data point visualization is one of the most important tools we have to examine data . But it ’s just as leisurely to mislead as it is to educate using chart and graph . In this article we ’ll take a look at 3 of the most common way in which visualizations can be deceptive .
Truncated Y-Axis
One of the easy ways to misrepresent your data is by mess up with the y - bloc of a saloon graph , line graphical record , or strewing game . In most cases , the y - axis ranges from 0 to a maximum time value that encompasses the range of the data . However , sometimes we alter the range to better spotlight the difference of opinion . Taken to an extremum , this technique can make difference in data seem much great than they are .
Let ’s see how this works in drill . The two graphs below show the exact same information , but use different scales for the y - bloc :
On the left , we ’ve constrained the yttrium - axis to range from 3.140 % to 3.154 % . Doing so makes it look like interest rate are rocket ! At a coup d’oeil , the barroom sizes mean that rate in 2012 are several times in high spirits than those in 2008 . But expose the datum with a zero - service line y - axis tells a more precise picture , where interest rates are staying static .
If this case seems overdone , here are some veridical - human race object lesson of shortened y - axis :
Cumulative graphs
Many people opt to make cumulative graphs of things like number of users , gross , downloads , or other crucial metric unit . For example , rather of showing a graph of our quarterly revenue , we might choose to expose a running total of revenue earned to date . rent ’s see how this might look :
We ca n’t tell much from this graph . It ’s moving up and to the rightfield , so thing must be going well ! But the non - cumulative graph paint a unlike picture :
Now thing are a lot clearer . receipts have been decline for the past ten twelvemonth ! If we scrutinize the cumulative graph , it ’s possible to tell that the slope is decreasing as metre goes on , indicating shrinking tax income . However , it ’s not immediately obvious , and the graphical record is fantastically shoddy .
There are lots of existent - world cases of accumulative graph that make thing seem a mass more positive than they are . A prominent example isApple ’s utilization of a accumulative graph to show iPhone sales .
Ignoring conventions
One of the most insidious tactics people use in reconstruct misleading data visualization is to violate standard practices . We ’re used to the fact that pie chart constitute region of a whole or that timelines progress from left to right . So when those rules get violated , we have a difficult time see what ’s actually going on . We ’re wired to misconceive the datum , due to our reliance on these conventionalism .
Here ’s an object lesson of a pie chart that Fox Chicago air out during the 2012 primary feather :
The three slice of the Proto-Indo European do n’t add up to 100 % . The sketch presumably allowed for multiple responses , in which lawsuit a barroom chart would be more appropriate . alternatively , we get the impression that each of the three candidate have about a third of the keep , which is n’t the sheath .
Another example is this visualizationpublished by Business Insider , which seems to show the opposite of what ’s really going on :
At first glimpse , it look like artillery deaths are on the descent in Florida . But a closer expression shows that the y - axis is upside - down , with zero at the top and the maximal note value at the bottom . As gun for hire deaths increment , the line splosh downward , go against a well establish conventionalism that wye - values addition as we move up the page .
There ’s a childlike takeaway from all this : be heedful when designing visualizations , and be extra deliberate when interpreting graphs created by others . We ’ve covered three common techniques , but it ’s just the surface of how people use data visualisation to misinform .
Do you have an example of a in particular poorly built visualization ? get us knowon twitter . Also , if you want to fall in us each workweek for more data - drive insight , enter your e-mail address in the form on the sidebar to sign .
This postoriginally come out onHeap Analytics ’ blogand has been republished with license fromRavi Parikh . For more from Heap Analytics , direct on over to their data blogor succeed Ravi onTwitter here .
DataScience
Daily Newsletter
Get the best technical school , science , and culture news in your inbox day by day .
News from the future , delivered to your present .
You May Also Like
