To Bolt’s readers – let us hope that you’re a little sharper than the author of the site from whence you came, understanding both a well known political meme when you see one, and the importance of empirical reality guiding political commentary. If you can also manage to leave your dramaqueens at the door and spare me the poor little victim bullshit, you’ll be four points up on Bolt himself.
Something else you might be interested in knowing – Bolt removed the link to this article in his latest spiel after the above little update was posted, reinforcing the “Undescended Testicle of the Internet” meme described below. After his blog was caught out and a few of us had much mirth about it on Twitter, it miraculously reappeared! You can read all about it and Bolt’s problems here. Andrew – man up and grow a pair will you, or at least get your online minder to grow some for you.
——
You know what shits me to tears, I mean really shits me to tears to the point where pulling someone’s arm off and beating them to death with their own severed limb starts sounding like one of the more civilised responses available?
It’s when MSM sites like this herd the dross of the internet into advertising, by tickling the bigoted little underbellies of their audience with pig-ignorant bullshit being masqueraded around as “fact”.
And there is no larger magnet for outright bigotry than asylum seekers.
With refugees it’s literally Moral Panic Bingo; Islam, terrorists, race, xenophobia – refugees are the ultimate canvas upon which the shallow end of the public affairs pool can paint their own preferred pathological animosities. If you don’t believe me, then undertake an experiment:
Write down 9 favourite themes of the small minded nutjob set, not specifically about any given thing, any old generically bigoted idiocy will do – then pop on over to the usual creatures that prey on such feeble minded antipathy and read the comments sections on any post they have about asylum seekers. Every time one of your predicted themes is mentioned by a commenter, mark it off – you won’t have to read far before you’ll be shouting “Wingnut Bingo!”.
Of all the Wingnut Bingo halls in the land, there is none bigger than that hosted by The Undescended Testicle.*
He started yesterday with his sneering innuendo, of asylum seekers being “Lured by Rudd to their deaths?”. There really are no boundaries that Bolt’s hysterical Rudd Rage refuses to cross – although the only thing really being “lured” here are miscreants by the bucketful into Andrew Bolt’s site –herding the dross of the internet into News Ltd advertising by playing up to their shallow and spiteful little fantasies.
Bolt’s argument is as predictable as it is convenient – Rudd is being soft on border protection and as a result, increasing hordes of refugees are arriving on our fair shores. He doesn’t quite go so far as to mention rape and pillage – he seems to save that meme for the Sudanese and Somali’s. In fact, he’s put on his Mister Caring and Sharing righteous Rudd Rage face today – it’s all about the lives of the people he’s so often demonised you see.. roads, conversion and Damascus – that type of thing.
His problem is that everyone else in journo land (apart from that wretched little Poison Dwarf seen on Sky Nooz Agenda yesterday, to accurately describe my thoughts on that creature’s performance is to invite a law suit) has actually made the effort to find out about the reality of the asylum seeker figures and have treated the current situation accordingly in their reporting of the matter. Quite a sophisticated, sensitive and highly responsible approach for which the entire industry, apart from the notable few, deserve a marvelous round of applause.
But oh no, not our Caped Crusader for the Cretinous – he accuses Philip Coorey of following government talking point memo’s for having the audacity to state the obvious, that “The number of asylum seekers has surged worldwide since 2007, and the greater number coming to Australia has been part of that trend.”
Bolt reckons Amnesty international is part of the conspiracy for quoting the UN High Commissioner for Refugees saying the same thing. Even Michelle Grattan is accused of being some government mouthpiece for writing “a refugee tide is swelling around the world…the modest rise (in Australia) reflects the general international trend.”
“But how true is all this really?” asks The Undescended Testicle – as if it were actually a real question rather than an invitation to the feeble minded, to vent spleen about shit they know not.
“Yes, the UN High Commission for Refugees reported a 12 per cent increase world-wide of asylum claims in 2008, but noted a much bigger increase in Australia and New Zealand – 19 per cent” he proclaimed, with a sort of mathematical naivete usually found in housebricks.
