Subscribe Bookmark


Jul 23, 2015

The Pictorial Superiority Effect with JMP

JMP benefits heavily from the pictorial superiority effect. To save you a moment of searching in Wikipedia, here's what that is: "According to the picture superiority effect, concepts are much more likely to be remembered experientially if they are presented as pictures rather than as words." I was reminded of this effect when a friend sent me a link to Alex Lundy's talk on "Chart Wars: The Political Power of Data Visualization." It is a rather humorous (although please be warned that the video contains some expletives) and educational talk espousing the benefits of tools such as JMP.

In the video, Lundy displays the names of a few of the experts in the field of data visualization. We pay a lot of attention to these data viz gurus at JMP; specifically, you may have noticed that we frequently invite Stephen Few to speak at JMP events. I would add Ben Schneiderman to that list, and there is certainly a more comprehensive list to compile.

Lundy also shows a quote from H.G. Wells -- "statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write" -- and adds the word "visual." Add the word "dynamic" to this statement, and now you are describing a state-of-the-art statistical tool called JMP. To demonstrate this, let's use his example of published White House staff salaries. You can get the data set yourself.

I was able to generate the same graphic Lundy shows and distribute it in a few minutes with JMP. Using a combination of the Distribution platform with the Data Filter and Tabulate, I quickly gleaned more information about the sum and distributions of salaries of different subsets of the White House staff.

For example, the combined salaries of staff with the status "Detailee" is $3,966,222.00, and they have significantly higher mean salaries ($123,949.19)...

...than the rest of the White House staff ($77,320.55), who have a combined salary of $35,180,848.00.

You can easily get this information in JMP, and it is not trivial to do this with other software. This is why I believe JMP is the best statistically visual software product on the market.

Article Tags