Probably because most people define 'urban' and 'suburban' in relative terms, and a neighborhood that might well be 'suburban' by NYC terms is 'urban' by the Seattle definition. Or a 3000sqft. house on an acre lot in Orlando can be 'downtown'.
Interestingly, if sidewalks measure "urbaness", most surburbs have extensive sidewalks and walking trails within a given development.
NYC in the example is very urban, but Staten Island is about 3x as dense as the (very suburban) suburb where I live -- and it gets very rural out here very quick (we're mostly converted farmland at this point). Staten Island seems fairly suburban to me, lots of single-story family homes with yards. It's almost the definition of a suburb except it happens to be administratively within NYC boundaries.
Manhattan is about 30x as dense as where I live and there's nowhere within 15-20 miles of where I live that even looks as urban as Manhattan. Mile after mile you get multi-story highrises or tiny squashed multi-story town homes that are mostly stairs. Even if you ignore the biggest skyscrapers, there's very few other places quite like it.
Part of it is that we're mixing two things that are a bit arbitrary.
How "urban" a city is and what the limits of a "city" are. Both of these are fungible by the community, as they should be. Urban living where I grew up would probably still include a backyard, in SF it probably includes a garage, in NYC it includes neither. The city of SF is pretty strictly defined, but the city of San Jose certainly has more fungible boundaries as well (santa clara vs milpitas vs san jose).
I am astonished at how much sidewalks cost per mile. We have engineered a system so complex that cities literally cannot afford to do much of anything.
I live in a rather nice area that has sidewalks only where new homes have been developed. There are huge stretches of roadway (some at 40mph) where riding on a bike equates to having a death wish. (Yeah I know bikes are supposed to be on the road, but at speeds above 20mph or so that is also complete bull!)
We have a mileage here where the city does all the sidewalk work. It's pretty awesome, no more tripping on cracks. I think it's like $15 a year for my $200,000 home.
It seems like the national consensus is that the threshold for "urban" is low enough to allow for a few more classifications of density which would probably be related to things like yards being squeezed out, single-family being replaced by high-rise, and presence of effective mass transit in spite of the American spite for such things. The term "downtown" would often fill such a role, though with even more variability than "urban". It's worth keeping in mind that our national capital doesn't even have any skyscrapers, and the term "urban sprawl" is widespread and widely understood to be something that actually happens with some frequency.
Sure, half the population is willing to give it lip service, but we're really good at coming up with excuses, barriers, and delays, and really bad at coming up with funding. Getting a new passenger rail line built is about as easy as getting a new nuclear reactor built.
Seattle has a huge city limits, but much of it is undeveloped parkland or somewhat industrial. I don't know of any populated neighborhoods that lack sidewalks.
Circa 2008, lots of areas around Greenwood didn't have curbs or sidewalks... everything north of 85th got annexed by Seattle in the 50's and none of it had sidewalks or curbs or storm drains. I bet most of the other sidewalk-less areas are similar annexations.
I dunno, Staten Island is pretty suburban. Same with parts of Queens. Hell, parts of Brooklyn have streets upon streets of single family homes with large yards that wouldn't look out of place in quaint Connecticut towns.
> Hell, parts of Brooklyn have streets upon streets of single family homes with large yards that wouldn't look out of place in quaint Connecticut towns.
I hear a tree even grows there!
No, that's really interesting. It would be amusing, more than anything, to try to define some kind of "absolute urban" by which all other locales can be measured for their urban-ness (or lack thereof... sub-urban-ness, I suppose) and, if found wanting, decorated with big-box stores and strip malls.
So... how urban is Midtown Manhattan compared to Calcutta or Dubai?
Dubai isn't dense, and Calcutta or Dhaka. Given Manhattan's extreme residential versus commuter population ratio (almost 2x), during the day Manhattan (if it were a city), may well be the densest in the world.
The sampling is wrong. Picking the largest municipalities isn't picking the largest cities; it's picking the regions with the best political organization. Jacksonville proper has about twice as many people as Miami proper, but you'd be insane to say Jacksonville is a "larger city": Miami's core is way denser and the metro area is four times as populated. Similarly, putting San Jose on the list instead of San Francisco doesn't make any sense. You could include northern San Mateo county and Oakland (which adds up to 1.5M) as "San Francisco" and you'd still get a higher ratio than in San Jose. Cities whose boundaries sprawl are concurrently more suburban and more populous: that's confounding.
