The updated version of Vote for Policies was released last week in preparation for the 2015 General Election. Having tried the previous version only recently I decided to take the new version for a spin. As before the site aims to extract party policies from the personalities and rhetoric of the surrounding campaign and lay them out next to each other so voters can really focus on the issues that matter to them. They claim it helps make “an informed, unbiased decision about who to vote for” for those who might want to vote with their head as much as their heart. It still works by taking you through a survey that presents policies for different issues and inciting you to select your preferred set of policies. At the end the survey presents results of which parties you prefer based on your policy selections.

Taking the Survey

The new site looks great visually. It’s clear and easy to see how to get started. There are ten issues in all of which you need to select at least four to start. Being a geek I selected all ten. The next screen asks you to select your country. This is so the policies can be tailored for your region so the five most relevant parties are used. This means voters in different parts of the country will have slightly different options. This is something the TV debates could perhaps consider? For England the parties are: Conservatives, Green Party, Labour, Liberal Democrats and UKIP.

The bulk of the survey is being presented with the policies issues by issues. Up comes the first issue and then 5 policy sets will be presented. An improvement on the previous version is you first shortlist the policies which makes comparison easier. So you review a policy set then are asked to say whether you would consider voting for these policies are not. This quickly helps eliminate policies you are not interested in.

Once you have decided on each of the five policy sets then all the policy sets on the shortlist of those you may consider voting for are presented on a single screen. You can then pay more attention to the policy details and compare them against each other before selecting which policy set you prefer. Having done this the survey moves onto the next issue and repeats for as many issues as you have selected.

This all takes a while, and I’m not sure if you can stop in the middle and come back, but progress through the survey is well signposted.

Reflecting on the Survey Process

I felt this worked well and helped reduce the amount of information at the comparison stage. For me this usually meant quickly eliminating the more right wing policies leaving me with a set of more left learning policies to compare. This isn’t surprising although there were a few areas where I selected four policy sets and one where I shortlisted all five. Some issues are closer to the centre than others it seems. I found it easy to quickly eliminate one or two policy sets from my decision. I found it much harder to select between my shortlisted options. I felt there were many good ideas and policies put forward.

I still felt at times it was still too easy to identify a particular party. As before it is admirable that VfP fillets the policies from the party literature and leaves it untouched but some rhetorical styles and phrases are so distinctive the party appears through the policies limiting the unbiased nature of the exercise. Neutralising the language would make the policies more unbiased. Whilst their is editorial risk in doing so I did feel that it would be possible to ‘clean up’ some of the rhetorical flourishes to make the policy statements more stark whilst avoiding accusations of impartiality.

Smart Policies?

The other thing a focus on policy statements demonstrates is how some are more policy like than others. There is variation between SMART (Specific, Measurable, Agreed, Realistic, Timed) policy statements and vague slogans for some issues that made comparison difficult. Overall, I always erred on the side of the party who offered SMARTer policies rather than vague and slippery aspirations. This is an area where I think Green policy can improve I think and where Labour policy is very good. Still, many people would agree that the Labour Party’s problem is one of style and communication rather than substance; they would easily win an election if more attention was paid to the good work they have done in government and the good work they pledge to do beneath all the canards and red herrings town their way and the shouty froth of political mediation.

Getting the Results

The final stage is to enter your postcode, for constituency comparisons and then the results are presented. This was the first time I’d seen any pie charts and I think it works well that you have to take the survey before getting and idea of how your constituency and national results look as this helps reduce bias.

The results are really well presented. There is a clear pie chart showing your overall breakdown. My result of 40% Green, 40% Labour and 20% Liberal Democrat was pretty much unchanged from my 2010 attempt.

Underneath there is an option to enter an email address so be sent a link to your results and send a reminder on polling day. There are also the usual social media sharing options to spread the word. VfP are hoping for 5 million completed surveys this time around.

In the final section the results are broken down by issue so you can see which party you voted for on which issue. This is where some of my results were a bit surprising. There was evidence of a left leaning consensus in my selections with variation at the level of implementation detail or phrasing rather than aspiration.

Overall, and not surprisingly, I felt the Green Party were better at articulating a social democratic vision whilst Labour in particular was better at translating vaguer ambitions into something that could be implemented and might actually work. There are some sticking points, Trident being an obvious one, but I finished feeling optimistic about the prospects of a leftish coalition and wondering yet again why the Liberal Democrats are members of the current coalition.

In fact, I would probably prefer a left leaning coalition with a broader spectrum of policies than perhaps a majority to better provide a mix of head, heart and soul. Certainly I think the Green surge, and to be fair the emergence of UKIP, is widening the Overton window and that is welcome.

The results are also then shown for my constituency where 61 surveys have currently been completed and nationally where 47,000 have currently been completed. I won’t say much about these results yet to prevent influencing those yet to take the survey but the results are still looking strange. I still have my reservation from my 2010 attempt about the demographic sample of this kind of tool:

An online tool like this appeals only to a minority of voters in certain demographics. There is no demographic breakdown for the constituency or overall responses. Those advocating the tool are likely to be tech-savvy, media literate and used to obtaining their information from diverse sources. Do these really represent the core voters who turn out and decide elections?

Policy Browsing

A welcome addition is a policies browser. This allows you to go back and look at all policies again, this time knowing which policies belong to which party. This allows a more reflective look at all the policies for a party or to compare policies by issue. The browser also shows which policies you voted for. There is no login for the site which suggests this information using cookies and local storage to remember your participation. If you attempt to start the survey again then the site prompts that you have already completed and provides a link to your results (the same permalink as used when sharing). However it does also give an option to forget you, suggesting you can erase the memory of your previous attempt and have another go. Not clear if you do this whether your previous result will be erased from constituency and national results or not or whether you could complete multiple attempts all adding to the overall verdict. It is this question that makes me have reservations about using any results as campaign leverage.

