entered into either, supposedly).  This is all a blatant double standard.  Rather than allowing an article writer to brazenly flout the standards applied to comments, IA should realise that if you want your commenters to maintain a certain tone and manner, then you should ensure that your authors do too.  They, after all, set the scene.   

Anyway, at least so far as opinion polling is concerned, "Independent Australia" is a website that betrays its own stated principles by publishing an author who continually churns out false claims, that is sloppy about correcting factual errors even when advised of them, and that wants to publish material about polling but isn't remotely serious about the rigour needed to do that sort of discussion adequately.

Independent Australia claims to comply with the MEAA Code of Ethics.  It doesn't say whether it's a member.  The articles by Ellis that it publishes are in blatant breach of item 1, and in my view also breach item 4. IA isn't serious (yet) about item 12 in this instance either.

Any site that wants to be an alternative to the mass media short-changing readers must stop short-changing them itself.

(* For those who were wondering, this site has no comments policy, though I've co-authored comments policies on other sites.  I've considered having one here, but I am not convinced it's needed based on my aims in running the site and the very small number of posters who create problems. If I moderate something and someone doesn't like it, I can explain why; if someone wants to know if I'd accept something contentious they can ask.  Indeed, I've only modded about four five posters since this site began.)

Just after this piece was written I noticed another par-for-the-course howler.  Ellis claims "Both surveys got Queensland wrong by two percent".  In fact, Ipsos did not even do polling for the Queensland state election!

Oh, and it looks like the Managing Editor of IA is just as bad as his attack dog on these matters:

Does he even know what push-polling is?

========================================================================

The above ends the article as such.  The updates below are just Twitter flamewar stuff with the chap running IA and his lone (so far) online mate, following this article.  For those not into such things, you can safely stop now.

------------------------------------------------------------------------------------------------------------------

Update (Wednesday): In a hole, Donovan immediately starts digging:

Ah, so apparently in Donovan's world it is not rude to call people crooks on the basis of accusations that are demonstrably false and where no attempt has been made to check facts, just passing them off on the basis of handwaves like "as I understand it".  It is apparently not rude in Donovan's world to mislead his own readers by publishing untrue claims about polling and failing to fix them when corrected.  And it is apparently not rude in Donovan's world to suggest that I "can GFH".  

Donovan's extremely peculiar and selective conception of rudeness aside, he's just plain wrong.  If Ellis says he understands something to be true, he is claiming that it is true, just as if I was saying "I understand Dave Donovan is a poor editor of polling articles", I would be claiming Dave Donovan was a poor editor of polling articles.  Donovan's claim that my comment was inaccurate is therefore false, and suggests that he is also a poor moderator.

And meanwhile, Donovan continues allowing his readers to be deceived into believing that Ipsos doesn't call mobiles.

Thursday: Donovan has made another tweet (language warning) insulting me again in a puerile and braindead if especially harmless fashion, but that's all we've heard from his towering intellect on the matter.  Meanwhile an evident ally of Donovan, the ranting Twitter nutcase Frank Calabrese (@frankscan65), a former problem poster on Poll Bludger, has been defaming me at high volume in vaguely similar style to Bob Ellis, but even more repetitive (and with a big CAPS LOCK problem to boot).  I did feed the troll until it had burst a few dozen times (a pet hobby) but have now put Calabrese on mute to stop his tweets flooding my notifications, so I won't necessarily see all his Twitter nonsense (and won't always bother replying).  The only part worthy of refuting here is that Calabrese falsely claims I threatened to sue him for defamation - in reality I just said I was thinking about blocking him for it.

Thursday evening: more DD insults (again showing DD will throw mud but refuses to engage with the core issue of his site publishing false statements about polls and failing to correct them while claiming to inform the public and correct errors).  Among the latest rubbish is that DD declares I am only a "psephologist" in my head.  Alas for him, the label was in fact conferred on me by Tasmanian Times, a site which he should regard as authoritative since its stated aims so resemble his own, its audience is just as infested with the lunatic end of the left, and its moderation policies are almost as bad.  But at least TT is run by someone who genuinely cares about informing the public about electoral issues.

Friday morning: The sheer stupidity and belligerence of Calabrese is best shown by the following.  Calabrese claims to be an ALP supporter.  His Twitter feed has this pinned tweet:

So I sent him some anti-ALP tweets, one of them saying the problem with Labor was it didn't expel people like him, and another accusing Labor of trying to destroy free speech in Tasmania.

He retweeted them!

-------------------------------------------------------------------------------------------------------------------------