As electoral stats wonks, it’s annoyed all four of us that there isn’t a decent single point of information about Scottish election results. The broadcasters’ pages aren’t terribly usable, although the BBC’s maps are nice, the Wikipedia pages are good but not purpose-built, and dear old alba.org.uk is a tad partisan, plays annoying sound files, and doesn’t have the 2011 results.
So, we thought (OK, I thought, before I twisted my colleagues’ arms) Better Nation could attempt to fill the gap, and as a result you can see a wee addition to the navigation above. It’s been a fair bit of work over the evenings and weekends, and we should also give proper thanks to Aaron (blog, twitter), who helped process some of the tables.
And now it’s ready. More accurately, it’s in beta testing now. You will find errors (honestly, there’ll be the odd + for a -, people’s names spelt wrong, all sorts, we’re sure -Â please do tell us in the comments to this post and we’ll fix them as promptly as we can), and you may find the odd note of inadvertent partisanship.
The national results are here, regions look like this, and as a sample, here’s the First Minister’s constituency page. All the constituencies are listed alphabetically here. For vote changes and vote share changes in the new constituencies, we’re using the Denver notional results from 2007 as the baseline.
In addition to finding mistakes, please do also let us know, if you find this useful, where you think we could take it next. Obviously doing historic pages for the three previous elections is on the to-do list, but that might take until the summer. Should we add each constituency’s regional vote breakdown to those pages? Should we ask all 129 MSPs for a short biog and have a page each? Should we try and do each and every ward ahead of next year’s locals?!
Who knows. Right now we’re a bit knackered and phase two is unlikely to start this week. But we hope you find it useful. Even if you don’t, we will!
#1 by rlemkin on May 30, 2011 - 9:33 pm
Very good.. though I would recommend adding the regional vote per constituency and % difference for each party vis-a-vis constituency vote.
#2 by Malc on May 30, 2011 - 9:36 pm
That is planned… but our eyes needed a break!
#3 by Tony on May 30, 2011 - 9:43 pm
Great stuff!
Would also recomend somekind of map with general description of socio-economic types, new estates, local factors that may have affected voting trends ect. Nicholas Whyte’s excellent site is a great example;
http://www.ark.ac.uk/elections/
A lot of work to replicate I’d imagine ;¬)
#4 by douglas clark on May 30, 2011 - 9:46 pm
Excellent work! I can believe your eyes need a rest.
#5 by Lily on May 30, 2011 - 9:52 pm
It’s fantastic, thanks for pulling it together. Will be most useful. Have some thoughts about additions/tweaks but will put them all together and post tomorrow. And would love to see every ward broken down for next year’s LA elections as have been advised previously that the info is not available drilled down to that level (by electoral folks at CEC).
Also interested in comment by Tony above re mapping socio-economic/demographic change….just to add to the to-do pile.
worth waiting for!
#6 by Malc on May 30, 2011 - 9:56 pm
Thanks Lily. We’d like to do council wards, but realistically… that’s LOADS of work. We wanted to get the 2011 results up, get some feedback etc and see what we could do. Think the breakdown of the regional results by constituency might be next, but I’m as keen as you to get the Council wards done… maybe we’ll be able to enlist some help?!
#7 by rlemkin on May 30, 2011 - 10:48 pm
Perhaps just an overview of the main councils to start with?
#8 by Aidan Skinner on May 30, 2011 - 9:56 pm
I would love ward level results. I would love ward level results so much I would help with the data cleaning if you can find a source for the data.
#9 by Malc on May 30, 2011 - 9:58 pm
We like comments like this.
#10 by rlemkin on May 30, 2011 - 10:08 pm
Equally I could help with that for Aberdeen City and Shire. I feel I should offer too after all my ‘helpful’ suggestions
#11 by Jeff on May 30, 2011 - 11:15 pm
I didn’t want to have to do this, but that’s a verbal contract. Start tomorrow at 9am?
Just joking (of course), but it does look like council info is the star attraction to go for. We (James) will regroup and roll out a plan I would wager…
#12 by Indy on May 31, 2011 - 6:25 pm
I really don’t think there is any external source for data on how electors voted in 2011 organised by ward.
There was data on the 2007 parliamentary election (both constituency and regional) that could be applied to wards because the Scottish Office published a breakdown of sub-constituency results i.e. the results in each ballot box (or where ballot boxes contained only a small number of voters they combined two or more together). That was because it was an electronic count.
However this time the count was done manually. Verification statements have now been provided. But although they give the turnout for each ballot box they do not give a breakdown of the votes cast in each ballot box.