Summary

Overall I think the site itself is much improved and I do hope it will be successful in introducing policies and a more thoughtful reflection on their voting preference to as many people as possible. Caveats still remain about its use as evidence.

Pros

• Clear and easy to use
• Like introduction of shortlisting stage to quickly eliminate some policies
• Result presentation much improved
• Addition of the policy browser makes it easy to compare all parties across an issue or see all policies for a party to better review/reconsider after taking the initial survey
• Don’t get to see national and constituency results until taken the survey preventing influence

Cons

• Still possibly open to gaming results so still a better tool for individuals rather than evidence of voting intention.
• Still too easy to identify policies for some parties because of their distinctive language and policies. The editorial/neutrality conundrum is not yet resolved

Vote for Policies is a site that aims to extract party policies from the personalities and rhetoric of the surrounding campaign and lay them out next to each other so voters can really focus on the issues that matter to them. They claim it helps make “an informed, unbiased decision about who to vote for” for those who might want to vote with their head as much as their heart.

How it Works

A simple survey allows voters to compare policies for issues they select. At the end of the survey the results will show which party’s policies you most agree with both overall and broken down by each issue. You can also see the overall results for your constituency.

How it is Funded

The site is run by volunteers and aims to be neutral and objective. It is funded using crowdsourcing via donations. It was created for the 2010 General Election and over 500,000 surveys have been taken so far. For the May 2015 General Election they are aiming for over 5,000,00

Trying it Out

It’s worth noting that presently the site remains the one created for the 2010 election. This means the policies are not current and the parties represented do not represent current developments of the major Scottish, Welsh and Northern Ireland parties. The site is being updating for 2015 and a new mobile friendly version with updated policies and parties is due in February.

I decided to try the current version out to see my current starting point and outlook based on the 2010 policies.

I tried out every issue and at the completion of the survey my doughnut chart revealed I was 44% Labour, 33% Green and 22% Liberal Democrat. This isn’t really surprisingly as it reflects many of the starting views I’ve outlined and accurately reflects how I voted at the last election and elections since then.  This is where I would expect to be or at least to have been in 2010.

This is not too dissimilar to my constituency picture with Labour, Green and the Liberal Democrats followed by the Conservatives and UKIP.

Which is nothing like how the constituency voted in 2010 or indeed in my living memory.

What Does this Tell Us?

It is not surprising that leftist advocates will happily point to vote for policies as a better way of deciding on your vote. The overall position consistently puts the Greens, Labour and the Liberal Democrats ahead and recently the Greens have surged. Many suggest this is indicative of the current green surge and whilst I’d like it to be so I’m not sure about using the data so uncritically. The picture for my constituency is totally at odds with what happened.

Some possible reasons for the difference and reasons to use Vote for Policies with caution include:

• voters only discovered Vote for Policies after the 2010 election and had they had tools to help them separate the policies from the personalities they may have voted differently?
• Vote for Policies is seen as interesting but mostly a bit of fun and established media and campaigns have much greater influence in voter decisions.
• The sample size is small. Vote for Policies didn’t reach that many people previously but may play a bigger role in the coming 2015 election.
• An online tool like this appeals only to a minority of voters in certain demographics. There is no demographic breakdown for the constituency or overall responses. Those advocating the tool are likely to be tech-savvy, media literate and used to obtaining their information from diverse sources. Do these really represent the core voters who turn out and decide elections?
• The algorithm for calculating scores is only simply described on the FAQ page and is quite simple. Is this kind of averaging the best way to represent choices.
• As far as I can see there is nothing to prevent people ‘gaming’ the system.
  • In some cases it is blatantly obvious which party is which. The policies are compiled without editing except to remove party names. This is admirable as editing brings its own problems but different parties have quite distinctive language. Their rhetoric is still discernible in their policies even out of context. More neutral language would make it harder to distinguish part policies by their language. Given this someone could go through and simply vote for the policies they recognise from their party rather than make an “informed, biased choice”.
  • There is nothing as far as I can see to stop people completing the survey multiple times to influence the results.

I also found it interesting to note that not everyone taking the survey is examining every issue. The issues that voters are selecting may tell us something about priorities with Health/NHS top with 404.2K votes followed by Education (392.1K) and the Economy (379.6K). Europe is least voted on (258.6K) so perhaps these agonised debates about the European Union have undue prominence. Bearing in mind the caveats above there are still little vignettes to chew over such as the Liberal Democrats leading on Democracy and reflecting on the possibilities for democratic reform they have squandered this parliament.

Going Forward

Whilst I want the Vote for Policies outcomes to be truly representative of how people would vote if they focused more on policies I am concerned that it’s not enough to hold up data that reflects your point of view and use it uncritically as a campaigning tool.  Despite the prominence given to overall results on the site I’m also not sure this is the intended use of the tool.

There are too many caveats and questions about how the data is compiled. If Vote for Policies are hoping to be more powerful this time around, or at least tolerate use of their data for campaigning, they might need to address some of these and be more transparent about their data and maybe even open it up. Certainly if multiple survey completions by the same people can’t be prevented the data isn’t that representative generally and those using it as a kind of opinion poll should do so with caution.

Where Vote for Policies works best is in the particular: as a useful tool for individuals that aids the engaged voter review, compare and contrast all policies on a particular issue in one place.

Despite my reservations I do appreciate all the work the team behind Votes for Policies are putting in to help present information differently and making it simple and clear for voters to compare parties on key issues. I am looking forward to seeing how they refresh the site and the policies over the coming weeks.

I will take the survey again once it has been revised to see how my views change and to critically appraise the update. I’ll also continue to keep an eye on the overall data and how people use both the tool and data during the campaign.