UK General Election 2015: Forecasting & Influencing the Outcome
Mark Twain popularised the term “There are three kinds of lies: lies, damned lies, and statistics” and attributed it to our former British Prime Minister Benjamin Disraeli back in the 19th century.
It’s arguable that at no other time in history has polling yielded so much influence and importance since the method became popular in the 1930s. With the frequency and disparity of polling being published in the lead up to the UK General Election, here is a check list to consider when evaluating them:
Popularity v Electoral Seats
Many polls hitting the headlines are publicising the intended voting patterns of a representative sample of the entire UK population. These are interesting to monitor the trended movements after key events, but this can have little relevance when transposing these percentages into the expected number of electoral seats for that political party.
For example, in Scotland our polling shows there is a concentration of voters in many electoral seats, such that the 4% that SnP is polling nationally could result in up to 55 MPs being elected. Conversely, UKIP are attracting three times as many supporters across the UK, but their supporters are geographically more widespread and likely to only yield a handful of MPs. In summary, it’s not how many voters nationally; it’s where they’re clustered.