Graph-Fix Round-up – Lloyd George knew my bar chart


12:22 pm - March 26th 2010

by Unity    


      Share on Tumblr

Last week, I promised you all a weekly round-up of some the worst examples of graph abuse appearing on election leaflets and although the number of submissions has been fairly low, so far, there’s been plenty of ‘quality’ material to play with.

Before kicking on with the round-up, a special mention has to go out to Mark Pack over at Lib Dem Voice for catching The Mirror’s art department playing silly buggers – nice one Mark.

Okay, that’s the preamble out of the way. Now on with the show, and we’ll kick things off with a fine effort from a minor party.UK Independence Party

Way down yonder in the garden of England, UKIP would like you to ignore the ‘small parties’ who have no chance of unseating the local Tory candidate…

They just don’t want you to know that they’re also one of the ‘small parties’.

Oh, and votes casts a Euro elections are, of course, not a realistic guide to voting intentions at a UK general election, but you have give them a bit of credit for trying it on.

Liberal Democrats

The Lib Dems, of course, bring a strong reputation to the table and, this week, they certainly haven’t disappointed.

Over in Horsham, for example, they’ve been a having a bit of problem locating the baseline…

Hey guys, it’s here… (corrected graph)

You’d think that Oxford, home of one of the world’s top universities, would be the one place where you could be absolutely certain of receiving a well drawn bar chart…

Or maybe not…

Meanwhile, in the Highbury East ward of Islington, the local Lib Dems appear to have forgotten something…

…The Green Party!

Never mind, we all have our forgetful days.

Labour

If imitation is the sincerest form of flattery then Bolton’s Lib Dems must have been absolutely chuffed to bits to get this one through their door…

Or maybe not, when the real figures look rather more like this…

Admittedly, its not as bad as some of this week’s entrants, but its still been tweaked to play up the Tory threat and downplay the LD’s chances.

I’m sure you’ll all agree that we’ve had some strong entries, this week, but there could be only one possible winner of this week’s award for sheer unadulterated bullshit and that’s

The Tories

In Birmingham Yardley, things went really badly for the Tories at the last general election, where they trailed in a very poor third behind the Lib Dems and Labour…

After a result like that, how do you convince the electorate that you’re in with a chance this time around?

You use a bar chart, of course, but not just any old bar chart… you put out this bar chart showing the number of occasions each party has won the Yardley seat since 1918 in the vain hope that they’re so dumb they’ll actually believe that you’re in second place…

Yes, in Yardley, Lloyd George really did know my bar chart!

    Share on Tumblr   submit to reddit  


About the author
'Unity' is a regular contributor to Liberal Conspiracy. He also blogs at Ministry of Truth.
· Other posts by


Story Filed Under: Blog ,Elections2010 ,Humour

Sorry, the comment form is closed at this time.


Reader comments


Your Oxford East graph aint right. Labour’s area is 19,210 pixels. Lib Dem’s area is 17,085. These numbers are the wrong proportion for the votes. Please add 924 pixels to the Lib Dem bar.

I don’t know how you find the time, I don’t think I even bothered to read the leafets delivered in my name, but this is brilliant. I hope you’ll do more in as the election nears! Top Blogging.

This is brilliant and hilarious stuff, that UKIP party one is the winner for me, bloody daft racists.

“Oh, and votes casts a Euro elections are, of course, not a realistic guide to voting intentions at a UK general election, but you have give them a bit of credit for trying it on.”

Well, yes….although you’d be even more upset if we started using the following point:

“UKIP vote up over 300%! Join up with the winners!”

(It’s OK, even when I was working there they never let me near slogan making for good reason.)

UKIP polled around 2% nationally last GE….looks like 6 to 7% this time around according to current polls.

So true but just a tad misleading as a slogan…..

“Oh, and votes casts a Euro elections are, of course, not a realistic guide to voting intentions at a UK general election, but you have give them a bit of credit for trying it on.”

This ostensibly light hearted jibe conceals the real democratic scandal.

Obviously the job of the Euro parliament is not the same as the job of Westminster, and so a small number of voters will vote for a different party in the Euro elections because of the different issues at stake.

