Bad graphics: manipulation or laziness?

Last week I sent this tweet of a graphic in the Daily Mail.

Some of the stack of re-tweeters said they were immediately going to use this in their school lessons, assuming like me that the choice to start the axis at 0.55%, rather than 0%, was deliberately to make the 0.1% revision in quarterly growth look as big as possible, and so confound the gloomy predictions about Brexit.

But others pointed out that these axes could possibly just be due to casual use of Excel. So I checked the default horizontal bar chart that Excel will produce when presented with just the two numbers 0.6% and 0.7%.

Default horizontal bar chart from Excel when given numbers 0.6 and 0.7

This turns out to look very much like the Daily Mail’s graphic, so if we were going to be very generous, maybe this was more laziness than manipulation? And perhaps we should be relieved they did not use the hideous default 3-d version, in which it’s impossible to even see that it represents 0.6 and 0.7.

Default 3-d vertical bar chart from Excel, which is supposed to portray the numbers 0.6 and 0.7

And of course this is nothing compared to what Fox News gets up to.

What might have been a better presentation? The Office for National Statistics (ONS), provided the series back to the 2008 crash, which the Daily Mail also used, which gave a fine context but rather loses the current story about the revision.

But ONS, bless them, provide a spreadsheet with all the data. Taking the series since 2014, using the stacked-bar option to show the revision, and adding some text boxes, the graphic below was produced in Excel in a few minutes.

ONS data, using Excel’s stacked-bar option to show the latest growth revision

While no doubt a proper designer could do a lot better, I think it is not too bad at showing the typical growths over the last few years, that there has been no short-term negative effect of the Brexit vote, and the appropriate size of the revision.

There’s a long debate about whether disasters in history are primarily due to conspiracy or cock-up, malevolence or incompetence. And just like history, this Daily Mail example may be a bit of both. Or maybe just a rushed job. But whatever the cause, it’s still depressing to see graphics that give a misleading impression of magnitude.

[This article was edited after publication to reflect that the Daily Mail did include the full ONS series in their article]

Like what you read? Give David Spiegelhalter a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.