The measure of a metro

In January, Charlotte had 1.8 million people. Today it has 2.3 million people. And no, there was no airlift of half a million residents from the Rust Belt or anywhere else. How can a city gain a half-million people almost overnight? How can a metro area vault from No. 33 in population to No. 23?

It’s all in how the U.S. government defines Metropolitan Statistical Areas. In a recent recalibration effective in February, the two-state Charlotte MSA gained five counties and lost one, and thus grew by half a million people.

Does this even matter to anyone beyond Charlotte’s ever-energetic booster crowd? It should – and not just because the Queen City gained impressive bragging rights. It matters for reasons of community image. And it matters, too, because the MSA changes are just one example of the importance of context when it comes to analyzing data.

Nationally and locally, enthusiasm is blossoming for data-based decision-making. You can call it Big Data, or Smart Cities, or one of several memorable nicknames. The idea is this: With so much information available now, decision-makers can be using more analysis to inform their decisions.

A recent New York Times article “The Mayor’s Geek Squad: A Group of Number Crunchers Analyzes Troves of Big Data to Try to Solve the City’s Problems” highlighted one example of a city government using large troves of publicly available information in new ways. New York’s Office of Policy and Strategic Planning has used records from city departments to track such things as where restaurant scofflaws are letting grease clog sewer pipes, where stores are selling bootleg cigarettes and where fires are more likely to occur.

It’s not just governments. As the Times article put it: “Data – or Big Data, as quantitative analysts will call it – is the tool du jour for tech-savvy companies that have realized that lurking in the vast pools of unprocessed information in their networks are solutions to some of today’s most pressing and convoluted problems.”

StateTech magazine recently proclaimed: “Big data promises big savings in both energy consumption and budgets.” All kinds of innovations and possibilities are in the air. A nonprofit called Code for America is working to invent online and mobile phone apps called things like “Where’s My School Bus” and “BlightStatus.”

But as the newly redefined MSA boundaries illustrate, just having statistics or clusters of information is not enough. One thing you also need is context.

Geographers and demographics experts know that what the government considers the boundaries of a “metro area,” change. That can make comparing today’s data apples to yesteryear’s data oranges tricky. But I suspect that not everyone who’s deciding whether to invest somewhere or where to move understands those complexities.

And it’s not just comparing year to year metro area statistics. The lines carving the country into metropolitan statistical areas (from the federal Office of Management and Budget and based in large part on commuting patterns) sometimes seem to ignore what you may think of as obvious metro areas. The Raleigh-Durham-Chapel Hill area, known globally as the Triangle, is not one MSA. Raleigh’s Wake County, plus Franklin and Johnston counties, are one MSA. Durham and Chapel Hill’s Orange County are clumped with Chatham (Pittsboro and Siler City) and Person (Roxboro) counties as another. Go figure.

North Carolina’s Triad (Greensboro, High Point and Winston-Salem) is not one MSA. To the feds, it’s more of a duad, with Greensboro and High Point in a three-county MSA (Guilford, Randolph and Rockingham) and Winston-Salem part of another, five-county MSA.

So even a term like “city” can mean different things, depending on context. Consider: Is Charlotte a bigger city than St. Louis?

It depends.

The city of St. Louis has 319,000 people; the city of Charlotte has 750,000.

But that’s just within city limits. St. Louis is hemmed in. Charlotte for decades was able to annex its surroundings. Look at metro area population and the picture changes. The 15-county St. Louis MSA was 2.8 million in 2010, ranking No. 18, when Charlotte’s then-MSA was 1.75 million and ranked No. 33.

The point of all this is that analyzing data precisely can be tricky. I regularly see analyses that compare cities and metro areas, coming from groups ranging from the well-respected Brookings Institution to a fun – but possibly not so precise – “statistical” ranking of the “Top 10 Least Hipster Cities in America.” (Charlotte was sixth “least hipster.”)

So while this isn’t quite a “don’t try this at home” warning, it is worth saying to look closely at data. As Kate Crawford argues in “The Hidden Biases in Big Data,” “Hidden biases in both the collection and analysis stages present considerable risks, and are as important to the big-data equation as the numbers themselves.”

What’s being measured, exactly? Are apples compared to oranges? What is the context and does it matter? Is there an implication that one thing causes something else, when the information only shows the two things co-exist? Does whoever’s analyzing it all have a particular ax to grind, looking only at information fitting a preconceived opinion?

So as we welcome the possibilities of Big Data, I have to admit to a Big Hope: That we’ll also welcome Big Education About Data. Because even a simple question like “How big is your city,” can have a surprisingly tricky answer.

Views expressed in this commentary are those of the author and do not necessarily represent the views of the UNC Charlotte Urban Institute, its staff, or the University of North Carolina at Charlotte.