shovell.com

I don’t, I think I know what I am going to vote for but I don’t profess to fully understand it.  I certainly don’t know what the outcome of any future election using the Alternative vote (AV) is going to be and I don’t believe that any so called experts on either the Yes or No side do either.

While I am at it, I personally struggle with a system which could get the next best alternative elected, I vote for the person or party that is most closely in line with my views and values as I feel they will represent me the best, how then can I nominate someone who I don’t feel strongly enough about in the first place as my runners up prize?

My problem is that I still don’t know if our current system is fair.

So putting aside the fundamental arguments about fairness of either system, I want to propose something different in the process for the referendum.

We live in an age where data is readily available, and when it isn’t we run tests, so online it’s all about the metrics;  Multi Variate Testing, Web Analytic’s, Transactional Data, Segmentation, Split tests, Sample groups and control groups, with new products and marketing concepts we can test them until we understand them taking a lot of the risk out of things.

But in something really big, like this referendum, where the outcome can have a massive influence over our lives we have to work with opinion and expert comments but the risk is still there.

So what I propose is this, for the next General Election, and it needs to be something big like that where the turnout is the highest and people take a real interest, we should run an AV experiment.

It works like this, people go to the Polling station as usual, and fill in an AV polling card in the correct way.  votes are counted but the actual election votes are done on a first past the post system (no changes to the expected result there).

The results are announced as usual but then we have the data.  With this data we can accurately model how the AV result would change things,  Nationally and locally, when these are published the voting public will be able to understand not just the concept of AV but also what it will mean in real terms to the result, then and only then will we be really informed enough to make a decision in the referendum.

Data can take the risks out of making decisions, and if they are big decisions then that’s got to be a good thing, so if you have the data, look at it, if you don’t set up an experiment to get it.