27 November 2015

Celebrating Thanksgiving through Statistical Analysis

I hope you all had a happy Thanksgiving yesterday, if you're American. As you may have noticed, I took the day off blogging. (Well, sort of, as I write these things anywhere between a day and six months in advance! Today's is being written on Tuesday, for your information. The picture below I edited in on Thursday night, though.)

Thanksgiving is my favorite holiday, and members of my family celebrate Thanksgiving not by eating turkey, but through three-way Cincinnati chili, a dish of mystical significance for people from my city-- though I'm not aware of anyone from my city who goes this far in their devotion to the dish.

This is actually a four-way, though: note the onions.

My love of Thanksgiving and my amateur interest in statistics meant that I was of course positioned to be deeply enthusiastic about a recent article on Nate Silver's website FiveThirtyEight, "Here's What Your Part Of America Eats On Thanksgiving." I was sorry to have missed the original poll where they collected the data, however, as none of the survey's 1,058 respondents reported eating chili, as I discovered when I downloaded the source data from GitHub: I was surprised that only 82% of respondents reported eating turkey, but most of the non-turkeyers do stuff like ham (though some eat lasagna, and many are vegetarians, hence, no turkey).

The real interesting discovery is the proliferation of side dishes; despite living in New England for seven years now, I've never eaten Thanksgiving here, and thus had no idea that 56% of New Englanders have squash as a side dish (compared to the natural average of 18%). The Middle Atlantic (my home region) disproportionately prefers biscuits and rolls; it's never occurred to me that you wouldn't have biscuits and rolls with a traditional American-style Thanksgiving dinner.

The other thing I read recently was this analysis of U.S. holiday travel, from the Bureau of Transportation Statistics, prompted by my wife and I wondering if it really was true that the day before Thanksgiving was the busiest travel day of the year. It turns out that the answer is no; this is only true for air travel: "Thanksgiving Day is a more heavily traveled day then [sic] Wednesday. Among those traveling more than 100 miles, travel is evenly spread throughout the Wednesday-Sunday period, with no statistically significant difference among the traffic flows during those five days."


There's lot of other fascinating data in there about Thanksgiving travel. This year we're travelling around 700 miles, well above the national average of 214! So I guess I am 3¼ times as devoted to Thanksgiving as the rest of you. (My kid sister, however, is 3¼ times as devoted to Thanksgiving as I am!)

Hope you enjoyed your Thanksgiving; even though I am writing this on Tuesday, I know I did!

No comments:

Post a Comment