In a manual count the votes are verified by ballot box – i.e. they verify that the same number of votes that are recorded as having gone into each ballot box at the polling station comes out of the same ballot box at the count. That is where the information on turnout comes from. And it is where counting agents can take ballot box samples.
But when they proceed to the next stage of the count I’m afraid they do not continue to group the voting papers by ballot box. So the information on the way electors voted n each ballot box (from which you can work out how they voted in each polling district and from there in each ward) just isn’t there.
(Of course this is assuming that the count is conducted in the same way everywhere in Scotland – I could be wrong about that but don’t think I am).
#13 by Indy on May 31, 2011 - 2:29 pm
I am pretty sure we will not get ward level results from any source other than our own ballot box samples.
We only got sub-constituency data in 2007 because it was an electronic count so it was a completely different procedure.
#14 by holyroodpatter on May 30, 2011 - 10:07 pm
As awe inspiring as I find the effort, i think as a parliament worker, james is being a tad harsh when discussing the dearth of information. Dry though it was, I thought the spice briefing was pretty informative
#15 by Jeff on May 30, 2011 - 10:10 pm
If Spice is what I think it is, it’s not point and click but Excel based and rather unwieldy. That said, I did use Spice to build a lot of my sp11 predictor model so it is certainly useful.
Wikipedia, of course, is really quite excellent, if limited in certain aspects.
#16 by James on May 31, 2011 - 11:28 am
SPICe briefing here. It is great, but it’s hardly the same thing.
#17 by La Lynne on May 30, 2011 - 10:28 pm
Nice work. A wee, quick fix suggestion: a glossary for any party name shortened – think globally! A non-Scot might be interested…
#18 by Lily on May 30, 2011 - 10:57 pm
Would be glad to help with the local stuff. Will follow up with you.
#19 by Rob on May 30, 2011 - 11:23 pm
Great job – and much needed! The regional results by constituency, previous results and council wards in one place sounds very good. As another election stats geek am happy to help out too…
#20 by Jeff on May 30, 2011 - 11:30 pm
Thanks, much appreciated Rob. The power of people could be the motto with this project thanks to all the offers of assistance. We SHALL be in touch…!
Council wards is definitely a key next target, 2012 isn’t so far away after all…
#21 by douglas clark on May 30, 2011 - 11:43 pm
Jeff,
As a longer term commitment, and not to be taken as a promise of a 9:00am start or anything, I have my own reasons for wanting to compare the 2007 baseline with the 2011 outcomes.
I would be willing, sigh, to have a go at providing you with that. On the basis that I think something very complicated happened. And I am not sure what it was. Perhaps working my way through that data might resolve it for me.
I am willing to start from the Denver numbers and show the actual votes constituency by constituency. And the change in 2011. Because the results – in terms of seats – don’t make sense.
If you are interested in me doing that, let me know.
#22 by James on May 31, 2011 - 8:41 am
Douglas, the constituency pages do do exactly that – comparing the Denver numbers constituency by constituency with the 2011 outcomes. One of us has misunderstood the other, surely!
#23 by fifs on May 30, 2011 - 11:37 pm
http://www.scotlandvotes.com/
#24 by Caron on May 31, 2011 - 7:28 am
Wow! Every political geek in Scotland will be queueing to buy the Better Nation team a drink.
This is exactly the sort of info I’ve been looking for and it isn’t available in one place anywhere else. Like others, I’d really like to see the regional vote by constituency. Fifs, the Scotland Votes site doesn’t have anything like this sort of detail.
If you think of anything useful I could do to help with future development, let me know.
I don’t think you need to have a page per MSP because the Scottish Parliament site does that quite well so it would be a duplication of effort.
Well done.
#25 by Dr Bill Reynolds on May 31, 2011 - 8:29 am
These data shows that not all of the defecting Liberal vote went to the SNP, and that the SNP took votes from everyone.
#26 by James on May 31, 2011 - 8:39 am
Agreed.
#27 by Chris on May 31, 2011 - 8:50 am
Wow, thank you!
My own favourite is coloured electoral maps a la alba.org.uk, but I can’t honestly call them useful, just great fun – it’s good to let your eye wander from one election’s map to another to another, getting a quick visual summary of the changes over the years.
I believe I’ve spotted an insignificant typo, but here goes: in the Lothian page, in the detailed breakdown of the vote allocation process, the Labour divisor jumps from 4 to 7 when Labour wins a seat near the bottom of the table.
#28 by James on May 31, 2011 - 8:53 am
Good spot! Thanks so much. All fixed now. And yes, coloured maps would be lovely. Maybe one of the good helpful people on this thread might be up for doing eight of them..