Let’s look at these numbers – between 2007 and 2008, global asylum applications increased by 41,580 while Australian applications increased by 779. The enormous disparity between those two numbers would give a numerate person pause to consider the meaning of the word “variance”. A very small change in the Australian numbers will have large implications for the percentage change in any given year. To show how sensitive these numbers are, in 2008 there were 4750 asylum applications made – a 19.6% increase from the 2007 numbers.
If 250 less people made applications, that percentage would have reduced to 13.3%.
What’s 250 people I hear you ask? Why, it’s one people smuggling operation.
So the percentage change in Australia of asylum applications in any given year is highly sensitive – to the point where we really have to look at the reality over a number of years to allow us to start drawing conclusions.
His argument is that global asylum seeker numbers have no bearing on Australian numbers, that it’s local factors that cause the Australian figures – local factors like, er, Kevin Rudd.
So let’s take Bolt’s pig-ignorant horseshit out the back and kill it.
First up, let’s compare global and Australian asylum number applications, yearly, since 1990. All of the following analysis, charts and bits and pieces are, unlike Andrew Bolt, my own work based on my own research for which I am professionally qualified to undertake.
That clearly looks like a bit or correlation there in the movement of the two series through time, so let’s change that exact same data into a scatter plot, to give us a different perspective and run a regression line through it to point out the bleedingly obvious.
OK, denying the numerical reality of correlation here is starting to border on the completely fucking ridiculous. But let us not stop here – we all know that correlation doesn’t necessarily imply causation, but we do have some statistical tools up our sleeve to go one better than mere simple correlation, we can look for the existence of a thing called granger causality.
Granger causality might sound a bit high falootin’, but it’s pretty easy to get your head around once it’s explained. Effectively, it’s correlation on steroids.
Let’s say we have two data series, x and y, where each series represents the value of some variable at some consistent period of time. The monthly unemployment rate is an example of such a series, quarterly GDP growth is another.
What granger causality tests do is firstly, determine how much of the current value of a series (say series y) can be explained by previous values of y, and then determines whether by adding lagged values of the other series x (lagged values, as in the values of last month, the month before, the month before that etc) we can increase that explanatory power.
If previous values of x can help predict future values of y at a statistically significant level, then we can say that x granger causes y. It’s not causation in the literal sense, but it is as close to statistical causation that can be proven – and it is far more rigorous than mere correlation.
We don’t have enough data to do this test with the yearly data set, but fortunately for us, and unfortunately for The Undescended Testicle and his piffle peddling, there is another, higher resolution data set at the UNHCR – asylum applications from 2000 to 2008, calculated quarterly.
We can then calculate from that data the percentage change in asylum applications, per quarter, for both Australia and globally – when we do, this is what it looks like:
This data may look messy, and to the naked eye it is – but what the naked eye might not show (depending on how professionally nerdy you are) is the vast wealth of information contained with in it, which is what we are going to extract.
This series gives us just enough observations to test for Granger Causality and be confident of its results – so breaking out the trusty Eviews econometric software, our test results come in like this:
The two series here are PCHANGEAUS which is the percentage change in quarterly asylum applications in Australia, and PCCHANGETOTAL which is the percentage change in quarterly asylum applications made globally. (for the statistical aficionados out there, the test results for the simple first difference of the quarterly numbers, rather than the percentage change results are virtually identical)
We are testing a null hypothesis that says “the quarterly change in Australian asylum applications does not granger cause the quarterly change in global numbers”. With a probability value of 0.46, we cannot reject this null hypothesis – we cannot reject the statement that changes in Australian application numbers do not granger cause changes in global application numbers. That’s to be expected – we don’t expect the world to be the slave of the Australian experience, but what about the opposite – is Australia a slave to the global experience?
