No Nicola, the SNP did not get the support of almost 50% of the population of Scotland.

In the aftermath of the recent Scottish Parliamentary election, a comment made by re-elected incumbent First Minister, Nicola Sturgeon, reminds one of a recent post I made here emphasising the difference between INPUTS and OUTCOMES.

The comment comes from this story, where Ms Sturgeon is reported as confirming her intention to relaunch her party's campaign in favour of Scottish secession from the UK.  Fair enough; I don't think either Ms Sturgeon's or the SNP's position on Scottish independence comes as a surprise to anyone, and if she/they want to try and convince more people it's a good idea they should be perfectly free to do so.

But don't go about it by spreading blatant lies.

The lie to which I refer is this howler:

"When asked if Scotland can put an independence referendum to bed for the next five years, Ms Sturgeon said: 'No, the position I put forward in the SNP manifesto got the support of almost 50% of the population.'"

This is demonstrably false and so far off the mark as to be laughable.

The position put forward in the SNP manifesto did not get the support of almost 50% of the population.

The population of Scotland, according to the 2011 census, is 5,313,600.

The SNP received 1,059,897 constituency votes in the 2016 Scottish Parliamentary election.

1,059,897 divided by 5,313,600 is not even 20% of the population of Scotland.

Of course, not all of those 5,313,600 people are eligible to vote.  Many of them are children, for example.  The BBC reports the electorate of Scotland, for this election, as 4,099,407.

1,059,897 divided by 4,099,407 is 25.9% of the electorate.

I think what Ms Sturgeon meant to say, was that the position put forward in the SNP manifesto got the support of almost 50% of those who voted.

This is, at least, a more reasonable interpretation of events.  Turnout for this election was around 56% of the electorate.  There were a total of 2,279,153 valid constituency votes and 2,285,752 valid regional votes cast.

Of these, the SNP won 1,059,897 constituency votes and 953,987 regional votes, or 46.5% and 41.7% respectively.

However, there is an issue even with this statement.  That is that people vote for a particular candidate or political party for a wide variety of different reasons and considering a broad range of issues.  It's not possible to know for certain how many of those 1,059,897 people support Ms Sturgeons position on a second referendum.  Maybe some of them voted for SNP candidates due to their position on healthcare, or education or taxes, or defence, or any number of other issues or combinations of issues, or because they have a particularly good local MSP.

Pick any random person off the street and any random political party's manifesto and you are likely to be able to find some things in it which they agree with and some things in it which they disagree with.  Just because a person has voted for a particular party doesn't mean they support everything in that party's manifesto.  All it means is that, on balance, they support more of that party's stated aims, objectives and values, than that of the other parties/candidates.  Often times it's a case of picking the lesser of several evils.

In this particular case, I think it's pretty likely that most people backing the SNP are in favour of Scottish independence, but it is simply not possible to know that from the number of votes alone.

To determine that you have to have a referendum.  Which we did, in 2014.  And in which we found that 1,617,989 people, or less than one third of the population of Scotland, were in favour of independence at that time (to use Ms Sturgeon's favoured way of expressing these statistics).

If she hasn't already, I think Ms Sturgeon should be obliged to make a full retraction of this statement and go on record to correct any misunderstanding that may have resulted.

The General Election & Electoral Reform

I had been expecting an interesting election and an interesting election is what we got.  For the first time since 1974, the second time since the end of WWII, we have a hung parliament.  We still don't know for certain who will be the next Prime Minister [EDIT: We do now, David Cameron has been confirmed as PM, with Nick Clegg Deputy PM in a Con-LD coalition government], all we know at this stage is that Gordon Brown is stepping down from the post this year.

Anthony Wells, over on UK Polling Report, has a thorough breakdown of general election statistics, both here and here.

The most unusual statistic perhaps being that whilst the rest of the UK saw large swings from Labour to the Conservatives, Scotland showed a small swing from the Conservatives to Labour, precisely the opposite direction from the rest of the country.  Despite this though, not a single Scottish seat changed hands, the results were so close as to make no difference to those of the previous election in 2005.  I had expected the Tories to do less well here than in the rest of the UK, but I had expected at least a few of the more marginal seats to change hands, if not to the Tories then to the Liberal Democrats.

Given the current situation we're in, I think now is an appropriate time to share my thoughts on electoral reform.

I was about to start writing about how our current voting system is unfairly biased towards the two largest political parties and cite the fact that the Lib-Dems have less than 9% of the seats in the Commons, despite receiving 23% of the popular vote*, when I came across this piece from Tim Harford, who explains that 8% of the seats is not the same thing as 8% of the influence in the Commons and in the current situation the Lib-Dems have as much voting influence as Labour.

