If those assumptions are false, whether we use a pie chart, bar chart or generalized linear mixed model, our results are no more likely to be true than just asking the magic 8-ball.

On pie charts -

http://www.militarytimes.com/blogs/broadside/2010/05/12/greenside-of-the-week-may-12-2010/

…mrt

wow, you are a bit upset over Jon’s comments. I am a news junkie and watch as much as I can of ALL networks. I get insulted when they pass off the kind of junk that Jon showed us. I am in the minority because I know better. Based on your comments you know better. Good for you.

When Jon says “My comment was as much about television news shows in general as about Fox News, about television viewers in general, and about the apparent decline in viewers’ IQs while the idiot box is energized. ” I have to agree. The sad reality is, most people of the world do not know how to read charts and graphs correctly nor do they know how to create them.

Jon Peltier should be applauded for showing these horrible data visualizations.

As far as insulting people watching TV news and the professionals creating the poor data visualizations, maybe Jon should. But that is just my opinion.

@dmgerbino

I’m a Fox News viewer myself. Not the national edition of Fox News, because I don’t watch any national television news. I watch the local Fox News out of Boston because of all of the local news shows, Fox has the personnel that seem less like plastic talking heads and more like real people.

Fair or not, it is a national pastime to make fun of Fox News. To illustrate, run Google searches on [___ News is a joke], inserting various identifiers in the blank. Only Fox News has the complete search phrase intact in the search results. Not ABC, NBC, CBS, MSNBC, CNBC, or CNN.

My comment was as much about television news shows in general as about Fox News, about television viewers in general, and about the apparent decline in viewers’ IQs while the idiot box is energized. However, you’re correct that some people may have taken my attempt at entertainment as an insult. To those readers I say:

I apologize for any comments that you may have found offensive.

As always, people are free to visit the sites they want, and to read the pages they want. If anyone unsubscribes from my blog for this or for any other reason, I will not be offended.

I should also label the chart axes and add a chart title, but the whole article talks about approval ratings, and I’m making a point about the chart, not about the data.

But I’ve cleaned up one of the charts as an example. To make this chart, I had to convert the percentages in the source data into whole numbers (i.e., multiply by 100), then use a custom number format of [<80]0_%;0"%" for the axis labels.

Quite frankly, you owe many of the people an apology for insulting them. If you do not apologize, you are not living up to the first sentence of your policy.