Median Rent SF: One Bedroom Apartment Listings

Click the picture above to get to the map. It shows median rent for a one bedroom apartment in the second half of 2012. You can see the median rent for a San Francisco neighborhood, as well as the nuance of where the pricey stuff is clustered, plus what you *should* be making to live there, based on the assumption that rent is 1/3 of your income.

Here is why my analysis of devastatingly high housing prices in San Francisco is better than all the others that came out in the past two weeks:

1. I used a dataset that spans a longer time range than one month, so that the sample sizes per neighborhood are larger. My data spans the second half of 2012.

2. I filtered for duplicates, so that the same apartment listed multiple times does not skew the data.

3. I used median instead of mean, which makes sense when you are dealing with a lot of outliers. You should be suspicious every time you see some shocking statistic about the “average rent” in San Francisco. The high end skews things quite a bit.

4. The fact that I am even explaining my methods and assumptions.

5. Instead of a silly choropleth, I made a super-pretty heat map based on the spatial variability of listing prices. The interpolation is based on ordinary kriging, and I underlay it with contours to bring out the variation a bit more.

As with several previous projects, I designed this map in Tilemill and am hosting it via MapBox, and overlayed it on a hard working OpenStreetMap. For the raster processing I used ArcGIS and QGIS. The data was kindly donated by PadMapper.com.

Pop Quiz: Can you spot the new development in Bayview? What about the Tenderloin? Can you see where that is??