This recent study used geotagged tweets to measure social connectedness within American cities. There are two measures: (1) concentrated mobility and (2) equitable mobility. The first measures the extent to which social connections (geotagged tweets) are concentrated in a set of places within the city. And the second looks at the degree in which people move between neighborhoods in roughly similar proportions. These measures are the y-axis and the x-axis, respectively, in this graph:

So how do you read this chart?

Well if you look at New York, you'll see that it is relatively high in concentrated mobility, but the lowest in terms of equitable mobility. This means that social connections are highly concentrated and that there's low connectedness to other neighborhoods within the city. Miami, on the other hand, is the opposite. It's also an outlier. Few hubs. But its social connections appear to cross neighborhoods and spread across the city.

Perhaps not surprisingly, the study found that the size of a city seems to have the biggest impact on social connectedness. Which makes sense -- it becomes harder to get around and so people start to localize. I am reminded of this whenever my friends in Los Angeles tell me they never go to the beach because it's simply too difficult and too time consuming to get across the city.

This also became clear to me after I started playing around with the Moves App back in 2015. The app no longer exists, but it was an activity tracker that allowed you to map where you, well, moved. And the more time you spent in one place, the more concentrated the activity would become. They depicted this through larger and larger circles. Example maps, here. My maps revealed that I need to branch out into different neighborhoods more often.

To download a full copy of the study, click here.

Chart: CityLab

Richard Florida is currently running a four-part CityLab series on the economic performance of America's cities. What makes this study somewhat unique is that it looks at cities proper, rather than at their larger metro areas. In some cases there may not be that much of a difference. But in other cases, the performance of the city proper could be very different from that of the broader area.

Here are the fastest and slowest growing cities from 2012 to 2017:

Here are the fastest and slowing growing job markets:

And here is the growth in share of adults with a graduate degree:

It's interesting to see Seattle at the top of the population growth list. It is not a sprawling sunbelt city. It is an expensive tech hub. And it is also interesting to see Miami's strong employment and education growth. Years ago, Paul Graham wrote an essay arguing that tech hubs have two prerequisites: capital and nerds. He went on to argue that Miami has lots of the former, but not much of the latter. Maybe that's changing.

Garrett Dash Nelson recently published a study looking at urban density on a cell-by-cell basis for a number of US cities. Each "cell" is a 30 arc-second grid cell, but you can think of them as being approximately one square kilometer. The goal of the project was to better define urban density and do it in a more granular way. City averages don't tell you a whole lot about how neighborhoods vary, and they can be skewed by the denominator you use. i.e. Where are you drawing the urban boundary?

You can play around with his interactive study, here. Each city can be explored according to its 200 most dense cells. One interesting takeaway -- though it is probably not all that surprising to this audience -- is that New York City is really a unique place when it comes to American cities. If you look at the above chart (sourced from CityLab), you'll see that most other US cities don't come close to it in terms of urban density. New York's 200th densest cell is still denser than the most dense cells of Boston, the Twin Cities, and of Dallas.

The y-axis is the total population in each grid cell.