However, for the vast majority of voters, the difference between the Euros and the Westminster is that for the Euros your vote for your real first choice of party has a much greater chance of actually counting. Hence it shows a good picture of people’s actual party preferences as at June 2009.

Whereas the Westminster election vote in 2005 already contains a large amount of tactical voting forced upon voters by the shite FPTP system. It suits the main parties (including the LibDems often) to pretend that this pattern of tactical voting is people’s real preference, instead of their ‘least worst in these shit circumstances’ preference.

A Euro vote showing a minor party in a good second place to one of the big three in a constituency is a very valid statistical platform for that party to make a play to squeeze the vote of other two of the big parties – especially if that was only ever theirs as a tactical vote anyway. So fair play to UKIP in this particular case, and of course to the Greens & Respect in a number of others.

Flat-out lying is worse than cherrypicking data. So I’d say the tories’ graph – absurd though it is – is not as bad as the others on display.

the Westminster election vote in 2005 already contains a large amount of tactical voting forced upon voters by the shite FPTP system. It suits the main parties (including the LibDems often) to pretend that this pattern of tactical voting is people’s real preference

Exactly. And the reason that Lib Dems are very keen on bar charts is because they want to encourage people to make this forced choice wherever possible.

I do find it ironic that the party that’s most competent at maximising their votes in districts where they’re strong and therefore actually fighting FPTP elections well is the least succesful in large district PR elections, despite them being the biggest proponents of change.

The LD Euro election campaign was, in virtually every region, bloody awful, they were fighting as if it was a Westminster campaign on a constituency basis.

However Strategist, while you’re right to say that the Euro election gives a clearer picture of ‘true’ preferences, it only does so for those that actually voted. Turnout was about half that of a normal General Election, and I still say that turnout will be higher this time, possibly even back up to ’92 levels.

I also say, from what I’ve seen, that the UKIP Euro vote represents pretty much all their support base, whereas other parties supporters are much less likely to vote in EU elections. And that’s notwithstanding the likely squeeze on their actual vote when campaigns get running in each constituency; the number of UKIP supporters who’ll actually vote expressively in the GE instead of tactically for their least disliked of the top two is likely to be a fraction of those currently supporting them in polls.

@Unity, very nice collection; one question, but without bothering to actually measure them, how far out are the Oxford charts? They look fairly close to me. FWIW, my local campaign organiser does like to draw his graphs so that it looks close for the top two, but we’re not actually losing one for the GE campaign.

We’re using a map. My idea, it’s completely accurate and honest as well, and will have a lot more detail about it on the website when I can be bothered to type it up.

For those of the “different elections don’t count” school, have to say, having studied this a lot, they do. In general circumstances, the level of support within local Govt for a particular party across a constituency will, over time, broadly be followed by support at a GE; this might not follow directly, especially if the sitting MP is popular, but if a party implodes in local govt and loses all its support and, crucially, local councillors to organise the activists, then GE support will likely crumble.

The converse is also true; if a party makes a foothold and grows in local Govt, that gives them an activist base and support base to then make a breakthrough in a Parliamentary election; the LDs are doing it in places like Burnley, having done it already in many other parts, and, worryingly, the BNP are trying it on in Barking as well.

One good reason to change the electoral system is to ensure that over-targetting of resources doesn’t lead uncampaigned in electoral deserts for the BNP and other fringe parties.

Gah! using one for the GE campaign. And it was supposed to say he likes his numbers to be accurate, but I either miscoded or got ahead of myself, d’oh!

OK, so in Horsham, we Lib Dems have modestly hid half our support behind some text. But anyway the Others 12% looks to me about 3/10 the height of the Lib Dem 40% as it should.

Unity’s version has the others’ 12% almost half the Lib Dem’s 40%. I don’t have a ruler to hand but that’s how it looks.

So can this be right? You’re criticising an accurate bar chart and your suggested correction is way less accurate?

So can this be right? You’re criticising an accurate bar chart and your suggested correction is way less accurate?

Apologies for that – Illustrator had a wobble when it rendered the jpeg, even though the AI and EPS files had everything at the correct scale.

Mat:

On the Oxford East chart, the LD is slightly out thanks to the pointed top of the bar, but the Tory bar is produced in a way that makes it appear undersized relative to the others, even the Greens.

