Cherry Picking

I was trying to be true to the ideals of RATIONAL urbanism last week while looking into the gun issue. Of all things it was a Facebook meme (Oh, Richard Dawkins, what thou hath wrought!) intended to be anti gun-control which forced me to dig a bit deeper into the issue. The posting in question compared Houston and Chicago and showed a series of facts about the two cities with the three major differences being the ease with which one could purchase a gun, the number of gun deaths, and the average daily temperature which of course leads the meme’s creator to conclude that “liberals” must believe that “cold weather” causes gun violence.

I’m sure many of us are amazed at the idiocy of things we see on social media, but having studied general levels of violence in the United States, by city, by region, by state, and overall, for decades, I knew that we not only had a fairly straightforward case of confusing correlation with causation, but I was also somewhat certain that this was also a case of cherry picking. Using Texas as an example of a place which has gotten to the bottom of the problem of violence and which should be seen as a model for the rest of the United States would be as ridiculous as doing the same in public education! (Oh, wait, we DID do that with NCLB only to find out that ALL of the supposed advances in education achieved in Texas were based on faulty, manipulated, and deceptive data…but we continue to go down that same road!)

Of course, cherry picking of data is one of the easiest ways to try to make a case for something, but it is also the most obvious demonstration that instead of being on a “quest for truth”, you are merely attempting to prove that you are right. What this anti gun-control zealot did with the Houston-Chicago comparison, pro gun-control activists could do using state to state comparisons between, perhaps, South Carolina and Vermont. You can see this same technique used when addressing the death penalty, welfare, and even abortion. I am absolutely certain that I make mistakes all the time. In the social sciences, in the study of cities in this case, it is impossible to know, let alone calculate, all of the variables. What I can do, is to do the best I can to seek out the data which seems the most significant, and which is the most compelling and use that to formulate hypotheses.

Confusing what we WANT to be true, with what IS true seems to be normal for most people. I happen to be an atheist because I see no compelling evidence for the existence of a god or gods. It isn’t that I know that there is no god because, of course, once someone posits the idea of an all powerful deity I must acknowledge that an ALL powerful being would have the power to exist without an Internet blogger knowing about it…perhaps he/she/it is hiding behind one of the rings of Saturn just yuck yuck yucking it up that “this guy” can’t see him/her/it! Atheism is a question of belief, not knowledge, and I am an atheist. In the absence of any compelling argument for the existence of a god, many people have expressed to me that they believe, even KNOW, that there is a god because they “wouldn’t want to live” in a universe where there was no god. What is amazing is that people really think that THAT is an argument for the existence of god.

On a podcast for the Center for Inquiry, “Point of Inquiry” Chris Mooney, normally the interviewer, but in this case the interviewee, discussing his new book on science and faith was asked about the minority of scientists who have a god belief and whether or not, in truth, it was because these scientists were able to compartmentalize their rationality and were obviously selectively irrational vis a vis their god belief. Mooney responded, as if it were an argument, that these scientists wouldn’t WANT to be told that they were only selectively rational!

What we WANT to be true has no bearing on what is true. The universe is under no obligation to make us happy, or conform to our previously held beliefs. The belief in “karma” in any sort of supernatural sense is as infantile as a belief in Santa Claus, the Easter Bunny, leprechauns, unicorns, or god. I would have preferred, I think, that forced integration, “busing”, be effective. It not only hasn’t been, it has made the specific problem it was created to address worse and has probably done more harm to cities than any other policy initiative in the last 50 years and THAT is saying something. But that isn’t because I want that to be the case, it is because the evidence shows that it is the case.

Anyway, apart from offending any “woo woo” readers I might have had up to now with my philosophical materialism and revealing six or seven topics for future blog posts what is the point of this screed? To reveal one particular major regional difference between what we call “cities” here in the United States. It isn’t that I wasn’t aware of this before, but it is that I was not aware just how enormous the regional variations were and therefore wasn’t as conscious as I should have been of its significance. What is this major component of statistical data which potentially skews the conversation every time a national urban issue is raised? The relative geographical size, and therefore the demographic make up, of what we call a “city”.

If you already have an opinion, and you don’t really care whether or not it conforms to reality, then understanding the data isn’t of much importance. You just want to use whatever it says to make your case. If, however, your opinion, even having an opinion, is contingent upon evidence, then truly comprehending the data, knowing what it says and what it doesn’t say, what it demonstrates and what it doesn’t demonstrate is exactly what the data is for. For me the subjective elements of this blog and website are that I prefer city living to suburban, rural, and wilderness living, and I want my mid sized northeastern city and others like it to thrive. Apart from those attitudes which are purely personal, the purpose of my efforts here is to augment and improve my understanding of the urban condition in 21st century America.

