Parks & Recreation… and Data

June 12, 2013

We are psyched to present to y’all a guest blog post from the inimitable Brian Dalessandro. For those of you that may not have been at our NYC Parks DataDive this past fall or at our last NYC meetup, Brian worked with the NYC Parks department as a DataKind Data Ambassador to figure out how their block pruning programs were working.  Lucky for us he wrote about his experience throughout this project, and we’re happy to share this great data-for-good journey with you all. Enjoy!

—-

When my former boss was retiring, he gave me some simple career advice. He said, “be sure to stop and smell the roses.” For some reason, I was expecting something more profound — after all, here was a man who had led such a successful career. It wasn’t until recently that I truly grasped the depth of his message.

I recently started volunteering as a Data Ambassador for an incredible organization called DataKind. Their mission is to match talented data scientists with non-profit and government organizations that have huge troves of data but lack the resources to analyze it. As a Data Ambassador, I was matched up with NYC Parks. Not only are they the ones to thank for the fabulous upkeep of Central Park and all the other glorious city parks, but they’re also responsible for each and every tree that brings shade, color and oxygen to our concrete-heavy metropolis.

Years ago, NYC Parks created a program for taking better preventive care of the city’s trees. This program involves a regular schedule of pruning and grooming large trees in an effort to reduce the risk of damage from storms and high winds. For years, the department kept a record of which blocks were pruned, as well as how many times they had to dispatch a crew to remove fallen branches and upended trees. Armed with all of this data, they approached DataKind with the following question: “Does pruning trees in one year reduce the number of hazardous tree conditions in the following year?”

If you are a regular reader of the m6d blog (admit it, you are), and/or a savvy advertiser, you might recognize that NYC Parks is asking a fundamentally causal question: “Does pruning cause a reduction in hazardous tree problems?” As a data scientist for a digital advertising company, I’m accustomed to fielding this type of question. A couple of years ago, we blogged about how we were able to estimate the causal impact of ads by analyzing our impression logs. When I signed up with DataKind, I naturally gravitated to this project, given both my desire to help the city I love and my experience analyzing these type of problems.

The project started with a few hours devoted to downloading, cleaning, merging and analyzing data. With the blessing of m6d, I was able to use our high-powered server infrastructure to run some intensive modeling, which produced some very interesting results. I found that pruning the trees at risk for certain types of hazards caused a 22% reduction in the number of times the department had to send a crew for emergency cleanups.

So not only did NYC Parks’ program have a positive impact, now they were armed with a tangible result to provide justification for past and future investments in their data infrastructure and data collection efforts.

m6d is a company that is fueled by data. I spend every day pumping this fuel through our system and I forget that it is kind of a luxury to be able to do so. But there are countless non-profit organizations that simply don’t have the resources that come standard here. Thanks to a facilitator like DataKind, the gap between the two can now be bridged. Although corporations often donate money for good causes, it’s often more rewarding to be able to donate human resources to an organization that can truly benefit from it.

I analyze data all day, every day. But this project was special. It reminded me how important it is to stop and smell the roses. After all, NYC Parks practically demands it.

 

by Brian Dalessandro