Travel, Tiny Data, and a little bit of Research

I hate tiny datasets.

Well, not really. I just find them difficult to work with. You’d think they’d be easy, just plot the data and you’re done. I’ve found trying to engage someone, or tell a story with a tiny amount of data, is a hard thing to do.

For week 6 of Makeover Monday we had a dataset with a grand total of 8 rows and 4 columns.


This was looking at figures comparing Chinese New Year with Thanksgiving. As I usually do, I started by throwing things around in Tableau. After just a few minutes I was already becoming resigned to creating some bar charts and calling it a day, or even not bothering this week.

But something struck me as slightly strange regarding the number of trips


Now I do realise that China has about 1 billion more people than the US, but here China just unbelievably dwarfs the US in terms of the number of trips. Something just doesn’t seem quite right. So I went back to the article that Eva and Andy provided us with. The relevant part of the article says:

According to data released by Bloomberg, Chinese people undertake 2.8 billion trips during their new year celebrations compared to 46.3 million for Americans during Thanksgiving.“

Ok, so after visiting Bloomberg (which requires a subscription, unless you’re quick with Ctrl+A, Ctrl+C, but I didn’t tell you that), the relevant part of the article says:

“About 2.8 billion trips will be made in China… …according to government estimates. In the U.S., AAA Travel had projected last year that 46.3 million Americans will journey 50 miles or more from home during the Thanksgiving weekend…”

So it looks like the figures are comparing “trips” vs journeys of over 50 miles. I can’t say if that’s a fair comparison or not as both figures are estimates from very different organisations. But for the sake of argument, lets assume they are fair comparisons. Sometimes you get data that just can’t be validated as thoroughly as you’d like, and sometimes you have to come up with some sort of analysis (it’s happened to me multiple times).

But I just couldn’t quite leave it there, I also found an article in the telegraph, here’s an interesting excerpt:

“Over the 40-day period of the Spring Festival, as it is known in China, which begins on January 28, the country’s authorities expect the nation to make some 2.5 billion trips by land, 356 million by rail, 58 million by plane and 43 million by sea in a phenomenon called ‘chunyun’…”

So it appears the estimates from the Chinese government are for a forty day period called Chunyun. That’s a bit longer than the roughly 4 days of Thanksgiving (here I’m assuming Americans get the Thursday and Friday off work, and then don’t work at the weekend after either).

So after that little bit of exploration, here’s my recreation of the original visualisation. Comparing population size, holiday length, and the number of trips per person, per day.

It’s also designed so it should fit on a mobile screen without any issue. Click on the image for the interactive version.