This article is from the source 'guardian' and was first published or seen on . It last changed over 40 days ago and won't be checked again for changes.
You can find the current article at its original source at http://www.theguardian.com/news/datablog/2014/may/12/lies-election-leaflets-five-tricks-european-elections
The article has changed 2 times. There is an RSS feed of changes available.
Previous version
1
Next version
| Version 0 | Version 1 |
|---|---|
| Lies, damn lies and European election leaflets: five tricks to watch out for | Lies, damn lies and European election leaflets: five tricks to watch out for |
| (35 minutes later) | |
| Election time in the UK means that our letterboxes are stuffed with political leaflets from parties eager to get our votes. While most of us take inflated political claims with a pinch of salt, spotting misleading or outright deceptive bar charts and statistics can take a bit of practice. Political parties often use stats to overstate their chances; to diminish their opponents’ apparent popularity; and to encourage tactical votes, even in elections using proportional representation voting where that doesn’t make sense. | Election time in the UK means that our letterboxes are stuffed with political leaflets from parties eager to get our votes. While most of us take inflated political claims with a pinch of salt, spotting misleading or outright deceptive bar charts and statistics can take a bit of practice. Political parties often use stats to overstate their chances; to diminish their opponents’ apparent popularity; and to encourage tactical votes, even in elections using proportional representation voting where that doesn’t make sense. |
| There’s very little to stop political parties from doing this. Under the current law, candidates are prevented from defaming the character of other candidates. Labour ex-minister Phil Woolas was ejected from parliament in 2010 for doing just that. | There’s very little to stop political parties from doing this. Under the current law, candidates are prevented from defaming the character of other candidates. Labour ex-minister Phil Woolas was ejected from parliament in 2010 for doing just that. |
| But candidates can say what they like about facts and figures such as previous election results. They can even mislead voters about how the voting system works quite legally. That’s all “within the rules”. | But candidates can say what they like about facts and figures such as previous election results. They can even mislead voters about how the voting system works quite legally. That’s all “within the rules”. |
| If you don’t want to get duped by a dodgy election bar chart, here are five common statistical dirty tricks – all based on real-life examples from this or previous elections – to watch out for. | If you don’t want to get duped by a dodgy election bar chart, here are five common statistical dirty tricks – all based on real-life examples from this or previous elections – to watch out for. |
| Let us know in the comments or through GuardianWitness if you’ve found good examples of your own. | |
| 1. Mystery data - “guess who-oo?” | 1. Mystery data - “guess who-oo?” |
| Data is only useful if it’s relevant, but many charts make this difficult to judge as they’re not labelled with the source. | Data is only useful if it’s relevant, but many charts make this difficult to judge as they’re not labelled with the source. |
| Even where charts are labelled, the source can be vague and misleading. Take this Tory leaflet for the Stonecot ward byelection for Sutton Council in December 2012: | Even where charts are labelled, the source can be vague and misleading. Take this Tory leaflet for the Stonecot ward byelection for Sutton Council in December 2012: |
| At first glance you might think this “Stonecot ward” data was from a previous Sutton Council election. In fact, it’s from the Greater London Authority Croydon and Sutton constituency member election where the voting breakdown is published per ward but the assembly member is elected across two whole boroughs. The Lib Dems went on to win the Stonecot ward byelection with 53% of the vote to the Tories’ 21%, contrary to what the Tories’ bar chart might suggest. | At first glance you might think this “Stonecot ward” data was from a previous Sutton Council election. In fact, it’s from the Greater London Authority Croydon and Sutton constituency member election where the voting breakdown is published per ward but the assembly member is elected across two whole boroughs. The Lib Dems went on to win the Stonecot ward byelection with 53% of the vote to the Tories’ 21%, contrary to what the Tories’ bar chart might suggest. |
| 2. Irrelevant data - “bait and switch” | 2. Irrelevant data - “bait and switch” |
| Even if data is adequately labelled it might still be irrelevant. Relevance in previous election results can be a tricky call to make. The cautious approach is to use the most recent results for the same political body in the same electoral district. But if that doesn’t help to make the case for a party they’re often happy to take anything they can get: older results, results from different political bodies (e.g. council results for a parliamentary election) and even results from a different geographical area to the one being contested. Essentially they’re making the case for product A using the performance data of product B. | Even if data is adequately labelled it might still be irrelevant. Relevance in previous election results can be a tricky call to make. The cautious approach is to use the most recent results for the same political body in the same electoral district. But if that doesn’t help to make the case for a party they’re often happy to take anything they can get: older results, results from different political bodies (e.g. council results for a parliamentary election) and even results from a different geographical area to the one being contested. Essentially they’re making the case for product A using the performance data of product B. |
