Voters are frequently asked about their opinions regarding certain issues or parties in various surveys, polls which give quick & somewhat comprehensive data about people's opinions.
However there are many problems as well: political polling history has shown that the polling data may vary quite a lot depending how the question is asked, people may not give an honest answer due to fear of getting marked, people like to be on winner's side and margin of error may effectively shadow the real figures & changes in them. At worst polls may give misleading, false information and effect on voting decisions.
Should polling be restricted or banned totally during elections, do we give them too much weight? Does polls affect your opinion about parties or issues at any way?
Heaven's gonna burn your eyes
16th April 2005
Interesting, but as they say, there's only one poll that counts.
We have a ban on polls in the final week before the election, and I think that works very well. As for issue polls, while they are interesting, it's not as important as a survey of the voting intentions of Parliament.
The problem with polling, and voting in general, is the amount of people that are politically uneducated, which drives me up the walls before every major vote, polls or none.
I'm way cooler than n0e (who isn't though?)
11th February 2003
I never trust polls(atleast not to certain extents) They should stay, just maybe restrict them(like what happens in Marsoe's New Zealand)
Besides they usually always ask a very small number o fpeople that they think is enough and slap on a margin of error :\
For example...do we REALLY REALLY know who will be a better democratic candidate(Obama vs Clinton) or the others for that matter? Its dumb and it does sway public opinion, because the public is generally dumb and follow whatever is on TV and beleive it.(WARNING THAT WAS MY OPINION)
In my opinion, for those who adhere to fierce party loyalties, poll numbers have a significant impact on primary elections. As you said, people like to be on a winner's side and for those who stick to a particular party, well they will vote for the person they think has the best shot at getting the party nomination (not to mention capable of beating the candidate(s) from an opposing party).
I think political polls are appropiate between elections and few weeks before them to give somewhat overall picture about the situation but margin of error should be kept in minimum, say 1.5-2 or 3 percent as absolute maximum. Political polling shouldn't be allowed 1-2 weeks before elections to minimize their negative impact on the electorate and election results. For example during 2007 Finnish parliamentary elections, a Finnish polling institute "Taloustutkimus" (trans. "Economy Research") gave some 20,4% vote share and reducing support for National Coalition Party (centre-right, liberal conservative party) at its election day poll. Actually the NCP got 22,3% vote share and while the result did fit into margin of error, the term was totally wrong as the party's support did went up a lot during couple of weeks before the election day. I find the US political polls quite problematic as the margin of error may be even 5.5 percent and the actual sample consists of just some 1 000 people in 300 million people country where as some 1 000 people sample is normal polling figure in Finland with just 5 million people.