On the 6th of May equimedia published its final election tracker looking at party and leader sentiment using data taken from a combination of social and intended polling surveys.
Alongside the national tracker equimedia have been running some fairly comprehensive reports at key marginal constituencies’ level. These reports compare the importance of key issues against the whole of the UK, along with voting intent.
Frankly, I have been fascinated with the insights these regular surveys have provided at a local level as it suggests that some candidates and their parties have been out of step with local concerns – but perhaps we can expect national party messages to be at odds with local concerns as local issues do tend to drive personal voting decisions. Added to this that on polling day itself the predication has been that 20 – 25% of voters will make their decision at the point of voting – it is proving tough to forecast the outcome.
And to quote the only bit of Cockney slang I know, would you “Adam and Eve it”*, there’s something missing – the views of women!
Our social tracker reviews comments made by those sharing their opinions – that’s the way it works. The flaw in this as a result predictor is that the data is heavily biased as the vast majority of comment that is shared comes from men voicing their support for their party. Perhaps this isn’t so surprising after all!
Here is a snapshot of the party social comment volume by gender gathered during this election compared to gender share of votes cast in the 2010 general election.
These findings could mean that all polls including social media citations are potentially skewed in favour of male bias.
The recent Scottish Independence referendum could offer another practical example of female voters being overlooked. With Lord Ashcroft’s exit poll showing that women were more likely to vote no than men, this may point towards why many polls showed a much narrower margin between ‘Yes and ‘No’ than the final results.
And in turn this is a big message to the world at large. Women as a group are not expressing their political opinions via social channels – they are keeping quiet. Very quiet which suggests to us that the outcome of the election when compared with the predictions will be determined by woman who have so far kept very quiet on their intent.
The outcome is not just a political one but could potentially lead to a redefining of how the pollsters collect their data based on loud opinions versus considered and personal voting preferences.
It will be interesting to track this gender voting pattern post the election.
*Adam and Eve it – Cockney slang for believe it