Hopefully you will have seen our East Belfast constituency poll last week, and as with all polls this has generated its fair share of comments and opinions. On practical point we've been asked about is how we decide what political parties to include, or not include, in the poll. With political polling there's always the argument as to where to draw the line in terms of how many political parties should be included in any poll. Already I've had questions as to why certain political parties weren't included in our East Belfast poll. Let's clear this point up - The accepted standard as advised by the professional polling organisations is to include all the parties that are currently represented in main local legislature, in this case the NI Assembly, and we did this.
All polls are snapshots. What a representative sample of people are thinking about a current issue at a particular point in time. This applies to the whole range of polling and market research services and not just political polls.
Some complain about the accuracy of polls - that e.g. political polls don't seem to be able to precisely predict the outcome of elections, even though the evidence shows that they actually do! But predicting elections isn't the main reason polls exist - they're meant to track ups and downs in opinion, and the ebb and flow of views on particular issues, in between elections, and during the build-up to elections. This provides information to the politicians and the public as to what issues are important, and not important, to the voters. This in turn informs and enhances the debate.