With the hypothesis “The change in global numbers does not granger cause the change in Australian numbers”, we have a probability score of 0.03, which means in very rough terms, we can be over 95% confident that this hypothesis can be rejected, we are 95% sure that this statement is false and that changes in global application numbers over the previous four quarters do indeed explain changes in the number of asylum applications that Australia experiences.
We can say, confidently, that changes in the numbers of global asylum applications granger cause the changes in the numbers of Australian asylum applications.
We can say that global levels of asylum seekers, and the global trends they create, do indeed significantly affect Australian asylum application numbers, and come as close to causing those numbers as one can statistically get.
So Andrew Bolt should shut the fuck up and stop spreading bullshit around when it is very clearly not true.
In fact, so seriously pathetic is this broader argument of his that we only need to look at the yearly data to clobber it.
Well look at that, asylum application numbers have been increasing since 2005!
Now there’s an inconvenient truth, and one that no amount of hysterical Rudd Rage will explain away.
Apologies to those that normally expect a more polite and civilised Possum, but honestly, sometimes you need to make a damn point.
* The Undescended Testicle.
Andrew Bolt doesn’t like criticism – fair enough, but the cure for that is to stop writing bullshit – anyways, he especially doesn’t like criticism when it pings him exactly for what he does. He really, really, really doesn’t like that kind of criticism when it comes from Crikey. A while ago there was a flair up with Bolt over Somali crime numbers, where Bolt couldn’t ignore what was said here (our readership is too widespread, and there was far too much mirth going on in the blogosphere at his expense over it all), but neither could he confront it properly because to do so would be to send his readers to a complete takedown of their hero, that describes many of them as the suckers they are. So the poor princess was in a bit of a pickle.
But rather than have the balls to link into the article here and address it properly, he used just one ball and quoted very, very selectively from the article (in comments too) in such a way as to remove all context and pretend I was actually agreeing with what he was on about.
He has enough balls to selectively confront his detractors, but not enough balls to do it in the open and transparent way that is the acceptable norm of the modern online world.
He is The Undescended Testicle of the Internetz
Elsewhere: Ken Parish at Club Troppo, Robert Merkel at Larvatus Prodeo, The Piping Shrike, Agmates, Pure Poison (of course!), Andrew Bartlett, The Political Sword, Jack the Insider with a terrific piece.
78 Comments
Pages: [1] 2 » Show All
You are a god.
Calling people undescended testicles, miscreants and bigoted idiots really does a disservice to your argument Polly.
Mathematicians 1 – Morons 0. Nice work Possum.
My only problem with this argument is that you let the mongrel off too lightly.
nice post Possum, your great when you are angry
Have you had a look at this post by Bolt?
http://blogs.news.com.au/heraldsun/andrewbolt/index.php/heraldsun/comments/what_world_are_the_models_on/
Am i taking crazy pills? Bolt seems certain it shows a trend from 1979 despite the first year being 1999 on the graph, am i missing something really obvious?
GP – sometimes it’s not worth staying righteous.
Well, I am someon who thinks Rudd’s change in policies have at least contributed to the increase, even if not wholly responsible.
But I do think Bolt’s “Lured by Rudd to their deaths?” was utterly appalling. So obviously written to get a reaction, I rate Bolt as the male counterpart to Ann Coulter.
Cyclosarin, can you please explain your reasoning and/or evidence?
Ta.
Nice statistics Possum.
Of course they will be ignored by the intended.
cyclo – read possum’s piece again to see the idiocy of your belief
No 6
I think you can mount a credible argument without reverting to the cretinous depths of certain News bloggers.
No 12
Andrew Bolt, Possum’s refutation of Bolt was hardly “professional” (according to your assessment), given how many unnecessary use of expletives and other offensive language. Obscene I tell you.
seconded
Pompous git!
I’m sorry Poss. but sodomizing Bolt with a calculator, is far too kind. An infinitely worse fate would be the one handed out to King Edward II of England as a result of the infamous Piers Gaveston affaire.