What I suspect has happened here is that the Green’s bar has been overlaid on that of the Tories, so the latter would be about the right size if shown in full, which is at least a fairly cute way of doing it.

Now you come to mention it, Joe, I’d be interested to know where Horsham LD’s got their figures from for that graph.

Horsham is currently held by Francis Maude, who got 50% of the vote at the 2005 GE.

At that election, the LD’s came second on 26.7% of the vote with Labour in third on 17.1%.

Is the illegible small print not a source?

14. Strategist

@11 “which is at least a fairly cute way of doing it”
… we should expect no less from the academic brains of Oxford..

@7 Thanks for your supportive comments, MatGB. To pick you up on one, you say:

“Strategist, while you’re right to say that the Euro election gives a clearer picture of ‘true’ preferences, it only does so for those that actually voted. Turnout was about half that of a normal General Election, and I still say that turnout will be higher this time, possibly even back up to ‘92 levels. I also say, from what I’ve seen, that the UKIP Euro vote represents pretty much all their support base, whereas other parties supporters are much less likely to vote in EU elections.”

Is that just your impression, or do you know if a psephological study has established that? Just asking, because my impression (not based on any academic knowledge) is that a rising tide of increased turnout seems to lift all boats.

Or in other words, people who sit out a Euro election but will turn out for a General are as likely to have UKIP as a first preference (or Green, or anything) as to be for one of the major parties. The fact that you don’t see this in the data is due to the tactical voting effect you mention: “notwithstanding the likely squeeze on their actual vote when campaigns get running in each constituency”.

Joe:

Yes it is – and it looks as if the figures refer to County Council elections in 2009 where the turnout was about 42-43% compared to 67% at the last GE, so not the best comparator.

In fact, the last time anyone other than a Tory was elected in Horsham was 1876.

Strategist, it’s based on a number of polls and exit polls I read, combined with growing up in and living for a long time in a strong UKIP area (specifically the Torbay and Totnes constituencies).

We know (from polling data and actual results on the ground) that nearly half the UKIP vote in an EU election actually comes from people who vote Lib Dem in the General and local elections in that area. But we also know that people that identify as UKIP voters in canvassing or surveys are almost certain to vote in an EU election, as it’s on the issue they care about) whereas most other voters, as can be seen by turnout, don’t.

For normal parties, increased turnout helps them all equally (although that’s not actually true, Labour gains most, Tories least, and I’ve got figures on that one on this machine), but UKIP isn’t a normal party.

While it’s trying to build itself as a general party with a specific emphasis, for many of their voters it is, very much, a single issue party. That issue is Europe. The European election brings them out in droves, and they actually campaign and spend money on it.

I know that last year, my local Lib Dem party delivered a normal council Focus leaflet once during the campaign, and the direct mailing was paid for by the regional party. UKIP actually worked at getting their vote out, and for the most part succesfully did so.

I can’t link you to a specific source for this; many years reading polling data, a fair number of years canvassing and running exit polls, and a lot of first hand experience, some of it anecdotal.

But the GE result in 2005 saw UKIP at very low numbers after they did very well in 2004; sure, this time around they haven’t got Veritarse playing silly buggers, but I suspect, outside of safe Tory seats their vote will get very squeezed come polling day.

It’s worth noting that their vote in the EU elections is inflated a lot; those LD that switch I mentioned earlier who vote for them in the EUs are frequently self-identified LDs who disagree on EU policy specifically but generally like their MP/candidate otherwise. So it’s single issue voting writ large. What proportion of them this applies to I can’t say though.

Unity; I think we’re just used to differing interpretations of bar charts; I was looking at the Tory/Green chart and thinking they were each the right height off the ground, whereas you were looking at the blue area.

TBH, I don’t actually care if the bars themselves are in the right proportions exactly; sometimes doing so makes for a very crap looking chart. What I care is that the numbers given are relevent and properly sourced, I’ve seen some very bad numbers used on occasions (including one where the campaign teams added up the total votes cast in council elections for each party since the prefvious general election and made a chart out of that FFS).

The reason charts are used is to squeeze third party voters into supporting you over the greater evil; Tories bang on and on about how this exact behaviour is a strength of the electoral system, then complain when a party that declares it a weakness does it better than they do.

