Not sure that these are markers of gentrifying neighborhoods -- just as an example, the big red area on your Austin map next to the University of Texas is largely student housing, and has always been this way: students have little to no income, often have roommates, and the parking policies of the university ensure that many students have few options other than to live near campus.

Likewise, the map of Philadelphia misses most of south Philadelphia, an area that is gentrifying extremely quickly with several hundred million dollar condo developments set to go up in the middle of the ghetto.

Interesting statistics, but I don't know that the conclusions that the authors are trying to draw are valid.

Actually UT heavily distorts the rental market, since while the students are poor, their housing is being paid for via loans or parents. So rents usually run higher than market rates near universities. This is why Universities frequently have expensive housing, but are unable to keep nearby stores in business.

All the images on this page aren't loading for me...

Google Cache:

http://webcache.googleusercontent.com/search?q=cache%3Ahttp%...

I think you're actually mapping areas where the market is being prevented from providing enough rental units for one reason or another.

or just where the data's bad. One of the highest ratios of median rent to median income (what they're trying to call their gentrification ratio) in the map of SF is shown in SOMA near 4th and Folsom. I used to live on that block, the median income is most certainly not $16k. It's just not. Half of that block group is non-residential, either Yerba Buena or office buildings, so maybe that's doing something weird to the data, but overall I think this attempt at a "gentrification" measure doesn't at all work.

I'd also expect walkable areas where people spend less money on cars to have higher rents compared to median incomes.

Take a look at Seattle: https://kwelia.com/maps/cbsa_census_tract?cbsa=Seattle-Tacom...

Looks like something is wrong with our median income data in Seattle. We're taking a look at it now.

Edit: we were missing the median income data for Washington state and a few other states. It's fixed now.

The Tampa Bay data has a weird outlier that it making the map kind of useless:

https://kwelia.com/maps/cbsa_census_tract?cbsa=Tampa-St.%20P...

Notice the one red spot at the top level, and everything else falling within the first 3 levels, which are 30% each. If the data is correct, perhaps the algorithm to pick the scaling factors could be improved to ignore outliers at the top and bottom?

These maps are really interesting, although I agree that whether they are actually measuring gentrification is debateable.

I would have found the site a lot easier to use if there were a big map view of the whole country where you could select from the supported regions. I was trying to see Mountain View, and finding the right setting in the drop down menu took a while (Mountain View and Los Altos are lumped in with "San Jose-Sunnyvale-Santa Clara" if anyone else is looking).

These maps don't work in Firefox :(

They aren't working in Chrome either. Is there some plugin that needs to be installed?

They worked in Chrome Canary for me. The pictures did not load correctly, but the actual maps that are linked to worked.

Another sign of gentrification: I'm seeing all these ad signs that they're buying cheap houses. These ads are getting more and more now. I say this is a sign because there are poor people that wants to move out, and others sees it as opportunity to flip the house for profit.

American Community Survey data is practically valueless. There, I said it. Symbolically, it may be interesting. When applied to cases like this, the conclusions one can draw are largely untenable.

Try the links- they work for me.