Otherwise a very good article. Cheers.
Do you disagree with any of Possum’s analysis GP, or is it just his language?
I should add, I think Possum should have held off on the name calling – it just gives Bolt an easy out to say that the piece is just an abusive rant.
I agree 100% with his findings (and methodology)
Glad I read your article, but geez your angry hurts my eyes.
Why Possum should be knighted and replace news ltd commentators
Quite simply the most comprehensive and righteous pwning I think I’ve read.
I dips me lid.
I’ve sworn off sarcasm and name calling. Glad everyone hasn’t. Go Possum!
An exceptional piece of snark, Possum.
Bravo! Poss. I was calling him the Rush Limbaugh of Australian media, but I like you undecended testicle tag much better. I won’t go to his page, I think that he and some of the commentators have gone over the far right side of nuttersville, and I consider myself centre left, which is probably on the right of the old democrats.
Sigh…I was hoping that Australia didn’t have to go through this shit again. Poss you forgot about the shock(ing)jocks they were flogging the airwaves today too.
GP, sigh … I sorry, but you need to grow up. If the swearing is all you can comment on, then you seriously missed the point.
But Possum, tell us what you really think.
You have to feel sorry for Bolt. He’s mathematically retarded and is only left with self-opinionation to support his prejudices.
Hi Possum
Interesting stuff, and great to see you making use of the data.
I hope you don’t mind, but I played around with the data a bit myself. The thing that took my interest is that while your granger causality tests using the quarterly data suggest that causation runs from the change in global asylum seekers to the change in Australian asylum seekers, in the annual data it looks suspiciously like the opposite is the case.
Granted that the annual data is problematic because of the very few degrees of freedom, but it goes further back and it is possible that there is some useful inference that can be made. Indeed, given the few degrees of freedom available in the annual data any statistically signicant results with that data would be worth thinking about.
So, I did a couple of things, first I did the Granger causality test using the annual data (with 1 lag). That test assigns a p value of 0.88 to the null that the change in the rest of the world does not granger cause the change in Australia, but assigns a p value of 0.043 to the null that the change in Australia does does not granger cause the change in the rest of the world. Thus, using the annual data, one arrives at the opposite conclusion from the quarterly data. The same thing can be seen if one simply regresses growth in the rest of the world on itself lagged a year and growth in australia lagged a year – the lagged Australia term is signficant at 5%. Those results seem to make sense to me given the fairly clear leading relationship that the Australian applications have over the sample period using annual data.
Now, this is where you have to go back to the correlation does not equal causation thing, because it would be difficult to write down a theoretical model whereby increases in Australian asylum seeker applications had genuinely causal implications for asylum seekers in the rest of the world. It is much more likely that there is an external factor that explains both and that for one reason or another, over the sample period, annual Australian applications have tended to lead those in the rest of the world by a year.
What it does suggest though is that you want to be careful in pushing those quarterly granger causality results too far as the truth might be a bit more complicated than the tests suggest.
So, I’d say yes, the argument that Rudd policies are to blame for the increase in asylum applications is clearly dumb and being peddled by Bolt et al for dark political purposes.
But I’d also say that the evidence that ROW applications granger cause Australian applications is more mixed as the quarterly data and annual data are in conflict. It almost certainly means that firm statistical conclusions should probably wait until many more years of quarterly and annual data are available, and there are data available on the causal factors that are likely to be explaining both variables. Perhaps you have more faith in the quarterly data, but I’d be very wary of reaching any firm conclusions on the basis of just 8 years of quarterly data (or 18 years of annual data).
Oh, one more thing. Because I wasn’t able to find the data quickly enough on the UNHCR website, I took your graph and estimated the numbers. That means óf course that when you replicate what I did with the actual annual data, you will get slightly different estimates, but I’d be surprised if the broad inferences were much different. Also, the choice of 1 lag in the annual data was simply chose to be consistent with the use of 4 lags in the quarterly data.