Look at this list of “mid sized” northeastern cities, their population, and the size of their footprint.

Springfield, Massachusetts: 155,000; 33 square miles
Worcester, Massachusetts: 182,000; 39 square miles
Bridgeport, Connecticut: 145,000; 19 square miles
Albany, New York: 98,000; 22 square miles
Rochester, New York: 211,000; 37 square miles

Now the Midwest

Toledo, Ohio: 286,000; 84 square miles
Springfield, Missouri: 160,000; 74 square miles
Peoria, Illinois: 115,000; 50 square miles
Madison, Wisconsin: 236,000; 85 square miles
Grand Rapids, Michigan: 190,000; 45 square miles

The Southeast

Birmingham, Alabama: 212,000; 152 square miles
Jackson, Mississippi: 176,000; 112 square miles
Tallahassee, Florida: 183,000; 103 square miles
Knoxville, Tennessee: 181,000; 98 square miles
Baton Rouge, Louisiana: 230,000; 79 square miles

The West

Boise, Idaho: 210,000; 64 square miles
Flagstaff, Arizona: 66,000; 98 square miles
Stockton, California: 296,000; 62 square miles
Reno, Nevada: 228,000; 69 square miles
Spokane, Washington: 210,000; 58 square miles

Do you see it? The midwest and the west clearly form a baseline of sorts, but the southeast and northeast are clearly different. I selected these cities at random, basically thinking of mid size, “2nd tier” or 3rd tier” cities in these regions and looking up their statistics on Google. The reasons for these regional variations aren’t of any particular interest to me right now, although I’m sure they are easily explained. The point is that when we compare cities to cities as we cross the United States, we are sometimes comparing apples to watermelons. The federal government accommodates what it can of these variations in their “Standard Metropolitan Statistical Areas” for just this reason, but often when statistics are compared by journalists or in magazines to create lists of “best of” or “worst of” what we see compared are the cities themselves.

If what we want is to really understand how these places stand up in questions of quality of life, policy decisions, economic advancement, and whatever else, it may be important to understand that, region by region, we really aren’t talking about the same animal, even when we use the exact same word. It may even end up telling us that these regional variations themselves have an impact on outcomes and that cities in the northeast should swallow up surrounding communities, or that cities in the southeast should devolve into separate communities, the case here isn’t for or against a particular standard, it is simply for an awareness that these divergences exist.

To think for just a moment about just what changes in our perspective understanding this data might cause. Take Jacksonville, Florida (Please!!), it has a population of around 800,000 people…but spread over almost 900 square miles: It is the largest “city” in the United States, over two thirds the size of Rhode Island. As a city of nearly a million people it has, for example, an NFL football team and views itself as a major league city. Springfield, Massachusetts has a population of just over 150,000 people spread over 33 square miles, but the metro Springfield area has nearly 700,000 inhabitants, with a much higher average income than Jacksonville, but would never think of itself as fertile ground for an NFL franchise or as any kind of a major player as places go in the United States.

Jacksonville

Greater Springfield at the same scale

What about federal money for cities? Often cities below the 150,000 threshold aren’t eligible for certain “monies”. It would be easy to envision an effectively less urban community in the south getting preference over an agglomeration of communities in the northeast which, but for these differences might have been preferred. In the case of the anti gun propaganda on Facebook which sparked this realization what is obvious is that Houston and Chicago are actually very dissimilar in ways which alter the impressions of the impact of gun control or lack thereof on urban violence and that is that what is nominally “Houston” is, in truth, what would be in the Midwest both Chicago and its inner rings (at least) of suburbs. (Houston, 627 square miles, population 2.1 million, Chicago, 234 square miles, 2.7 million population), the lower rates of gang and drug violence of the enormous Chicagoland would alter the calculation significantly. Greater Chicago has a population of nearly 10 million people, greater Houston, just over 5 million…and therein lies the entirety of the (flawed) argument of the Facebook meme! It actually looks as though Chicago and Houston are almost identical in terms of gun violence, in spite of the fact that their respective gun laws are so different.

What that probably means is that, in a country where freedom of movement is the norm, the only gun laws which could have a significant impact on gun violence would need to be national in nature, an unfortunate fact perhaps, given that it is reasonable to suspect that meaningful nationwide gun legislation will be impossible to achieve without creating horrible social unrest in much of the country.

For the purposes of this essay, it’s enough to show that, in the United States, a city is not a city is not a city. What makes up a city in one region can be very different from what constitutes a city in another. Good to know.