March 17, 2014
The Huge Correlation between Median Income and Recycling in NYC

A few weeks ago, the city announced the addition of a new regularly updating dataset: Monthly Tonnages, from the Department of Sanitation. The data shows the total tons of residential garbage and recycling hauled away each month, broken down by Community District.   I’ve seen some sporadic data on this in the past, but it’s great to see it updated monthly and I look forward to seeing how the numbers change over the coming years.  

With just a few months released thus far, it’s difficult to measure any changes over time. So I decided to explore varying recycling rates across the city for the month of February, 2014.  To do that, first I made a map of the percent of total refuse that was recycled in each Community District (the level of granularity released by the city):


The map shows that the relatively affluent district of Brooklyn 06 leads the pack on percent of all garbage that is recycled, coming in at 30%, (Carroll Gardens, Gowanus, Cobble Hill, Park Slope, Red Hook, South Slope).  The lowest rate was 8%, seen in Bronx 01, (Mott Haven, Port Morris and Melrose). 

The map led me to believe that there may be correlation with median household income, which I show on this map: 


The maps don’t match exactly, but they certainly seem correlated. 

To explore the relationship between income and recycling rates, I made a scatterplot of median income per district and the recycling rate:


(The Bronx is Red, Staten Island is White, Brooklyn is Orange, Queens is Yellow and Manhattan is Black in the above plot.)

The recycling rate and median income are incredibly correlated.  In fact, the correlation coefficient (r) is 88%!  I’ve been looking at city data for years, and rarely do you see two datasets, which measure such different things, have such a high correlation.

It’s important to note that this is a correlation, not a causation, and so we can only speculate as to the reason.  Among other possibilities, It may have something to do with the type of garbage created in each neighborhood, or it might be about awareness of what can be recycled. And remember, it takes time to recycle, and time is money.  Either way, if the city wants to increase the recycling rate, this sort of analysis might help it target where it may have the biggest impacts. 

Maps done in QGIS.  Analysis/Plot done in IPython. CD Income data from OpenData, and is as of 2008.  Garbage data is from Open Data and is as of Feb 2014.   

Filed under: nyc opendata income recycling qgis DSNY 
  1. medeasdragons reblogged this from iquantny
  2. erikct1 reblogged this from iquantny
  3. nagler reblogged this from iquantny
  4. iquantny posted this