Extended Thoughts about Convincing with Statistics

Yesterday I wrote about my nervousness towards the extreme weather happening these days and implied how upset I was with people who didn't believe that climate change was real. So when I started to read the articles assigned for class tomorrow, I was surprised the articles addressed my concern of statistics effectively convincing a wide range of audience (coincidence??)

In this post in Professor Gelman's blog, Malcolm Gladwell's book David and Goliath is discussed and criticized (somewhat). Professor Gelman points out that a lot of people can be convinced without a straightly statistically valid arguments (like the Andrew Carnegie story). What I learned from this article was that the style of storytelling matters, the tone that is written.

But it was also interesting that Professor Gelman had problems with Gladwell’s stories not being messy enough to represent the real thing because: Over-smoothing = less messy = less real. To Gladwell's comment, "There is a place for storytelling, in all of its messiness. . . . narratives sometimes begin in one place and end in another," Professor Gelman replies, "Try resisting the urge to tie every story into a bow. Let some of the loose ends hang out." So I guess while a nicely arranged narrative with some statistical support can be effectively convince people, there is a deeper level peril of not "really" arguing a "truth". I have to think more about the purpose of using statistics when convincing other people.