A better list would be: New York, Los Angeles, Chicago, Washington, San Francisco, Boston, Philadelphia, Houston, Miami, Dallas.
Yeah, no one (at least here) thinks SF is a bigger and more important city than San Jose. And for all its wanna be bragging as a big city, San Jose is not the capital of Silicon Valley-at a minimum SV is diffuse and spread out throughout Santa Clara County, spilling out into the surrounding counties.
Hmm. Certainly no one should think San Francisco is bigger than San Jose. But more important? Certainly all my friends who live on the peninsula think SF is more important! SJ is not exactly renowned for its urban features.
> "How many people who live within a city's limits are really going to say they live in the suburbs?"
Tons:
> "Looking only at respondents in the larger principal cities (those with a population greater than 100,000) of larger metropolitan areas (those with a population greater than 500,000), the breakdown was 56 percent urban, 42 percent suburban and 2 percent rural. That means close to half of people who live within city limits describe where they live as suburban."
"City limits" is a really poor indicator of "urban".
Good point. This study seems like it would be biased by a region's perspective on city living. My guess is that people who want to live in the city will self-identify as urban much more often than those who want to live in suburbs.
The Houston graphic is interesting. Conroe is actually quite rural. The Woodlands, however, is shown as being a smaller area, but it's very densely populated, being one of the prime places for Houstonians to move to.
As others (and the article itself) note, first define a "city." Is it the arbitrary, and very political boundaries (of which each state, at least in the US has various and different rules how cities come to be, annexation, etc.)
For fun go google "world's largest cities". I guarantee you will get a different answer depending on the source, because each source defines a city differently. Of course there will be some overlap (Tokyo will be up there, along with a few other cities, but there is a rather large amount of variation). How can we not come to a consensus over something that seems so relatively easy to figure out? How difficult is it to count people in a city?
Again, what is a city? The political boundary? Is it the metro area? What type of metro area? The parts in which people commute to the center of "the" city for work? The surrounding counties that have economic ties to the "central" county? (This is roughly how the Census Bureau defines Metropolitan Statistical Areas, even though parts of the counties will have connections, but other parts of the counties will not). The urban area only? Pick whatever you want to call the city, just be consistent.
Let's say we go with a metro area, because it's more true of what a "city" is than the rather simple political boundaries. Which is most likely going to include if not other towns and cities, at least exurban areas that (economically if nothing else) act like an independent towns, and all manner of other interesting urban/non-urban forms of land use with people that all surround some sort of central place.
Let's say that you've defined your geographic area that constitutes a city.
Now let's get to the very difficult problem of defining the differences between urban, suburban and rural.
One of the more common answers to define "urban" is "population density," but then of course you must create a threshold for what constitutes high population and urban and what isn't. Are we talking about daytime population (workers) or nighttime population (residents)? Because there are still a major number of cities who's downtown will cease to exist after 5pm.
But defining what is center city/urban is actually not too terribly difficult, it's defining the edge cases that it becomes a difficult question. Where does the urban end and suburban begin? In most cities that are quickly growing, the old inner ring suburbs are now becoming quite urban in character. Old rural lands are now suburban. But where does one end and the other begin?
It's one of those things, like porn, you know it when you see it. You can probably point to one area and say "that's urban." "That's rural." But you (and me, and others) will all have slightly different definitions that while most will overlap, we will always disagree on the cases that could be both urban/suburban, suburban/rural or even (like parts of Detroit now) urban/rural.
If you can do all that, please give me a call, because congratulations, you're probably going to get nominated for a nobel prize in economics for precisely defining a city as well as coming up with a generalizable and replicable form of defining urban/suburban and rural.
That's one of those typical statistical fallacies. Depending on how you define 'urban' even Los Angeles (one huge blob of suburbia) suddenly becomes "Los Angeles — despite its reputation for sprawl — is 87 percent urban."
Your understanding of Los Angeles is incorrect. The average Angeleno lives in a denser neighborhood than in all U.S. cities except New York and San Francisco.
You can't just casually link to that page and claim it's talking about San Franciso. It's comparing the LA mega sprawl to SF-Oakland-Fremont. Fremont! The very definition of suburbia.
So yeah, if you head off from urban areas and include their low density suburban outliers, LA looks kind of sort of urban. But now I'm just repeating the grandparent's point: people who care about urban-ness are likely to consider both Fremont and LA suburban hell holes.
My point is that it's wrong to think of Los Angeles as the epitome of sprawl. It is not. It is an outlier of density compared to most American cities.