#29 by Malc on May 31, 2011 - 8:55 am
That’ll be a Malc bad…
#30 by Doug Daniel on May 31, 2011 - 10:30 am
I noticed the use of “2nd votes” on regional list vote tables. I’d use “list votes” myself…
#31 by Doug Daniel on May 31, 2011 - 10:31 am
Other than that though, great piece of work. It puts the BBC to shame, but then most of their election stuff was half-baked anyway.
#32 by James on May 31, 2011 - 11:24 am
We’ll stick with 2nd vote. It’s what the Scottish Government uses. But thanks..
#33 by David Morris on May 31, 2011 - 12:59 pm
I’m starting to get more interested in Scottish politics (I’m an England resident though). Elections and voting methodologies are major interests of mine and have been for some time.
What was the reason for having both constituency and regional MSPs? Does this not make it difficult for people to know who to contact, or do they have different duties?
#34 by James on May 31, 2011 - 1:06 pm
Good question. The AMS system used is a compromise. It allowed Labour’s leadership to placate their anti-PR stalwarts, while giving a proportional element to keep the Lib Dems and others happy. In practice, people are supposed to contact their constituency MSP first and only if they don’t get a satisfactory answer (or by mutual consent) should a regional MSP take up their casework.
Labour used to call the PR side “the assisted places scheme”, and complain about “second tier politicians”. Now most of their MSPs are elected off the regional lists I expect that chat to dry up.
#35 by Jeff on May 31, 2011 - 1:10 pm
Good question David. The reason for regional MSPs was to allow a ‘top-up’ to the constituency results in order to balance out any unfairness in seats won compared to votes received.
In Scotland, one party can win as much as 70% of the First Past the Post seats with only 40% of the vote. The constituency MSPs ensure that their share of MSPs moves a lot closer to the 40%.
And yes, this makes it difficult to know who to contact but overall it seems to be beneficial for a region to have different parties it can contact if they do not wish to contact their constituency MSP (who is usually the first port of call and, arguably, a lot busier than your average regional MSP).
Scotland hasn’t sussed it all out yet but we’re getting there and like to think we’re streets ahead of the Westminster system!
EDIT:
(Sorry, didn’t realise James had already answered you)
#36 by James on May 31, 2011 - 1:11 pm
Ha!
#37 by Jeff on May 31, 2011 - 1:24 pm
(My answer was better)
#38 by James on May 31, 2011 - 1:32 pm
A better answer, but to a slightly different question 😛
#39 by Dr Bill Reynolds on May 31, 2011 - 2:29 pm
I think that voters should have a choice about who to approach.If they do not fancy the constituency MSP,why should they not choose from the regional liast MSP’s?
In theory a regional MSP should have a greater workload than a constituency MSP.Whether that happens probably depends on the MSP.I know that some of the SNP,regional MSP’s have been extremely hard working (eg Christine Graham and Dave Thomson). Nearly all of them are now constituency MSP’s.I suspect that their work for their constituents as regional MSP’s helped them to win constituencies.
#40 by James on May 31, 2011 - 2:33 pm
I don’t fancy any of my MSPs. Oh, I see what you mean.
Yes, and beyond that. In 1999-2003 Robin was regarded by people across Scotland as their Green MSP, which certainly added to the workload.
#41 by Dr Bill Reynolds on May 31, 2011 - 2:56 pm
Well if you don’t fancy any of your MSP’that too bad.My point is that people have options and can choose to go to their constituency MSP or a regional MSP.
While I can understand that people accross Scotland might view Robin as their Green MSP,I doubt whether he would have wanted to take cases outside of the region (Lothian) that he was elected to represent.I am sure that he was very active in that region.That is my point.How you got elected may have little to do with how hard you work.It is possible that some constituency MSP’s are lazy and some regional MSP’s are extremely hard working.I know that some SNP regional MSP’s were hard working and got themselves know.That probably cointributed to getting them elected by first past the post in certain rural constituencies.I think tht Robb Gibson ,Dave Thomson and Christine Graham are good examples.It is also likely that some regional list MSP’s did not work,or make their faces known.The success of the individuals that I have named (and others) show that to be a mistake.If a proportional system is designed to give more people representation,then we are entitled to expect politicians elected under a PR system to be active in their communities and to represent voters who used their second vote to elect them.
#42 by James on May 31, 2011 - 3:05 pm
I’m sorry, I was being flippant. I am one of those lucky eighth of Scots who’ve always had a Green MSP.
And I personally think if we broke the regions into three and elected them by STV we’d get a better Parliament.