As discussed on Scrapbook at the weekend, the Liberal Democrats aren’t ones to let the facts get in the way of a misleading bar graph. A reader has since supplied a copy of a chart distributed by “Fib Dems” in Batchwood ward in St. Albans.

The graph on the left was what councillors Allan Witherick and Amanda Archer put their name to. The one on the right shows how it should look:

Perhaps eagle eyed readers can spot the difference?

  1. Pingback: Duncan Stott
  2. Pingback: L DTUC
  3. Pingback: Estelle Hart
  4. Pingback: Rob Oxley
  5. Pingback: Wes Streeting
  6. Pingback: Rob Oxley
  7. Pingback: Elliot Page
  8. Pingback: David Hodges
  9. Pingback: Tom Miller
  10. Pingback: Roxanne Ellis
  11. Pingback: Sarah Brown
  12. Pingback: Tom Copley
  13. Pingback: Jonathan Pryor
  14. Pingback: Louise Haigh
  15. Pingback: Ben Furber
  16. Pingback: Max Bell
  17. Pingback: Ben Folley
  18. Pingback: Tom King
  19. You make an assumption, which appears to invalidate your claim on this one. You assume that the bottom of the visible bars is zero. If it is about 1500 then the graph is correct. Dishonest, but correct

  20. Actually, I don’t think this is dishonest or misleading at all (and I have seen misleading ones!). The subtitle of the graphs is “Only one vote in it” and it is concentrating on the difference between the vote margins. In fact if you look at the representation of the one vote difference then it is being overly kind to the Conservatives when representing the 80 vote difference. By your argument using the same scale the Conservative bar should be somewhere off the bottom of the leaflet!

  21. @Tom: We’ve double checked. Our source says: “There’s no continuation of the graph below. ‘Only one vote in it’ is the bottom of the textbox”, adding (correctly): “There is no mathematical reasoning that can explain a Lib Dem bar graph. They just make shit up!”

    @LibSync: Well it isn’t off the bottom of the leaflet. This is the totality of the graph! Therefore this *is* completely dishonest and misleading. QED!

  22. Pingback: Rachel
  23. Pingback: Matthew Skinner
  24. … as for the argument going “the Y scale doesn’t necessarily start at zero”, there would need to be some signifier of this. This is usually a number OR a zig-zag line at the bottom of the axis.

  25. Aye, looks awful to me. If, that is, you ignore the fact that the bargraph is not to scale and the fact that the figures are also given. A funny sort of dishonesty right there…..

  26. Pingback: Ryan Duffer
  27. “Well it isn’t off the bottom of the leaflet. This is the totality of the graph! Therefore this *is* completely dishonest and misleading. QED!”

    You miss my point.

    By your reasoning above it is massively under-estimating the difference between the Tory and Labour vote (the gap should be 80 times bigger than between LibDem and Labour) and I assume your argument is that it is making the Tory deficit look larger than it is to encourage tactical voting. But it is being TOO kind to the Tory vote! Therefore if it is “misleading” (which I don’t think it is for the reason stated above (and the massive figures printed on the bars also help!)) it is in the opposite way than you imply and actually mitigates against the “lie” that you think is being perpetrated!

  28. What intellectually dishonest tribal crap. You are invoking some imaginary form of bar graph which doesn’t exist. I.e. one where “the Conservative bar should be somewhere off the bottom of the leaflet”. The correct form is indicated on the right of the image in the post: everything is to scale!

  29. “What intellectually dishonest tribal crap”

    Right back at you!!

    We could go round in circle all night arguing this but my point is that the bar chart is all about the one vote difference (as evidenced by the massive subtitle and the massive numbers!).

    Would you say the same chart with just the LibDem and Labour bar would be misleading? (what with the massive numbers and the subtitle an’ all)

    If you do then, well… (we’ll never see each other’s point)

    However if it isn’t then would the addition of the Tory vote using the same scale be misleading (the Tory vote would be 80 times further down)?

    Then artificially increasing the Tory vote so it fitted on the graph, would this mislead you further into thinking the Tory vote was way down lower than it was to encourage Tories to tactically vote?

  30. You play down the significance of the Tory bar being wrong by saying “the bar chart is all about the one vote difference” between Lib Dems and Labour. In fact, the purpose of the chart is to disguise the a three way marginal by misrepresenting the Conservative showing. You say that it’s fine for the bars to be the wrong size provided that they have figures on.

    The size of the graph is as it appears in the images above. There is no “virtual graph” extending off the leaflet to accommodate an invisible Tory vote that “would be 80 times further down”.

    You really are Through the Looking Glass on this one.

  31. A more accurate and clear breakdown would be by %.

    LD 32.51%
    Green 4.59%
    Lab 32.48%
    Cons 30.43%

    Total Votes 3904

  32. Pingback: Nick Sutton

Leave a comment

Your email address will not be published.


Comments are limited to 1000 characters.