"So yeah, if you head off from urban areas and include their low density suburban outliers, LA looks kind of sort of urban."
This is explicitly the thinking I dispute. Look at the data. It is weighted density for each metropolitan area. It is a statistic used to eliminate the problem of arbitrary borders making data noisy.
I agree that it looks like a much more useful statistic than ordinary density, and is relatively insensitive to arbitrary inclusion / exclusion decisions of outlying areas like the grandparent complains about. Thanks for linking that site.
Glad you found it useful. The census bureau itself shifted to tract weighted density with the 2010 census, but I couldn't find a good city ranking on their site.
To be clear: I'm a fan of weighted density. It's better than simple density. It's just that it still suffers from selection effects. I think it's still just an unsolvable problem. Which suburban areas get glommed on to the dense core? Weighting reduces the impact of low density glommers, but if you glom on as many people as the urban core has from medium density suburbs, with a different culture and self identification, you get an inaccurate picture.
It's true that LA has some urban bits. And it's suburban bits aren't nearly as suburban as in other places.
It's more about the central problem of defining urban. To me, Fremont isn't even vaguely the same region as SF. SF has cultural attractions, Fremont does not. In Fremont, you need a car (I once walked 3 miles through Fremont to get to my bed for the night. I got a lot of bemused stares).
SF, and Boston, and Manhattan, feel like they're cities, and people who live in the outlying regions don't claim they live in the recognizable urban core. LA has a few dense zip codes, but everyone, including all the folks in relatively low density areas, feels like they live in LA.
LA has some of the densest zip codes in the country. If we look at the top zip code by metro, the ranking is NY, Boston, SF, Seattle (???), LA, Chicago, Washington.
This was an interesting visualization. Nothing wrong with it as long as you understand what is being measured, but I think it does leave out degrees of density in very urban areas. Chicago and New York are both 100% urban, and that does indeed tell you something, but there's still an interesting measurement to take above the threshold used here for what qualifies as "urban".
These maps show distinctions at very high density levels, so you'd see not only how Chicago differs from New York, but distinctions within these cities at a very high level of density.
33 comments
[ 4.3 ms ] story [ 73.0 ms ] threadProbably because most people define 'urban' and 'suburban' in relative terms, and a neighborhood that might well be 'suburban' by NYC terms is 'urban' by the Seattle definition. Or a 3000sqft. house on an acre lot in Orlando can be 'downtown'.
NYC in the example is very urban, but Staten Island is about 3x as dense as the (very suburban) suburb where I live -- and it gets very rural out here very quick (we're mostly converted farmland at this point). Staten Island seems fairly suburban to me, lots of single-story family homes with yards. It's almost the definition of a suburb except it happens to be administratively within NYC boundaries.
Manhattan is about 30x as dense as where I live and there's nowhere within 15-20 miles of where I live that even looks as urban as Manhattan. Mile after mile you get multi-story highrises or tiny squashed multi-story town homes that are mostly stairs. Even if you ignore the biggest skyscrapers, there's very few other places quite like it.
How "urban" a city is and what the limits of a "city" are. Both of these are fungible by the community, as they should be. Urban living where I grew up would probably still include a backyard, in SF it probably includes a garage, in NYC it includes neither. The city of SF is pretty strictly defined, but the city of San Jose certainly has more fungible boundaries as well (santa clara vs milpitas vs san jose).
I live in a rather nice area that has sidewalks only where new homes have been developed. There are huge stretches of roadway (some at 40mph) where riding on a bike equates to having a death wish. (Yeah I know bikes are supposed to be on the road, but at speeds above 20mph or so that is also complete bull!)
We have a mileage here where the city does all the sidewalk work. It's pretty awesome, no more tripping on cracks. I think it's like $15 a year for my $200,000 home.
http://m.metro-magazine.com/news/292617/poll-americans-favor...
I hear a tree even grows there!
No, that's really interesting. It would be amusing, more than anything, to try to define some kind of "absolute urban" by which all other locales can be measured for their urban-ness (or lack thereof... sub-urban-ness, I suppose) and, if found wanting, decorated with big-box stores and strip malls.
So... how urban is Midtown Manhattan compared to Calcutta or Dubai?
A better list would be: New York, Los Angeles, Chicago, Washington, San Francisco, Boston, Philadelphia, Houston, Miami, Dallas.
How many people who live within a city's limits are really going to say they live in the suburbs?
