Hard News: Laying Down the Law
79 Responses
First ←Older Page 1 2 3 4 Newer→ Last
-
I wonder if the Herald will be concerned about Key trying to get journalists fired?
-
It's nice that Mr McCully has noticed that I join the fun at the Hero Debate. I've only been doing it for three years.
So you now admit you're a long-time supporter of Labour ... the plot thickens etc.
To be fair to McCully - you supporting Labour via the Hero Debate isn't news until you get something in return. Donor loans money to the British Labour Party - who cares? Donor receives peerage - news.
Lord Russel of Brown, Honorary Consul to the Interweb.... hmmm, it has a certain ring to it don't you think!!
-
I guess dropping a small thermonuclear bomb on Tauranga would be beyond the pale?
I could really get on board with this, except for that whole 'people who live in WTF electorates shouldn't throw nukes' thing.
-
The one think I'll never forgive Margaret Wilson for is splitting the anti-Winston vote just enough in '99, that He Who Must Not Be Named beat Katherine O'Regan by sixty three votes.
I'm not so sure if Wilson is as much to blame as any one of about 50 retirement villages. I would go with the napalm to be honest.
-
Right, I am sooo out of touch with kayway culture. "ELV" means "Expendable Launch Vehicle", right? Could someone tell me whether the sacred ritual of Rhugbi is performed with a racquet or a bat?
The Herald's business pages don't necessarily bother with facts when it comes to climate science.
Well, I've always thought it indicative of the intellectual tone of the Dompost that the horoscopes are in the business section.
I wonder if journalists ever wonder why they are despised almost as much as politicians whenever polls are taken on the relative esteem of various 'professions'? It's pompous crap like the Harold's recent edict that tends to reinforce this impression, shurely.
-
Russel Watkins was the Libz candidate for Tauranga last election. What's he done...
heh.
-
Just for fun I have calculated the confidence levels for the figures Russell quotes from the Herald. There's a 99% chance that the actual percentage of New Zealand First supporters who wish NZ First to go into coalition with National is between 64.5% and 115% (!). For ACT and United Future its between 42% and 156%, and for the Maori Party between -2.6% and 92.5%.
These statistics are based on the extremely dubious assumption that the people surveyed provide a representative snapshot of the range of opinions of the supporters of the four parties. The stats are, of course, meaningless as Russell and others point out. It's not simply a case of OVERinterpretation here - rather, it's a case of basic innumeracy. You can't conclude anything at all from the survey about coalition support.
BTW - for the purposes of arriving at the ACT and United Future figures I had to use the figure of 99% support for coalescing with National rather than 100% - the stats can't be calculated using 100%.
-
For the record, here's the Tauranga result from 1999:
BALDOCK, Larry Future New Zealand 1456 4.31%
BERRY-EVANS, Vivienne NMP 24 0.07%
GROVER, Frank Christian Heritage Party 738 2.18%
HEPBURN, John Independent 73 0.22%
O'REGAN, Katherine National Party 10162 30.08%
PETERS, Winston New Zealand First Party 10225 30.27%
SUMMERHAYS, Karen Green Party 1131 3.35%
WICKS, Tekarehana Alliance 453 1.34%
WILSON, Margaret Labour Party 9519 28.18%So Wilson is no more guilty of vote-splitting than O'Regan. You could just as well argue that in 1999 people were voting against National, and if Wilson had not stood, half her votes would have gone to Peters. Who knows? Maybe it's all Larry Baldock's fault.
But for real (and endless) arguments about vote-splitting, and who let who in, see "first past the post" elections, a system we're well rid of.
-
Paul Rowe,
Yes I'm talking Tony Ryall. One of those family values types...
When he's not in Wellington -
Just for fun I have calculated the confidence levels for the figures Russell quotes from the Herald. There's a 99% chance that the actual percentage of New Zealand First supporters who wish NZ First to go into coalition with National is between 64.5% and 115% (!). For ACT and United Future its between 42% and 156%, and for the Maori Party between -2.6% and 92.5%.
These statistics are based on the extremely dubious assumption that the people surveyed provide a representative snapshot of the range of opinions of the supporters of the four parties. The stats are, of course, meaningless as Russell and others point out. It's not simply a case of OVERinterpretation here - rather, it's a case of basic innumeracy.
There's a high level of irony in a post critiquing someone else's statistics, while making the claim that 115%, 156%, and -2.6% (!) of a group of people want something.
Clearly this proves that not only can statistics say anything, they can also say nothing at all.
-
I wonder if the Herald will be concerned about Key trying to get journalists fired?
Well, Snowy, I suspect if The Herald did want to follow up Cullen's allegation (faithfully repeated by The Standard) they'd have a hard time finding a source willing to go on the record. Funny how often that happens, doesn't it? But I guess on the record attribution is so last millennium.
