For sure you have noticed Infographics on top web pages just like Mashable, TechCrunch, and Information is Beautiful. Probably you might have even noticed them reblogged over Tumblr and retweeted through Twitter. They are usually common posts for bloggers. These colorful, graphic representations of data gain a great deal more consideration and go viral much sooner than the same information written out. At the same time, the pitfalls are not as understandable and many top ranking sites publish beautiful infographics that supply little insight. As you start designing a visual representation, ask yourself first: is it significant?

To promote their own research arm LinkedInsights, Linkedin created a wonderful infographic listing the top 3 names for female and male CEOs. The issue? Simply because there are more CEOs named ‘Jack’ does not mean that this information is significant in any respect. Consider this example: we have one hundred CEOs in crowd, 3 are named ‘Bob’, two named ‘Peter’, two named ‘Jack’, the other 93 CEOs all have distinctive, unmatching names. As soon as we gathered these details we could with certainty claim that ‘Bob’ was the top name among CEOs … but so what?

Sometimes a trend does not reveal an insight but merely an inconsequential anomoly. Test for statistical significance before creating that infographic.

You will need a statistical program for example SPSS produced by IBM or even the completely FREE PSPP in order to do this. Each offers several testing choices when examining and studying the data. Decide carefully on the test that matches the data.

Will you be comparing means or several components? Are your factors dependent? How convinced do you think you are? Better figures are often had by accepting lower than 95% confidence rate, though ultimately this does your study a disservice.

Dan Gordon knows that good data can do amazing things. Like write better books. If you’re a writer and want to use analytics to sell more novels you’ll find the best advice with F.S. Publishing.