Poss,
When you catch him – will you sell tickets?
“Andrew Bolt, Possum’s refutation of Bolt was hardly “professional” (according to your assessment), given how many unnecessary use of expletives and other offensive language. Obscene I tell you.”
If you can’t attack the argument, complain about the language.
Nicely done, P.
Possum,
A beautifully constructed and delivered slow execution of Bolt’s pathetic and ignorant position.
Well done!
LO,
You raise a really interesting point that caught my eye as well when I started playing around with this.
That paradox between the data sets seems to be mostly as a result of the arbitrariness of the delineation points of the yearly data.
With the yearly data effectively being a group of four consecutive quarters, but where those four consecutive quarters are arbitrarily aggregated into a thing called the “year”, any dynamics that operate over any period less than 1 year become unmeasurable, and any yearly patterns that transcend that Dec-Jan delineation period become washed out.
Because we take a rolling look in the rear view mirror for what happened in the year previous to any quarter with the quarterly data, we pick up on a larger array of dynamics – obviously.
The yearly data also has some pretty amazing compositional issues. Most of the applications come from Western Europe, and up until 1999, the application numbers from eastern and western Europe tracked each other perfectly – but from 2000 onwards they went in completely opposite directions and have acted almost independently since – on the yearly data. The broad “others” group – the global numbers minus total european minus Australia and NZ – have asylum application numbers that look, again, literally independent compared to Western European numbers with the yearly data.
But with the quarterly data, granger causality is still flying around between all of those groups.
Most of that ‘Oz leading the world with the yearly data’ thing seems to be an artifact of where we draw the lines that start and end a year rather than anything else – but I might have a ferret around and look into it more.
What I suspect is happening is that when some event occurs that creates a bunch of refugees that move across the world seeking asylum, the transition period between the event and the actual asylum applications of that group being lodged in other countries differs by a quarter or two among those countries, and where that parcel of refugee movement may last for a year or two. So when that transition period for both the start and the end of that refugee surge crosses the Dec-Jan period of any year, the yearly data just doesn’t capture what is actually going on.
It would certainly explain why the yearly data at the regional level of composition acts with little broad correlation in most cases, but the quarterly data at the same level is filled to the brim with granger.
GP has been trained well – first reaction (actually multiple reactions) is to attempt to make Possum’s language the issue rather than the conclusions of a well constructed argument.
Nice one.
Why do I feel like I’ve woken up back in my 1991 2nd year econometrics lecture?
(furiously scribbling notes he knows he doesn’t understand… heteroscedasicity he hears… sounds like a made up word, but writes it down anyway…? )
filled to the brim with granger? is that the statistical equivalent of a brimful of asha?
Perhaps the data should be seasonally adjusted, in the truest sense of the word. Prevailing weather conditions could affect the sailing of boats across the Timor Sea, sailing across the Mediterranean Sea, driving container loads of refugees across Europe, walking across the US-Mexico border.
Possum, “pulling someone’s arm off and beating them to death with their own severed limb” best sums up how I feel about the garbage Bolt and the Boltoids have posted on this topic. I don’t like to use the ‘F’ word on line but find I can’t reply to their small minded ramblings without using it out of frustation. My family origins come from people arriving here in the mid 1800’s, my Great great grandfather and his brother arrived here from Scotland and used their skills as Master Mariners to open up coastal shipping along Queensland and northern NSW. My other great grandfather on my fathers side escaped British oppression in southern Ireland to try his luck on the Australian goldfields. On my mothers side my great great grandfather and his extended family left everything in Ireland and dissapeared overnight to escape the British. His son, my great grandfather made significant contributions to the Toowoomba area and it is said that he kept a few Darling Downs bookies honest. My white celtic catholic forebearers certainly had little respect for british laws or titles and as such would have been a bit different in colonial life. Therefore I come from economic/political refugee and skilled migrant stock, they all arrived on boats and I am proud of it. BTW, why did Bolt’s school teacher father pack up his family and emigrate to Australia. The only inhabitants of Island who had comething to fear from boatpeople were the original ones when three boatloads arrived in the Sydney area in 1788.