And I'd respectfully suggest that I've heard some interesting gossip about certain Minister's offices who've threatened to "make life difficult" for insufficiently deferential hacks. Any sensible hack or editor, in my experience, ignores them.
But for real (and endless) arguments about vote-splitting, and who let who in, see "first past the post" elections, a system we're well rid of.
simon -- simple reality is that if a hundred or so people had voted for O'Regan over Wilson (while still casting their party votes for Labour or anyone other than National), Winston Peters and his squalid personality cult would have been gone. That's something we would be well rid of.
-
There's a high level of irony in a post critiquing someone else's statistics, while making the claim that 115%, 156%, and -2.6% (!) of a group of people want something.
That the error bars cross the limits of reason just shows the sample size wasn't big enough to be useful. 0.1% +-4% is meaningless, other than showing the result to be very likely somewhere under 4%, assuming the sample set is random, which it isn't.
Really, these tiny samples in the MMP environment are all a bit of a joke, or would be if the media didn't promote it as being so accurate to the public all the time. They barely even mention error margins these days, and have never shown them clearly in the numbers. -
I do feel that when Labour gets back in after the failure of the Key government, it'll be time for utu against the Herald and Dom Post (not to mention the Press). [...] They should propose this pre-election, and have a quiet word with the owners of APN and Fairfax as to what will happen to their long-term investment if they persist in being propaganda sheets for the National Party.
Rich: You're extracting the urine, aren't you?
-
Y'know, I'm usually such a politics junkie, but the heads-up on those new Velvet Underground tracks has thrown everything else clean out of my head...
Awesome stuff!
So awesome in fact that after I downloaded them I wrote a busy little blog post all about my Velvets love, and then trucked on over to eBay to check and see how much those 100 green vinyl copies had actually sold for... and lo and behold there were a couple still for sale (or maybe a new batch if I'm being cynical).
So I bought one. For twelve quid, how can you say no?
w00t! Thanks Russell!
-
Must be some pretty funky maths that gives statistics under 0% or over 100%. I think we can be 100% certain that the real values will not be outside that range.
-
Hey Russell, How's the TVNZ7/Hard News/Public Address relationship to be handled - in a way you're both public broadcasters...
Could you f'ristance sign off with "By the way, you might of noticed a Murray McCully press release regarding this show - I blogged about it on Public Address - a search for 'sloppy politicking' or 'malicious falsehood' if you'd like a look at it...
-
On the spam topic, Gmail has a useful way to determine who is spamming you: sign up (e.g. to an ACT mailing) as yourname+ACT@gmail.com. It will still get delivered to your inbox, but if they start spamming you, just add a rule that sends all emails to that address to your junk folder.
Now I must go listen to that VU; downloaded it after seeing it on Metafilter a few days ago, still haven't got round to unzipping and playing it.
-
Must be some pretty funky maths that gives statistics under 0% or over 100%. I think we can be 100% certain that the real values will not be outside that range.
I think thats the whole point of providing the funky numbers..... to show how poor the original data is...
-
Kyle:
Clearly this proves that not only can statistics say anything, they can also say nothing at all.
Precisely.
-
I think thats the whole point of providing the funky numbers..... to show how poor the original data is...
mmmm....I'd still be surprised how you could conclude that support is negative, no matter how poor your data. But I know it's just a joke anyway, and merely saying that the standard deviation in the data is extremely wide on the sample given.
-
That was my point exactly - the formulae I used are those used by the research companies. They basically shouldn't be used with small sample sizes. The reason some of the confidence limits (lower and upper percentages) extend past 0 and 100 is that the 'margin of error' calculations are phenomenonally large when using tiny samples.
There's nothing wrong with the maths - it's the research design that's the problem.
-
Rich: You're extracting the urine, aren't you?
Not entirely. The Herald (and Dom Post, Press, etc) are monopolies. They don't have a "dominant market position" or a "complex monopoly" - they are the only choice if you want a paper that covers the city you live in.
I think there's been a tacit acceptance that they're allowed to get away with this in return for some degree of political impartiality (even if this has always been mythical).
If we are now going to have a monopoly paper that acts as a propaganda sheet for the National Party, shouldn't something be done about it?
-
There's nothing wrong with the maths
Can I agree to disagree? Any maths that generates a negative possibility is just...broken.
-
No Ben, you can't really disagree with established math any more than you can disagree with gravity or evolution.
You're just looking at it the wrong way. This isn't some sort of distribution curve, it's just a couple of limits. It doesn't override the other limits, like the natural minimum of 0.000001% (the guy who answered for UF) and maximum of 99.9999% (all the people who didn't answer at all, plus that one guy).
-
Tussock, the minimum of zero and maximum of 100% are mathematical too, and you can't override those. I get that a lot of approximation formulas will sometimes do that, but that's because they are not 'pure'. They are approximations.
Post your response…
This topic is closed.