“Hey, let me tell you how you really want to vote”

(Definitely not my results!)

With the recent announcement by Theresa May that a snap election will be held on June 8th the UK has plunged itself into an unexpected and partially unwanted election season.  Far from the energising mood of democracy in action that sometimes accompanies a chance to voice our opinions electorally, the mood as we approach the third big national vote in two years seems to largely be a resigned shrug of the shoulders and a collective “Oh.  This again”.

Or at least it is on my social media feeds which probably isn’t the most representative source of public opinion, given the fashion for polarising bubbles that we all seem to buy into even more readily than craft beer and lumberjack beards.  But that fashion (social media, not lumberjack beards) is set to have a big influence on this election.

This election will for me, and an increasing percentage of the UK electorate, be the third social media general election.  By which I mean it is the third time where, in the run up to a general elections, a significant part of the discussion we have about it will take place on social media.  We’ll share articles and comment on them and give our opinions on campaigns and events and subtly shape each other’s thinking.  The fact that we’ll be doing this amongst a specially pre-selected group of people who mostly think very much like us (with the occasional bout of reacting very angrily to people who think marginally different things to us) is a problem that has already been well documented in a hundred other places and doesn’t need re-examining in any great depth here.

What I do think needs closer inspection is the election time trend of apps or websites that purport to help you make up your mind as to who to vote for by asking you your opinion on a range of questions.  This might take the form of a questionnaire about general political values or a “blind test” asking you to pick from a range of policies and the revealing the party that proposed them.  There is some variation in how it’s presented.  Generally there’s an option to share the results on social media, creating a flood of posts from people commenting about either their surprise or lack thereof.

In 2010 I can distinctly remember expressing a radiant optimism towards this new tool.  It seemed to be the very best of the technological revolution in action.  Our democracy and public discourse were being strengthened by the ability to efficiently collect and process information relevant to the decisions we make.  Surely the inevitable result was better collective decision making?  My optimism was almost certainly born in part, at least, from the confirmation bias that accompanied a set of results showing the party I was a member of at the time was massively outperforming their polling numbers in terms of popular support for their policies.

In 2015 I was more critical.  The questions seemed to me to be too simplistic and I wondered how well they could represent my, or anyone’s complex set of influences, values and beliefs let alone give an accurate picture of the equally intricate political parties that we were being matched with.  I still saw them as a handy guide but now hoped they served more as a starting point for further discussion rather than some kind of deterministic crystal ball of science whose advice we should all follow.

In 2017 the now familiar option appeared again.  “Click here and by completing a relatively detailed survey and telling us how important each question is to you we’ll tell you how aligned you are to each political party in a neat little representative percentage figure”.  This time I was more that critical, I was concerned.  This is potentially dangerous.  Who designed the survey questions?  How were they tested?  And what about the processing – how does the weighting of each question work?  Is it independent or relative to how important I say other things are?  Why isn’t this information easily accessible to anyone taking part?  But more than just a sceptical curiosity about the methodology and design there was another question.  Who gets this data?  In an age of Cambridge Analytica and the associated (sometimes overhyped but often very real) new methods of big data inspired targeted marketing should we be so keen to share our deepest held values and beliefs with a third party developer for the sake of a probably-inaccurate result?

In a way this actually quite small issue is representative of a much bigger social problem, one that feeds into a lot of the current trends towards the growth of populism and the retreat of empiricism as a commonly valued trait.  The technological revolution and the potential of big data have the potential to transform our civic institutions and with them our society as a whole.  Because these things are so new and still evolving, though, we have no model or template for how to best employ them to engineer outcomes in the public interest.

There are no rules to play by when we use these new tools.  We’re still catching up with our own capabilities.  It’s incredible to many of us that some people will click on any link regardless of the source, and unquestioningly accept what is written in whatever crazy, conspiracy-theory backwater of the internet they find themselves.  But we must also begin to act incredulously towards the instinct to click on a link such as these election questionnaires, volunteer deeply personal information and accept the results of some opaque assessment of our answers.

We should carry a healthy amount of suspicion and engage critically with ANYTHING that essentially markets itself as being able to tell us how we really want to vote.  It’s not that such tools can’t exist, it’s that they are operating within the realms of something so sensitive and important as to be expected to meet the highest standards of transparency and accountability.

Perhaps I was being unduly concerned in this instance.  The organisation that ran the latest quiz I described, isidewith.com, describe themselves as unaffiliated to any interest groups or political parties and a basic google of them doesn’t throw up any obvious red flags to suggest this mightn’t be true. They also provide a very basic and US-centric outline of their methodology on their website.   A lot of the questions remain unanswered though.  How do they store the data they receive and who do they share it with?  How did they design the question wording?  And how many of us really engaged with these kinds of questions before taking the quiz?  I definitely didn’t, so keen was I for the sweet sugary hit of an electronic confirmation of my progressive political beliefs.

Maybe it’s the barely breathing but still somehow clinging to life little optimist in me that, try as I might, my blackened heart and dead core can’t quite kill off but I still believe that it can be like I imagined it in 2010.  I think technology and the mountains of information that we now have at our fingertips cans serve to strengthen our democracy, enliven public debate and, despite all the current evidence to the contrary, bring us together.  The vast majority of the British body politic share very similar values and desired outcomes, they differ on their ability to form a coherent picture of the flood of often conflicting statistics and narratives that have suddenly entered their lives in a way our species has never experienced before.  Education and technology can help us organise, analyse and engage with these new inputs, all of us.  Right now, though, that’s not happening, and in the absence of that the space can potentially be filled by anyone.  It could be concerned citizens trying to put something helpful together or self-interested opportunists and we need to learn how to identify that.  Imagine, for example, a questionnaire app like the ones discussed here into which all political parties were able to input to the design of, under the careful eye of an independent panel of software developers, data scientists and political academics?  The input data would be completely anonymised before being published online for public analysis and the product itself would serve as a genuinely useful tool for engaging and informing a citizenry.  An idealistic idea perhaps but whenever we see a new quiz online claiming to help us know who to vote for perhaps that is the standard we should examine it against before we agree to take it.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s