| The Lib Dems are trying this trick in London in this year’s Euro elections. Here’s a bar chart from the front page of their booklet delivered in the Sutton and Cheam area: | The Lib Dems are trying this trick in London in this year’s Euro elections. Here’s a bar chart from the front page of their booklet delivered in the Sutton and Cheam area: |
| As the label says, these are the results from the last general election – a first-past-the-post race for the local constituency. However, elections to the European parliament use proportional representation – a system the Lib Dems strongly support – across the whole of London. Under that system, every vote can count: Labour can and probably will win seats here. | As the label says, these are the results from the last general election – a first-past-the-post race for the local constituency. However, elections to the European parliament use proportional representation – a system the Lib Dems strongly support – across the whole of London. Under that system, every vote can count: Labour can and probably will win seats here. |
| Sutton and Cheam makes up only 1% of London’s voters. The “race” for the Euro parliament isn’t a local one at all. Even if the Lib Dems top the Euro poll within Sutton and Cheam it probably won’t make any difference to how many Lib Dem candidates get elected among the eight MEPs in the London region. Likewise, whether Labour can “win” the Euro vote in Sutton and Cheam doesn’t matter. Local Labour voters will have their votes added with other Labour voters across London and will almost certainly help to elect at least one Labour MEP. The seats in London in 2009 were: three Tory, two Labour, one Lib Dem, one Green and one Ukip. | Sutton and Cheam makes up only 1% of London’s voters. The “race” for the Euro parliament isn’t a local one at all. Even if the Lib Dems top the Euro poll within Sutton and Cheam it probably won’t make any difference to how many Lib Dem candidates get elected among the eight MEPs in the London region. Likewise, whether Labour can “win” the Euro vote in Sutton and Cheam doesn’t matter. Local Labour voters will have their votes added with other Labour voters across London and will almost certainly help to elect at least one Labour MEP. The seats in London in 2009 were: three Tory, two Labour, one Lib Dem, one Green and one Ukip. |
| This leaflet also shows the Lib Dems, who support proportional representation and backed a referendum for it in 2011, encouraging first-past-the-post style tactical voting against Labour in an election where every Labour vote will count. The Lib Dems’ suggestion that a Labour vote would be futile isn’t true at all. The Euro election in London is at least a five-horse race where the Lib Dems are at risk of losing their single seat, regardless of how well they do in Sutton and Cheam compared to other parties. | This leaflet also shows the Lib Dems, who support proportional representation and backed a referendum for it in 2011, encouraging first-past-the-post style tactical voting against Labour in an election where every Labour vote will count. The Lib Dems’ suggestion that a Labour vote would be futile isn’t true at all. The Euro election in London is at least a five-horse race where the Lib Dems are at risk of losing their single seat, regardless of how well they do in Sutton and Cheam compared to other parties. |
| 3. Inaccurate drawing - “the elastic ruler” | 3. Inaccurate drawing - “the elastic ruler” |
| Here’s another way to put the squeeze on your opponents - shrink their bars as much as you like. This leaflet from the Lib Dems in Dundee West for the 2010 general election greatly visually exaggerates the number of Lib Dem MPs compared with their rivals. Labour’s bar is drawn at just 39% of the scale of the Lib Dems’ one and the SNP’s bar is scaled down to 72% of what it should be relative to the Lib Dems. | Here’s another way to put the squeeze on your opponents - shrink their bars as much as you like. This leaflet from the Lib Dems in Dundee West for the 2010 general election greatly visually exaggerates the number of Lib Dem MPs compared with their rivals. Labour’s bar is drawn at just 39% of the scale of the Lib Dems’ one and the SNP’s bar is scaled down to 72% of what it should be relative to the Lib Dems. |
| This standard of presentation would get you fired from a job in finance or engineering and it certainly wouldn’t get you through GCSE maths, but if you’re standing for parliament it’s supposedly just fine. | This standard of presentation would get you fired from a job in finance or engineering and it certainly wouldn’t get you through GCSE maths, but if you’re standing for parliament it’s supposedly just fine. |
| (Hat tip: BuzzFeed) | (Hat tip: BuzzFeed) |
| 4. Outrageous optimism - “the future’s so bright I’ve got to wear shades” | 4. Outrageous optimism - “the future’s so bright I’ve got to wear shades” |
| What do you do if you’re trailing a distant third after the last election? | What do you do if you’re trailing a distant third after the last election? |
