Using a name to predict someone's age

I enjoy reading the FiveThirtyEight blog and always mine it for great math content. I came across this article and thought it was perfect for a box and whisker plot lesson:

Source: http://fivethirtyeight.com/features/how-to-tell-someones-age-when-all-you-know-is-her-name/

I will post more once I develop the lesson.

Female Named Hurricanes Kill More than Male Named Hurricanes

I heard this story yesterday morning on WNYC morning edition and thought it was perfect for a lesson on causation versus correlation. Essentially the data shows a strong correlation between fatalities and the gender of a hurricane name. Essentially they found that female named hurricanes are much more deadly then male named hurricanes. We can present this to our students with the obvious question, "does that mean that the names are causing higher fatalities?" The goal is to get students to recognize how absurd it would be if we thought this was causal. That would mean that we could simply make a hurricane safer (or more deadly) by changing its name. So this gives students a clear distinction between correlated and causal.

So what is happening? What is a possible cause? If you review the story, articles and data below, you can see that people respond differently to a storm based on its name. It is this response that seems to be causing the difference in results. If we are more afraid of a female named storm, we might take more precautions, etc.

This still leads to more questions and class discussion, why do we respond differently? Why do we give names to storms at all?

Here is the radio story:
http://www.npr.org/2014/06/03/318414845/research-americans-less-fearful-of-storms-with-female-names

You can also find this story on their site:
http://www.wnyc.org/story/research-americans-less-fearful-of-storms-with-female-names/

http://www.wnyc.org/story/study-americans-less-fearful-of-storms-named-after-women/


Article on WNYC: http://www.wnyc.org/story/do-female-named-hurricanes-need-to-lean-in/

Article on the Washington Post:
http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2014/06/02/female-named-hurricanes-kill-more-than-male-because-people-dont-respect-them-study-finds/

Original Study:
http://www.pnas.org/content/early/2014/05/29/1402786111.abstract

Download Data from Original Study:
http://www.pnas.org/content/suppl/2014/05/30/1402786111.DCSupplemental

You can also find the data key here:
https://drive.google.com/file/d/0B5FTxsu-S1CoQ1ZSM2VSR3Yxc3c/edit?usp=sharing

And the original data here:

https://drive.google.com/file/d/0B5FTxsu-S1CoQVJIVzl4SV9yb0k/edit?usp=sharing

Colbert, Causation and Correlation

Interesting problems on correlation and causation are not easy to find. Perhaps I am not looking in the right places, but when I start to search, this is what I find:

Source: http://www.regentsprep.org/regents/math/algebra/practicetests/MCTest1.htm

This problem offers a nice contrast between causation and correlation, but it feels manufactured. There is nothing compelling about it. Nothing that shows why the causation-correlation contrast is worth thinking about. When we discuss the answer, students understand that the rooster doesn't cause the sun to rise. They know this is ridiculous and wonder, "why are we talking about this? No one thinks that a rooster causes the sun to rise." 

And this is where I am struggling. I want to use examples that show a meaningful contrast between correlation and causation, because I know I can find problems that reflect common assumptions about causation and thus spark a class debate. 

How many times do people assume that a correlation means causation? I know there are endless examples of this, but I need to start compiling them.  

I was watching Colbert this week and found one of my first examples. With some major editing, I am going to show this as my opening:

This clip captures answers a lot of the "so what" question as well as a hilarious example of correlation and causation:

I am now going to think about how to turn the autism and vaccine connection into a investigable lesson.