Body & Soul3 mins ago
Opinion Polls,anyone Been Asked?
10 Answers
Just wondered if any on AB have actually been asked to participte in an opinion poll about an election?
Also how do they choose the people to ask? I am not sure the selection of participants is random, for example if it is a phone poll , those like me who only have a mobile will never be asked.
What steps are taken to ensure that all sections of society are equally represented?
Also how do they choose the people to ask? I am not sure the selection of participants is random, for example if it is a phone poll , those like me who only have a mobile will never be asked.
What steps are taken to ensure that all sections of society are equally represented?
Answers
Best Answer
No best answer has yet been selected by EDDIE51. Once a best answer has been selected, it will be shown here.
For more on marking an answer as the "Best Answer", please visit our FAQ.I think they are stratified samples and they go to great lengths to get a representative mix of social groups. Polls often cover only around 2000 people I recall- the accuracy doesn't change much by taking samples of 10000; so I think the chances of a person being asked at any given election are pretty small given that there are, I would guess, around 30-40 million voters
telephonic polls never really recovered from wrongly predicting the Presidential election in ? 1945
Chicago Trib carried the headline "Dewey by a landslide"
and everyone knows Dewey was never president
oops
I knew that pollsters skipped various people - you dont want 70% unemployed in your poll or no OAPs because OAPs are renowned for getting out of bed and voting on polling day...
but google
opinion polls quota filling
debate I had not realised has been raging for fifty years on that one
( enough links to make your eyes glaze over )
Chicago Trib carried the headline "Dewey by a landslide"
and everyone knows Dewey was never president
oops
I knew that pollsters skipped various people - you dont want 70% unemployed in your poll or no OAPs because OAPs are renowned for getting out of bed and voting on polling day...
but google
opinion polls quota filling
debate I had not realised has been raging for fifty years on that one
( enough links to make your eyes glaze over )
// If I was asked I would give a wrong answer just to mess up their system. //
a poll of one is gonna be pretty unrepresentative - a poll of 100 is better and a poll of 1,000 even better
so the larger the poll the less the wobble ( plus or minus 3% seems common ) Jim'll fill in the technical bit - standard error of the mean I think.
Thatcher's election was wrongly predicted - by 2%
and that has happened to Netanyahu recently - too close to call and then he won a clear majority....
IN those two, the pollsters agonised ( coz they lose money if they call it wrong see ?) and put in a wobble for small poll numbers and then a wobble for quota filling and then a wobble for dont-knows not following the distribution of the do-knows
and they still didnt get the right answer on the day and concluded....
people had been intentionally misleading the pollsters
a poll of one is gonna be pretty unrepresentative - a poll of 100 is better and a poll of 1,000 even better
so the larger the poll the less the wobble ( plus or minus 3% seems common ) Jim'll fill in the technical bit - standard error of the mean I think.
Thatcher's election was wrongly predicted - by 2%
and that has happened to Netanyahu recently - too close to call and then he won a clear majority....
IN those two, the pollsters agonised ( coz they lose money if they call it wrong see ?) and put in a wobble for small poll numbers and then a wobble for quota filling and then a wobble for dont-knows not following the distribution of the do-knows
and they still didnt get the right answer on the day and concluded....
people had been intentionally misleading the pollsters
Related Questions
Sorry, we can't find any related questions. Try using the search bar at the top of the page to search for some keywords, or choose a topic and submit your own question.