LinkBack URL
About LinkBacksThat may be what you wish but that does not make it true. SF are moving forwards.Originally Posted by Cork1234
That may be what you wish but that does not make it true. SF are moving forwards.
RIRA not in my name-Traitors to Ireland MMcGuinness; People are entitled to cultural & social equality MLMcDonald; We have a length to go understanding unionism GAdams
Neither does Eurochickan.
You can even delete the "internal FG" phrase and it'll still be true.
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
I'd like a bit of clarification on this point, if some of the pollsters could enlighten me, I'd be very grateful. I've just read a tweet saying that all of the changes in this poll are within the 'margin of error', and can be disregarded as such. I'm not entirely sure if that means anything, because ....
During the course of the UK general election just gone, I heard a TV commentator, can't remember who, say that margin of error with regard to polls is effectively meaningless - can't remember the specifics of the argument. It was something to do with that individual polls should be compared to similar polls by the same pollsters.
I'd like to know if there is any truth to this argument, or what it actually is.
My search of statistical websites has left me none the wiser, but my basic questions are:
Does margin of error actually mean anything in political polls?
Surely we should be looking at trends instead?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
It is frustrating that Eurocitizen does not know SF's score.
RIRA not in my name-Traitors to Ireland MMcGuinness; People are entitled to cultural & social equality MLMcDonald; We have a length to go understanding unionism GAdams
Forwards just not up
Fixed that for you.
A poster of some consequence...
Well, simply put, polling is not an exact science. They accept that. They allow for some latitude in accuracy, plus and minus, in each sampling, and the latitude, in percentage terms, depends on the sample size. So, in this sampling of 1000 pax., and I don't have the exact theorem to hand, I think they allow the margin of error to be 3%, plus or minus.
Therefore, if any party is up or down, from their last poll, if that poll used the same sample and put the same question, to a maximum of 3%, technically the poll can be disregarded as being "within the margin of error".
I am sure I did not explain that very well!!
"......... we must sometimes listen to those who, consumed with zeal, have scant judgment or balance. To such ones the modern world is nothing but betrayal and ruin.........We feel bound to disagree with these prophets of doom who are forever forecasting calamity -- as though the world's end were imminent."
This is what (I think) it might say:
FF 25%
FG 30% (the drop reflecting concern that the childrens-referendum is actually a ploy to frustrate deportations of bogus asylum seekers).
Labour 18% (the public-sector vote outweighing that concern)
SF 10% (as a protest-party and reflecting concerns about the role of the bureaucrats in Brussels in causing the recession via pro-cyclical interest-rates that overheated the economy.)
Greens 2%
Ind 15% (reflecting the failure of the entire political-establishment in these recessionary times).
By your logic, CM, why do we not just dispense with the rigmarole of elections at all? Why not do it the FF way and just annoint your successor without bothering with elections (and in particular, by-elections?). And please don't come back with the 'they're all the same' nonsense, leave that for the likes of EC.
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
Well, simply put, polling is not an exact science. They accept that. They allow for some latitude in accuracy, plus and minus, in each sampling, and the latitude, in percentage terms, depends on the sample size. So, in this sampling of 1000 pax., and I don't have the exact theorem to hand, I think they allow the margin of error to be 3%, plus or minus.
Therefore, if any party is up or down, from their last poll, if that poll used the same sample and put the same question, to a maximum of 3%, technically the poll can be disregarded as being "within the margin of error".
I am sure I did not explain that very well!!
Thank you, that does make sense, and you did explain it quite well. So, we can effectively disregard the changes in this poll, as they all fall within the margin of error, and probably most polls, for that matter too.
However, the day on which a party avoids making political capital because of statistical rigour is probably a long way off ...
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
Reply With Quote