The reason LDs do it more than other parties is, in large part, because even LD supporters will go and vote “tactically” for one of the big two, even in areas where the LDs are strong, unless they’re given the message, again and again, that the LDs can actually win locally.

The sooner we get a proper preferential voting system, the quicker we can get rid of the bloody things, but until we do, they’re an essential tool, that is unfortunately abused a bit too often.

Here we go, Lib Dems worst offenders among the parties. Altho in this occasion they are beaten to the post by UKIP. That has got to be flat out the worst distortion ive ever seen on a leaflet.

I think the real objection to the Horsham graph is the data source rather than the rendering of it – using county election results.

Perhaps Labour should use whatever the general election results are in bar charts at the next district elections…

I was astounded to find a LibCon columnist pointing out a few of the barchart distortions for which LibDems are notorious until I checked the “Home” page and found that Unity is anonymous because he is a member of the Labour Party. That may explain why only one of the many Labour distortions are included (and that one within the norm of PR publications – it just cuts the chart so the x-axis is at the 10% level). how about things like the unemployment data where Brown quotes the “Claimant Count instead of the 5.1m claiming out of work benefits (which still ignores those who have lost entitlement to claim because they have been out of work for six months and have some savings left, or have given up trying)?

Photo of a Labour flyer I got through the door here in Dunfermline yesterday:

http://www.facebook.com/photo.php?pid=1263903&id=1175064874

The usual “it’s a 2-horse race” crap – ignoring the pasting they got here in the by-election a couple of years ago, and the SNP, etc,etc, but anyway, as for the graph..

According to the numbers, the Lib Dems’ got 62.6% of Labour’s share of
the vote, but my trusty ruler revealed the wee yellow rectangle is only 53.6% of the size of the red one. Pathetic, isn’t it? Labour are apparently petrified of a mere
2.5mm extra height to make the bar chart accurate. Ho hum.

ah, sheesh, how unobservant am I? missing exactly the same bloody leaflet tucked-away in the middle of the original piece.

D’oh!

Ah well. I could always try using single parenthood & the intellectual distruction wrought upon me by my mini-monkeys as an excuse? no? oh, ok.

it’s a fair cop.

🙂

Scottish Socialist Youth are joining in;
http://ssy.org.uk/2010/04/graph-magic/

Oh dear. Quite apart from producing a pamphlet with entirely uncalled-for references to “the Labour candidate from Cumbria”, my local Dunfermline & West Fife mp, LibDem Willie Rennie, felt the need to shorten the SNP & Tories by 5mm each:

http://www.facebook.com/#!/photo.php?pid=1317489&id=1175064874

Particularly annoying when I’m almost certain to be voting for him 🙂


Reactions: Twitter, blogs
  1. Chris Paul

    RT @Unity_MoT: RT @libcon: Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn – This week's batch of dodgy electio …

  2. Gareth Winchester

    RT @libcon Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/dnN8oD

  3. Patrick Hadfield

    RT @libcon: Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn >and the prize for worst pun of the goes to…

  4. Eben Marks

    Excellent post on some barefaced examples of chart abuse from all parties http://is.gd/b0iXR

  5. Liberal Conspiracy

    Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn

  6. Unity

    RT @libcon: Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn – This week's batch of dodgy election graphs

  7. Two Seven Two

    The Tory "wins since 1918" is priceless RT @libcon Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn

  8. uberVU - social comments

    Social comments and analytics for this post…

    This post was mentioned on Twitter by libcon: Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn

  9. Tweets that mention Liberal Conspiracy » Graph-Fix Round-up – Lloyd George knew my bar chart -- Topsy.com

    […] This post was mentioned on Twitter by Liberal Conspiracy, Gareth Winchester, Chris Paul, Chris Paul, Patrick Hadfield and others. Patrick Hadfield said: RT @libcon: Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/8XOoNn >and the prize for worst pun of the goes to… […]

  10. Jae Kay

    RT @libcon Graph-Fix Round-up – Lloyd George knew my bar chart http://bit.ly/dnN8oD – The Tory example is a lulzy #epicfail





Sorry, the comment form is closed at this time.