Among the many current-year estimates we build from the American Community Survey and decennial Census is ancestry. The ACS asks the question in the following manner –

“What is this person’s ancestry or ethnic origin?

(For example: Italian, Jamaican, African American, Cambodian, Cape Verdean, Norweigan, Dominican, French Canadian, Haitian, Korean, Lebanese, Polish, Nigerian, and so on.)”

Is ancestry race based? Is it nationality based? Is it language based? Is it culturally based? Religion based? Is it based on where I was born? Where my great-great-great grandmother was born?


Me? Glad you asked. I was born in Canada, am now a dual citizen of the United States and Canada, but grew up being told I was a “mongrel” – a mix of German, Dutch, Scottish, English, Irish, French, a few other European ancestries of which one or both of my parents chose for reasons unbeknownst to me not to disclose, and based on 19th century photographic evidence of a man who looks strikingly like my father sitting with his wife (definitely not dressed in the Mennonite fashion of the time), Iroquois.

My family’s North American roots on both sides go back over three hundred years, and first to the original Dutch settlement of New Amsterdam. That Dr. Keirstead ended up in the rugged moose infested interior of New Brunswick was presumably slightly more appealing to him than the parting of company of head and body offered as the alternative by the British soldiers who were probably hired Hessian mercenaries anyways.

But I digress.

I was distressed by two things in the census question. First, that the concept itself is rather squishy, and second, that in order to be thorough, I would need much more space than they provided. Some people quite understandably leave it blank. Others write down everything they can think of in the space afforded.  Some invent ancestries which don’t even appear in Wikipedia.

The results are, of course, quite interesting. Because we pander to the mapping community here and are thus obliged to have at least one map per article, we mapped people who in one sort or another identify as having Canadian ancestry – including the Acadians (Cajuns) who were unceremoniously shipped from Nova Scotia to Louisiana after the French were defeated in those parts. This might offend them, who don’t view themselves as being of Canadian ancestry at all. Even worse, if you identified yourself as French Canadian, you did so because you believe it to be a different ancestry than either Canadian, French, or more properly French (excluding Basque who have been part of France since the time of Charlemagne but aren’t French), or Cajun.

And herein lies the whole difficulty with this data.

Who knew there were so many Canadians in Albuquerque? Sorry, again I digress.

So, why do we include this data?  Because sometimes it passes from interesting to valuable.  Just ask the folks at Chevy who had the cool sounding Nova. Great name. Focus groups in Michigan loved it. Not so much in Spanish speaking countries, especially since it was the era when that was unfortunately an appropriate label for an American car.

Because, really especially because, ancestry is self-reported, it matters to those who reported it. Identifying as “Cuban” rather than “Hispanic” matters, and if you are trying to locate facilities, name your products, or just plain sell stuff to people, you had better know these things. So the next time you are tempted to skip past the page long ancestry report for a trade area, please take a look in case it turns out to be valuable information. If it isn’t, it will at least be interesting.