“In the end the debate between online and phone polls, and different methodologies, proved irrelevant. ”
Much has been, and still continues to be written, about how the pollsters could have been so off the mark in the recent elections. The article below is a prime example, pondering on the relative merits of online versus telephone polling.
But what it and many similar pieces fail to do is to question to what extent the nation should be quite so reliant on one specific methodology and even research itself to give totally accurate predictions – effectively to tell the future. David Ogilvy once said “People use statistics as a drunk uses a lamppost — for support rather than illumination.”
At Muse we have a model which enables us to forecast the number of visitors likely to attend a particular cultural event.
To find out more, visit:
www.musestrategy.com/modeling and evaluation.
It takes into account a large number of contextual factors. It is used by a many cultural organisations around the country, which is testimony to its effectiveness. But the important thing is that it is used by our clients as an aid to judgement when business planning, not as a sole predictor of events.
So perhaps in this instance the political pundits could look to the cultural sector to learn some best practice.
And in truth we understand that the polls got it so wrong because they used the same methodology as in previous elections when voters choices were wider than in previous elections and so different methods and samples were needed: we read a number of articles from academic polling experts who predicted the polls would be wrong before the election ever took place.
