Written By:
James Henderson - Date published:
10:08 am, May 2nd, 2012 - 14 comments
Categories: polls -
Tags:
OK. I’m officially confused. Is National plummeting or holding? Is Labour gaining or falling? Are the Greens becoming the second party or slipping back? 5 polls: 5 different answers.
The best poll is Left 48% vs Right 44.5%. The worst: 41% vs 52%.
Either the public opinion is in extraordinary flux, we’re having a rogue concentration of rogue polls in both directions, the polls are broken, or the Gods are toying with us.
The events are so big that everybody is reacting. They’ve made reef fish of us all! Back to the reef for the latest…
Greens becoming the second major party, that’s hilarious
Not sure how you see that from the results, as there is no combination where the green bar is higher than the red.
Yes, Mighty Kites, though not so much “hilarious” as something to smile happily about. Do you keep up with their newsletters and policy statements? This is the one alternative that could be viable. But the Greens will likely appeal only to the intelligent, which would exclude rather a lot of voters in this country. (They even bother about “ethics” in politics!).
My view,there is the appearance of the polling organizations messing with the polls using ”the margin of error” to skew what gets broadcast to the public,
Why would NZFirst,for instance,have suddenly lost half its 2011 election support,or for that matter the Greens suddenly suffer a radical drop in support from November 2011,
All these media polls have a margin of error + or – of at least 3% so its that easy to manipulate the polls,
NZFirst with an election result over 6% can simply be reduced to half such support in these polls by applying to that Party a % of support from the minus side of the margin of error,
I watched NZFirst quite closely throughout 2011 in terms of where they were at in polled %s and on one particular web based polling site the manipulation downward of NZFirst % of support was quite apparent and ongoing throughout the year,
Why would ‘they’ go to such lengths to manipulate such polls, simply because people DO read and watch them and some of those people DO make voting decisions based upon such polling,so, if you KNOW that by publishing X as a poll result for A particular party you will get Y as a result and you have a vested interest why wouldn’t you do so,
Doing a bit of polling itself the Electoral Commission has come up with some interesting facts on this, another 6% of registered voters decided to NOT vote in the 2011 election,
When asked 2% of those voters stated that they had not voted in the 2011 election because of the ”done deal nature” of the election and that poll results had helped form their impression that the 2011 election was a ”done deal” so they didn’t bother to vote,
My view,considering the nature of such polls ability to in effect attack the very democracy that they purport to be measuring such publishing of polls should be banned for at least a year out from an election,
The political party’s do their own polling to see how they are faring in the electorate so the media polls would not be missed,
‘I see a clinic full of cynics, trying to twist the peoples wrist, they watch every move we make, we are all included on their list’…
You appear to be suffering fron delusions about what the polling companies report.
The percentages they give for a party are exactly the numbers they get from their sample, nothing more, nothing less.
The margin of error is a measurement that gives an estimate of the amount, at a 95% confidence level, that the population, ie all voters, figure might be expected to differ from the sample that they are reporting.
If 90 people in a thousand sized sample say they would vote for the Green party they give the result as 9% with a margin of error of about 3%.
They are saying that they expect the real, population, value to be between 6% and 12% with a probability of 95%.
They do NOT fiddle the results in the way you suggest and give a result for the Greens of 6% (or 12%) depending upon their political bias.
I’ve often thought it would be better if the media reported the bounds of the 95% CI rather than the sample mean. Saying “the Greens are on 9%” gives a false sense of precision and certainty, especially when the margin of error is only briefly mumbled at the end and ignored by most viewers. It would be closer to the true meaning of the poll to report “the Greens are somewhere between 6% and 12%”, which actually conveys the uncertainty in a way casual viewers would take in.
“As flies to wanton boys are we to the Gods. They kill us for their sport.” Gloucester in Lear.
Polls are in fact broken. As a country we’re over-polled and under-educated about political issues. This creates huge problems for a functioning democracy…
The polls will systematically over report and under report their metrics compared to one another.
Someone with a bit of a statistics head can get into it and understand a bit better what the underlying truth probably is.
There are several problems with polls that simply cannot be fixed. Every person called can choose whether or not to participate, so self-selection bias is introduced. Those with strong political views are probably more likely to want to give their opinion, while those who only express moderate or no preference will not care for being polled as much.
On top of that the polling companies don’t seem to go to a lot of effort to actually seek out a representative sample of the population at large. As Bomber has been saying for quite some time, the poor are under-represented in the land-line holding population, as are young people.
The combination of these plus other factors makes landline polling an absolute farce, and they should be largely ignored in my opinion.
The public arent fooled by Labour side shows what have they really got to offer in the way of policy. Nothing new the new labour really does look like old labour in drag
Jturd Key slippery’s teflon is not going to stop his slide down the greasy poll’s