| There was no way that the Lib Dem candidate in Cardiff North in 2010 could use these previous results to make the case for being a serious contender. The answer? Apply some outrageous optimism and dredge up anything that suggests that the Lib Dems’ prospects are stronger than they are. | There was no way that the Lib Dem candidate in Cardiff North in 2010 could use these previous results to make the case for being a serious contender. The answer? Apply some outrageous optimism and dredge up anything that suggests that the Lib Dems’ prospects are stronger than they are. |
| So the Lib Dems used the number of councillors in the area covered by the parliamentary constituency and came up with this: | So the Lib Dems used the number of councillors in the area covered by the parliamentary constituency and came up with this: |
| This is as devious as it is contrived. | This is as devious as it is contrived. |
| Counting the number of councillors rather than the vote share across these 21 wards eliminates support for second and third-placed parties in these wards entirely. Even though Labour holds the parliamentary seat it has no councillors in the Cardiff North area as the first-past-the-post system gives no prizes for coming a close second in many of these wards. | Counting the number of councillors rather than the vote share across these 21 wards eliminates support for second and third-placed parties in these wards entirely. Even though Labour holds the parliamentary seat it has no councillors in the Cardiff North area as the first-past-the-post system gives no prizes for coming a close second in many of these wards. |
| This is irrelevant data and it’s also misscaled even if you want to take it on its own merits. The Tories had 13 councillors, the Lib Dems five, and there were three independents, yet the Lib Dems’ bar is more than half the size of the Tories’ and more than twice the size of “Other”. | This is irrelevant data and it’s also misscaled even if you want to take it on its own merits. The Tories had 13 councillors, the Lib Dems five, and there were three independents, yet the Lib Dems’ bar is more than half the size of the Tories’ and more than twice the size of “Other”. |
| How did the outrageously optimistic claim that “only the Lib Dems can beat the Tories” (never mind Labour) stand up? The Tories gained the seat with just 194 votes more than Labour. The Lib Dems’ share of the vote stayed constant at just over 18% placing them a distant third again. | How did the outrageously optimistic claim that “only the Lib Dems can beat the Tories” (never mind Labour) stand up? The Tories gained the seat with just 194 votes more than Labour. The Lib Dems’ share of the vote stayed constant at just over 18% placing them a distant third again. |
| 5. Missing data - “Uncle Joe’s bar chart” | 5. Missing data - “Uncle Joe’s bar chart” |
| Stalin knew how to deal with political rivals: He’d take them out the back, shoot them, and airbrush them from any old photos they appeared in. The Uncle Joe Bar Chart method is a great way to make inconvenient political opponents disappear. | Stalin knew how to deal with political rivals: He’d take them out the back, shoot them, and airbrush them from any old photos they appeared in. The Uncle Joe Bar Chart method is a great way to make inconvenient political opponents disappear. |
| Here’s a Labour leaflet for the local election in Herne Hill ward, Lambeth Council in 2010. The bar chart shows Labour, the Tories and a tactical squeeze on the “no chance” Lib Dems. Seems legit. | Here’s a Labour leaflet for the local election in Herne Hill ward, Lambeth Council in 2010. The bar chart shows Labour, the Tories and a tactical squeeze on the “no chance” Lib Dems. Seems legit. |
| But this is how Herne Hill voted at the previous council election in 2006: | But this is how Herne Hill voted at the previous council election in 2006: |
| Not only are the Greens the main challenger to Labour in Herne Hill, one of the Green candidates topped the previous poll to win one of the two council seats in this ward. Yet Labour’s bar chart usefully makes the Greens vanish entirely. If there was a “two-horse race” in this ward in 2010 it was between Labour and the Greens not Labour and the Tories. | Not only are the Greens the main challenger to Labour in Herne Hill, one of the Green candidates topped the previous poll to win one of the two council seats in this ward. Yet Labour’s bar chart usefully makes the Greens vanish entirely. If there was a “two-horse race” in this ward in 2010 it was between Labour and the Greens not Labour and the Tories. |
| Perhaps the trick worked. Labour won Herne Hill decisively in 2010 with the Greens a distant second. The Tories were pushed down into fourth place behind the Lib Dems. | Perhaps the trick worked. Labour won Herne Hill decisively in 2010 with the Greens a distant second. The Tories were pushed down into fourth place behind the Lib Dems. |
| While bar charts show a snapshot in time from a particular election, line charts of voting trends over time are much more useful to see where the vote’s likely to go in future. Election results prior to the most recent one are the missing data that is most conspicuous by its absence. | While bar charts show a snapshot in time from a particular election, line charts of voting trends over time are much more useful to see where the vote’s likely to go in future. Election results prior to the most recent one are the missing data that is most conspicuous by its absence. |
Previous version
1
Next version