Tons:
> "Looking only at respondents in the larger principal cities (those with a population greater than 100,000) of larger metropolitan areas (those with a population greater than 500,000), the breakdown was 56 percent urban, 42 percent suburban and 2 percent rural. That means close to half of people who live within city limits describe where they live as suburban."
"City limits" is a really poor indicator of "urban".
For fun go google "world's largest cities". I guarantee you will get a different answer depending on the source, because each source defines a city differently. Of course there will be some overlap (Tokyo will be up there, along with a few other cities, but there is a rather large amount of variation). How can we not come to a consensus over something that seems so relatively easy to figure out? How difficult is it to count people in a city?
Again, what is a city? The political boundary? Is it the metro area? What type of metro area? The parts in which people commute to the center of "the" city for work? The surrounding counties that have economic ties to the "central" county? (This is roughly how the Census Bureau defines Metropolitan Statistical Areas, even though parts of the counties will have connections, but other parts of the counties will not). The urban area only? Pick whatever you want to call the city, just be consistent.
Let's say we go with a metro area, because it's more true of what a "city" is than the rather simple political boundaries. Which is most likely going to include if not other towns and cities, at least exurban areas that (economically if nothing else) act like an independent towns, and all manner of other interesting urban/non-urban forms of land use with people that all surround some sort of central place.
Let's say that you've defined your geographic area that constitutes a city.
Now let's get to the very difficult problem of defining the differences between urban, suburban and rural.
One of the more common answers to define "urban" is "population density," but then of course you must create a threshold for what constitutes high population and urban and what isn't. Are we talking about daytime population (workers) or nighttime population (residents)? Because there are still a major number of cities who's downtown will cease to exist after 5pm.
But defining what is center city/urban is actually not too terribly difficult, it's defining the edge cases that it becomes a difficult question. Where does the urban end and suburban begin? In most cities that are quickly growing, the old inner ring suburbs are now becoming quite urban in character. Old rural lands are now suburban. But where does one end and the other begin?
It's one of those things, like porn, you know it when you see it. You can probably point to one area and say "that's urban." "That's rural." But you (and me, and others) will all have slightly different definitions that while most will overlap, we will always disagree on the cases that could be both urban/suburban, suburban/rural or even (like parts of Detroit now) urban/rural.
If you can do all that, please give me a call, because congratulations, you're probably going to get nominated for a nobel prize in economics for precisely defining a city as well as coming up with a generalizable and replicable form of defining urban/suburban and rural.
(I could go on and on, but I won't but for extra fun, go read up on the Spersopolis. https://muse.jhu.edu/journals/sgo/summary/v035/35.2.hart.htm... )
Seriously now?
http://www.austincontrarian.com/austincontrarian/2012/09/the...
So yeah, if you head off from urban areas and include their low density suburban outliers, LA looks kind of sort of urban. But now I'm just repeating the grandparent's point: people who care about urban-ness are likely to consider both Fremont and LA suburban hell holes.
"So yeah, if you head off from urban areas and include their low density suburban outliers, LA looks kind of sort of urban."
This is explicitly the thinking I dispute. Look at the data. It is weighted density for each metropolitan area. It is a statistic used to eliminate the problem of arbitrary borders making data noisy.
I agree that it looks like a much more useful statistic than ordinary density, and is relatively insensitive to arbitrary inclusion / exclusion decisions of outlying areas like the grandparent complains about. Thanks for linking that site.
It's true that LA has some urban bits. And it's suburban bits aren't nearly as suburban as in other places.
It's more about the central problem of defining urban. To me, Fremont isn't even vaguely the same region as SF. SF has cultural attractions, Fremont does not. In Fremont, you need a car (I once walked 3 miles through Fremont to get to my bed for the night. I got a lot of bemused stares).
SF, and Boston, and Manhattan, feel like they're cities, and people who live in the outlying regions don't claim they live in the recognizable urban core. LA has a few dense zip codes, but everyone, including all the folks in relatively low density areas, feels like they live in LA.
http://www.zipatlas.com/us/zip-code-comparison/population-de...
LA has some of the densest zip codes in the country. If we look at the top zip code by metro, the ranking is NY, Boston, SF, Seattle (???), LA, Chicago, Washington.
I actually find this review more revealing:
http://beyonddc.com/?p=4808
These maps show distinctions at very high density levels, so you'd see not only how Chicago differs from New York, but distinctions within these cities at a very high level of density.