Recycling In New York City
Morgan Alcalay, Ronen Cohen, Nirothama Datt, Yoon Jung Kim

The NYC datamine offers a wealth of information about New York City.  One interesting data set contains information about recycling rates throughout the five
boroughs.  Careful analysis of this data reveals many interesting facts about recycling trends.  The goals of our analysis was to determine differences in recycling rate and eco-friendliness across boroughs, community districts, and income levels, as well as whether there are any differences in the recylcing rates of paper (which doesn't have a redemption value) and bottles and cans.  

The first two pie graphs show the individual city-wide recycling rates for both paper and bottles and cans.  The figures represent the percentage of materials recycled.  As you can see, the rate for bottles and cans is 10% higher than it is for paper.  We theorize that this difference is primarily due to the redemption values for bottles and cans.  People have a monetary incentive to recycle these materials, whereas there is no incentive for paper.

Paper recycling ratesBottles and Cans Recycling Rate
The next bar chart shows the differences in average total recycling rates for each borough.  There appears to be a large amount of variance - more than 20% - between the most eco-friendlyStaten Island and least eco-friendly Bronx.

Recycling Rate by Borough

We hypothesized that this difference between the boroughs might be related to differences in income levels.  To test this, we first needed to see whether there is actually a relationship between income level and recycling rate.  To do this, we first joined the community district and recycling tables, created a query to return community district, average recyling rate, and median income, and then displayed the information graphically.  The results confirmed our suspicions.

Recycling Rate Versus Median Income

There appears to be a positive relationship between income and recycling rate.  This relationship can be formally tested with Excel's regression data analysis tool.  The output is reproduced below:
Regression Output
With an R-squared of 44.55%, nearly half of the variance in recycling rates can be explained by the differences in income.  The income variable coefficient means that every $10,000 increase in median income is associated with 5% increase in the recycling rate.  Finally, we constructed a chart of the averaged median incomes across boroughs, to see if there is sufficient variance income to explain the differences in recycling rates.
Average income by borough
The data roughly supports our hypothesis.  Other than the switch between Manhattan and Queens, the average income graph shows the boroughs in the same order as in the recycling graph.  It is important to keep in mind, however, that correlation does not prove causation.  Both income and recycling rate might be significantly affected by a third variable which we have not examined.  

Data used in this analysis can be found at the NYC data mine website.  Two database files were used (and joined for that matter).  The first, Demographic, Social, Economic, and Housing Profiles by Community District can be downloaded here.  The second, Recycling Diversion and Capture Rates can be downloaded here.

If you want to find out more about community districts in New York City, please visit the New York City Department of City Planning's website.  More information about recycling can be found at the Department of Sanitation website.

Contact Information
If you have any questions, please email the authors at:
Morgan Alcalay
Ronen Cohen
Nirothama Datt
Yoon Jung Kim

Return to main page.