Jonathan Corum for The New York Times examines word usage by President Barack Obama in his State of the Union addresses and the words used by Republican candidates in their debates. Many of you will be happy to know that no word clouds were harmed in the making of this graphic.
Words used in SOTU and Republican presidential candidates in debates
“Many of you will be happy to know that no word clouds were harmed”
Are you sarcastic
Are you sarcastic? (sorry for the missing question mark)
dead serious.
Hey nathan, I’ve been thinking about word clouds a bit recently and agree that they have some serious drawbacks. That said, I don’t think they should be given up on. See http://blog.fellstat.com/?p=101
I would say that reducing debates to words isn’t a good idea.
In France they do the same.
It’s just gonna make candidate work like people code their description of website for search engines robots.
Everybody will answer to my comment “but it’s just an overview”, but who really thinks that? You read this, you think Ron Paul isn’t a good candidate because he doesn’t say the word job, meh.
Plus this is hard to read! how can we compare 3 years to 3 months? there are lot of flows.
We’ve gone from “sounds bites” to “keywords”. Soon we’ll have politicians that just say “Work, work, work, work, work, work, and tax, Thank You.”
Where is “Of the People, For the People, and By rhe People”? from ANYONE!!!!
I was thinking yesterday it would be great to see how many times the candidates mentioned the fault of others versus placing the fault (of anything) on themselves.
The scale for word count is very confusing. So the lines are once that they used the words 5 times per 1000, three times that the used the word 5 times per 1000??? The text highlighting Mitt Romney’s use of jobs is arbitrary since the other highlights all explain the highest count in their series. Also, the two text highlights for Newt Gingrich only state the obvious from the graphic and don’t add anything. Add text on a few high points or anomalies that explain a data point, but be consistent and let the graphics speak for themselves.
I now see that “—Used 5 times per 1,000 words” is not a scale or legend, but simply the label of the X axis for “5” for the first word. Everything is crammed at the top and combining the explanation of the X axis into the label does not work. Just give space above it to label the column. Also, is “frequency of word usage” the label to the first column (would be incorrect) or the title of the chart? Again, bad placement and spacing.