« Artistic license | Main | Lessons from propaganda »



while I agree with your proposition for this dataset given its structure, I have a problem with the data proper. the question is which is the single best method of discipline. and the answers add up to 100%. but in a real-world situation, parents would talk first, then if ineffective, go for a harsher punishment, then eventually a third one. Apart from the psychotic fringe which doesn't represent 1% of the population, no one would beat their child before talking to them about what they did. another problem with that question is that it really suggests a "good" answer. Parents are tempted to report that they can solve all their problems with talking.
now AFAIK parents split into two categories, those who never use physical punishment ever, and those who accept it as a possibility. There is a strong geographic divide between the two (there was a recent post on strange maps about that)
If I were to investigate on the subject I would ask questions like "how often do you resort to " + various methods with possible answers including "systematically", "less than once a week", "less than once a month", ..., "never". the shape of the data would be very, very different!

Robert Kosara

I wonder why you chose line charts for your small multiples redesign, when the x axis is clearly a categorical variable. Bars would have been a better choice there.

Good discussion otherwise, though I think Jerome is spot on with his criticism of the source data.


Jerome: Agreed on the data. Their conceit was that these polls have only one question; that is probably why they didn't ask anything else. I'm most disturbed by the presumption that parents treat their sons and daughters with the same tactics.

Robert: With 4 bars in each of 16 charts, we have 64 bars in all, and that is just too much for readers to process, especially since the profiles of the responses are not that different. I have voiced frequently on this blog that I don't like or follow that particular rule. (See my posts on profile charts.) What is the advantage anyway of using bars instead of lines?


Okay, I agree that slavishly sticking to rules without questioning them is not always the right choice. But in this case, I think it would have made thing clearer. There is a lot of overplotting, so using grouped bars would have been easier to read.

I guess the lines work here because there is an order to the types of response, which vary right to left from least to most severe. I still think that bars are a better representation fro this type of data.

(Trying the twitter login because typepad says "we cannot accept this data" if I try to submit with the same credentials I used above)

Andrew Gelman

Ditch the alphabetical order, dude!


EagerEyes: I really don't like grouped bar charts because they make you literally and visually jump over barriers to compare data.

Andrew: Good point. I did think about the order but there are four data series, and I didn't think any particular order is more sensible. The profiles all look remarkably similar.


Online cash loans are becoming the new wave of the future and people are applying if only to establish themselves with these services in case of an emergency. So if you find yourself short of cash " Click Here


Start Earning $300 Today - Every Single Day!!! Easy2Earn Money on Internet Without Investment... Spend 30 Minutes a Day on internet..plz visit once time

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Marketing analytics and data visualization expert. Author and Speaker. Currently at Columbia. See my full bio.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Good Books

Keep in Touch

follow me on Twitter