Misleading charts and data

[Jon von der Heyden]

Hello board

I am preparing a presentation on misleading charts and data (with the purpose that it be easily identified). The charts subject is relatively easy, especially since Bill has so many of the 'little lies' documented here: <meta http-equiv="Content-Type" content="text/html; charset=utf-8"><meta name="ProgId" content="Word.Document"><meta name="Generator" content="Microsoft Word 11"><meta name="Originator" content="Microsoft Word 11"><link rel="File-List" href="file:///C:%5CDOCUME%7E1%5CJONVON%7E1%5CLOCALS%7E1%5CTemp%5Cmsohtml1%5C01%5Cclip_filelist.xml"><!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning/> <w:ValidateAgainstSchemas/> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables/> <w:SnapToGridInCell/> <w:WrapTextWithPunct/> <w:UseAsianBreakRules/> <w:DontGrowAutofit/> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><style> <!-- /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-parent:""; margin:0cm; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:12.0pt; font-family:"Times New Roman"; mso-fareast-font-family:"Times New Roman"; mso-ansi-language:EN-GB;} a:link, span.MsoHyperlink {color:blue; text-decoration:underline; text-underline:single;} a:visited, span.MsoHyperlinkFollowed {colorurple; text-decoration:underline; text-underline:single;} @page Section1 {size:612.0pt 792.0pt; margin:72.0pt 90.0pt 72.0pt 90.0pt; mso-header-margin:36.0pt; mso-footer-margin:36.0pt; mso-paper-source:0;} div.Section1 {page:Section1;} --> </style><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]-->http://www.mrexcel.com/tip142.shtml. And there are all sorts of other examples on the web.

I am eager to gather as many examples of 'bad practise' as possible. I have many chart examples but I am especially looking for more examples on the 'data' subject. I am preparing a list of tips to make sure that data can be easily interpreted.

Example - insist that number formats show zero as hyphen and -tive numbers are formatted to red font and surrounded by brackets:
EG: #,##0,;[red](#,##0,);-

One notable advantage is that only TRUE zero's appear as "-". Using the example above 499 will appear as "0" suggesting immediately that there is a balance possibly worth scrutinising. I have frequently seen colleagues use rounding to 000's to conceal costs.

Do you have any other experiences that may be worth noting?

Thanks in advance.

 

[Long Nose]

Charts have frustrated me in the past, particularly adding too much data and obscuring any gleaming gold nuggets. Does anyone know, how much is too much? Our company charts look like they need charts to explain themselves.

I kid you not, we have charts with 15 series of data on one chart PLUS a secondary Y axis on the right! Am I just terribly small minded? I can't get my head around what is going on there.

 

[Andrew Fergus]

Hi Jon

In the late 1970's one of our former Prime Ministers, Robert Muldoon, would point to economic graphs with a rising trendline and claim all was well. Closer inspection of the graph (which would have the scales in the most miniscule of fonts) found the x-axis had been inverted. It looks like some people are still doing it : http://www.thestandard.org.nz/the-polls-and-petrol-prices/

Andrew

 

[Gerald Higgins]

Hi Jon. Are you familiar with the work of Edward R Tufte ?

His books include many examples of both very good and very bad charting.

 

[PaddyD]

classic bad data practise - percents of percents

 

[Jon von der Heyden]

Thank you alll.

In the late 1970's one of our former Prime Ministers, Robert Muldoon, would point to economic graphs with a rising trendline and claim all was well. Closer inspection of the graph (which would have the scales in the most miniscule of fonts) found the x-axis had been inverted. It looks like some people are still doing it : [URL="http://www.thestandard.org.nz/the-po...petrol-prices/"]http://www.thestandard.org.nz/the-po...petrol-prices/[/URL]

Thanks Andrew, would you believe I have even come across this.

Are you familiar with the work of Edward R Tufte ?

Only through my recent research. I like his LIE FACTOR ratio. Thanks

percents of percents

Paddy, what sort of scenario have you witnessed regarding this? Sounds ominous!

 

[cornflakegirl]

In the late 1970's one of our former Prime Ministers, Robert Muldoon, would point to economic graphs with a rising trendline and claim all was well. Closer inspection of the graph (which would have the scales in the most miniscule of fonts) found the x-axis had been inverted. It looks like some people are still doing it : http://www.thestandard.org.nz/the-polls-and-petrol-prices/

I don't think that graph is misleading though. He states clearly in the text that he's done it, and he's only done it to illustrate the correlation more clearly. (Not that I think it does show a correlation, but that's not the point.)

I've been thinking about misleading data (rather than graphs). I think most of the stuff that I see (and produce!) is to do with poor labelling, not updating text when the data changes, assumptions not clearly noted etc. Probably not the sort of stuff you want for a presentation!

 

[Long Nose]

Last year my boss looked at me as if I had two heads because I'd never used a second Y axis. Anyone here ever used one?

 

[cornflakegirl]

Long Nose: Yes. Any time that I want to graph a margin % against a £ value over time, for example.

 

[Long Nose]

Well, I stand corrected. I've not run across this in past reports.