Tim is right, the maths does back him up - commentators are wrong to suggest that the outcome of this election was a bad one for the Liberal Democrats.  However, this is but one election and the Lib-Dems are but one party of many.  The fact that they hold more power than seats in this circumstance is surely not the point though?  (Tim acknowledges this fact in an update where he also points out that many people have missed the point he is making.)

Peter Henley has written here about the difference which would be made to the outcome of this election were it conducted under the Alternative Vote (AV) or Single Transferable Vote (STV) systems, as opposed to our current First Past The Post (FPTP) system.  However, these numbers will involve some pretty big assumptions, so can't be taken as a reliable prediction.  I do agree with the general conclusion made by Peter though, that there would be less of a difference from the existing system with AV than with STV.  This should be pretty obvious, since as a voting system AV resembles a mixture of both FPTP and STV.

First Past The Post (FPTP)

The problem with FPTP is that a party can gain a lot of public support and win a lot of votes, but due either to geography or gerrymandering can win relatively few, if any, seats.  The FPTP system favours Labour, whose voters tend to be concentrated in urban areas, at the expense of smaller parties and parties with less geographically concentrated supporters.

I have only heard two arguments put forward in favour of the FPTP system, one of which isn't even a proper argument:

1. "We do it this way because we've always done it this way."
2. FPTP gives you a representative local MP, something which you don't get with proportional representation.  (i.e. Parliament taken as a whole may be unrepresentative of what the nation wants, but at least each individual MP is representative of what their own constituents want.)

As I've mentioned the first of these is not even an argument, just because something is tradition or has always been done a particular way is not a good reason to continue, or to not seek to improve it in some way.  The second point has more substance to it, but even then there are systems of proportional representation which can go most of the way to addressing this issue.

Alternative Vote (AV)

This is the system promised by Labour in their latest manifesto and the system the Tories have discussed implementing to gain the support of the Lib-Dems in Westminster.

In an Alternative Vote system, also known as Instant Runoff Voting (IRV) instead of putting an 'x' in the box next to your preferred candidate you simply rank the candidates by preference (put a '1' in the box next to your favourite candidate, '2' next to your second favourite, etc.)  This is simple for the voter and the system for counting votes has the advantage of being simple too.

The advantage of AV over FPTP, and the point which is generally used to argue in it's favour, is that it minimises "wasted" votes.

The way it works is very simple:

• In each constituency an absolute majority (i.e. more than 50% of the votes) is required to win the seat.
• If no candidate has an absolute majority the candidate with the least votes is eliminated and their votes are re-allocated based on their next preference.

This process continues until one candidate has an absolute majority and is declared the winner of that seat.

Under AV it seems likely that more people would vote for who they actually wanted to win (i.e. less likely to vote tactically), since they can specify a second, third, fourth choice, etc.

Also, more people would be happy with the outcome of the vote in their own constituency, with every MP having gained an absolute majority of votes.

The disadvantage is that it still doesn't necessarily lead to a more proportional share of the seats in Westminster compared to the proportion of votes won for any particular party.  A party could secure a huge number of votes, but come a close second in the vast majority of seats, winning hardly any for themselves.

Single Transferable Vote (STV)

This is the system currently favoured by the Lib-Dems.

With STV the voter votes in exactly the same way as with AV (ranking candidates by preference).  The difference comes in the way constituencies are arranged and the way in which votes are counted.  AV is actually a special case of STV, where only one candidate is elected per constituency.

Rather than having small(ish) constituencies like we do now, with STV the country would be split into much larger constituencies or regions (maybe 5 or 10 times** bigger than our current constituencies).  Say, we combine 10 existing constituencies into one super-constituency, assuming the number of MPs stays the same we then need to elect 10 MPs from this one region.

The system for electing these MPs is the same as that used in the AV system, only instead of requiring an absolute majority each candidate requires the number of votes specified by the Droop quota in order to win a seat.

It works in a similar way to AV:

• In each constituency, any candidate who has reached or exceeded the quota is declared elected.
• If a candidate has more votes than the quota, their surplus votes are transferred to other candidates, proportional to the next preferences listed on that candidate's ballot papers.  Votes that would have gone to the winner instead go to the next preference listed on their ballot.
• If no one new meets the quota, the candidate with the fewest votes is eliminated and that candidate's votes are transferred.

This process repeats until either a winner is found for every seat or there are as many seats as remaining candidates.

Similarly to AV, people are more likely to vote for who they actually want to win and more people are likely to be happy (or at least not dismally disappointed) with the result.  The advantage over AV is, providing the constituencies or voting regions are large enough, the composition of MPs in Westminster will be proportional to the total number of votes received by each party nationally.

In case it is not apparent from my discussion of the voting systems above I am in agreement with the Lib-Dems and in favour of switching to the STV system.  If you disagree with me I'd be keen to hear from you in the comments.

* The Lib-Dems actually gained in terms of total number of votes and share of the vote, but lost seats.

** Maybe more, I use these numbers just as an example.