Fuck me, that was beautiful
Brilliant post. Brilliant.
My only criticism of your piece Poss, lies not with the argument which you so coherently construct, but its conclusion!
Why should Bolt be let off with so compact a device? And one which he clearly cannot fathom the use of?
No, he should be treated with an abacus! Much more fitting, (or, er, ahem, not!) don’t you think?
G’day Possum,
Great Post.
I said in my Agmates piece (thanks for the link) that any Politicians (especially the Liberals) looking for a political opportunity by stirring the ugly Australian Xenophobic monster is contemptible.
You have gone the next step by lining up Bolt. ‘The Undescended Testicle of the Internetz – I love it.’
Had always thought he was a bigot, until I saw him recently on ABC Q&A. I said after that performance that he had about as much credibility as a radio ’shock Jock’.
That he has a large readership is cringe worthy and disturbing.
Hey as an aside is your stuff on Twitter (I know Crikey is)? If you are I’d like to follow your posts.
Would it upset the cheer-squad to ask Polly whether he thinks border policies have any impact at all on Australian asylum applications?
What about the regional origins of asylum seekers? I don’t think an increase in, let’s say, Bosnian emigration would have notable impact on the Australian intake – but an increase in Timorese would.
I hope I haven’t overstepped the boundary here. I wouldn’t want to be told to “shut the fuck up” and be referred to as a race baiting, poisoned dwarf, small minded, undescended testicle bigot. That’s the open-minded, tolerant left for you.
I guess the AFP are in Bolt’s pocket by the looks of it too. Racist, innumerate, fools!
@Pollytics
There’s also a link up top somewhere
Would it upset the cheer-squad to ask Polly whether he thinks border policies have any impact at all on Australian asylum applications?
Patrick – I can only speak for myself, but your question doesn’t upset me at all. I believe it has been comprehensively answered in the above piece, but I doubt that many readers here would shirk from any honest discussion of asylum seekers. The key word is ‘honest’.
The only remaining implements in the toolbox of the right-wing side of politics in the USA and Australia are racism, bigotry, xenophobia, fear and dishonesty. They have nothing else.
The previous Bush/Howard era trashed their credibility on most things and left us with this hollow casing that cracks are revealing to be a cesspit of old tired tricks.
All the right-wing side of politics and their faithful zombie followers have left are a hand full of dog whistles. And faithfully they will roll out each whistle in turn to see if it the prick the ears of a few inattentive Australians. And they wont give up, they will whistle and mislead all the way. They have no choice, they have NOTHING else except these tools of the sick.
Oh where are the years of overt bigotry and racism and hate gone. Oh woe is the Liberal Party.
An American commentator hit the nail on the head when he referred to the behavior of all these people as the delirium and lashing out of a dying of syphilitic king.
I have been waiting for this post, Scott, and I am glad to say that you did not disappoint.
Presumably you have looked at total asylum seeker applications here. What I’m interested in is the comparative numbers of asylum applications by arrival method. You point out that the number of applications in Australia per year is very low, a few thousand. What proportion of that is made up by illegal boat arrivals, as opposed to, say, people arriving by plane, or even people over-staying their visas.
I imagine the numbers would be so small as to be of poor statistical value. But since illegal boat arrivals appear to be the major problem for these right-wing dirt bags, it would be interesting to see just how ridiculous the numbers we’re talking about here actually are.
GP #2: Good point.
Possum #6: Good point.
How many times have i told you young poss
Dont play with the neanderthals
seriously good post tho
You know you’re really hot when you’re angry.
Love it!
Possum,
You’re gonna make yourself and Crikey famous* if you keep this up.
Bolt has shot his bolt prematurely over this one.
Great work as usual. Thanks.
___
*more famous
Pages: [1] 2 » Show All