• SR_spatial tweets

Proposed Congressional districts for NYS available in GIS format

UPDATE June 25, 2012

We launched a companion map featuring Congressional districts with statistics on eligible voters by race/ethnicity compared with total population.

UPDATE March 6, 2012

We’ve added Congressional districts as proposed by District Court Judge Hon. Roanne Mann to our interactive redistricting site. Here’s a link that compares District 9 (Rep. Turner, in NYC) with one of the proposed districts that it would become under her proposed lines: http://t.co/01K4hMu8

We also submitted a letter today to the court [PDF] suggesting that they can use our maps to visually compare the different proposed lines.  Hopefully they’ll put our online maps to good use as they review the different Congressional district proposals.


UPDATE March 5, 2012

We’ve made two updates the information below.

  1. We’ve added the Congressional district data in shapefile and KMZ formats based on Common Cause’s submission to the court.  We think this will be especially helpful since the court has asked the intervenors to compare their maps with Common Cause’s proposal.
  2. Now you can visualize the proposed districts based on the mapped data below at the Center for Urban Research’s interactive redistricting site.
    1. compare with existing Congressional districts;
    2. easily switch among the Congressional proposals from Common Cause and the Senate & Assembly majorities; and
    3. view the proposed districts in relation to block-level demographic maps (do any of them appear to “pack,” “crack,” or dilute the potential voting power of minority populations?) or local voting patterns (click the “More Data” tab at the bottom right).

Here are some examples:

Today the New York World posted an analysis of how these different Congressional district proposals might impact Rep. Charles Rangel’s current district 15.


Original Post

If you’re hoping to use GIS or any of the online mapping tools to map the Congressional district lines in New York State that were proposed late yesterday, you’ll have some work to do.  The maps were released in PDF format as well as “block assignment lists” for the proposed districts.

But if you’d like to use shapefiles and/or KML files, you’ve come to right place!  Our team at the CUNY Graduate Center has created them and posted them for downloading here:

http://www.urbanresearch.org/news/proposed-congress-districts-in-gis-format/

We hope to add these soon to our interactive redistricting map. Stay tuned!

NYC’s open data legislation: reading between the lines

TL; DR (i.e., the summary)

NYC is about to adopt what some are calling “landmark” and “historic” legislation regarding open data.  Does the hype match the reality?

I offer the analysis below not as a critique of the City Council.  I think they probably tried to negotiate as good a bill as they thought they could achieve.  I offer it more as food for thought for those of us who will be seeking the data that may eventually become available because of the legislation (and for those of us who rely on data that’s currently available that may become less so due to the bill).

Hopefully my concerns represent a worst case scenario.  If the bill’s implementation indeed lives up to the “landmark” status bestowed on its passage, that would be a great thing.

For example, the Council’s committee report on the bill [Word doc] suggested that substantial city data sets such as the Building Information System (BIS) or the Automated City Register Information System (ACRIS) would be made available in open, accessible formats due to the legislation. If that happens, that would be great.  But for each of the handful of examples like that suggested at yesterday’s Council committee meeting, I could offer several more that I believe might escape the requirements of this bill.

My overall sense is that somewhere during the two-plus years the bill has been on the table, the details got in the way of the original vision embodied in this proposal.  And, as they say, the devil is in the details.  If you’re interested in my take on those gory details, please read on.


An important step

The bill is important, in a way. It’s an acknowledgment by the City Council (and the Mayor, if he signs it) that city agencies need to provide public access to data sets online, in a standardized electronic format.

In doing so, it goes a step beyond FOIL — the New York State law since the mid-1970s that has required agencies (including local government) to provide public access to data.  Though FOIL has adapted to the times to some extent — the courts and policymakers now understand that FOIL applies to electronic data as well as printed material — it is still a reactive approach.  You have to submit a FOIL request (and have a good idea of what data you’re requesting) for an agency to respond and give you access.  New York’s Committee on Open Government describes it as “pull” vs. “push”. [PDF]

Some smart agencies have realized that posting data electronically saves money, time, and effort. By posting data online proactively, before the agency even receives a single FOIL letter  (“pushing” it so people don’t have to “pull” it), it avoids having to respond individually to FOIL requests.

So the City Council bill acknowledges that pushing is better than pulling.

Those devilish details

But will the legislation require agencies to post data online?  To some extent, yes.  But how far that goes depends on how it’s interpreted, and how aggressively it’s implemented (and perhaps how strongly the public reacts, since it seems like the only enforcement mechanism is public reaction).

The first substantive part of the bill says that within a year, agencies need to post their data at the city’s online data portal.  But let’s look closely at the language.  Section 23-502(a) says that within a year, agencies don’t need to publish all their data to the portal.  Only “the public data sets that agencies make available on the Internet” need to be included in the portal (emphasis mine).

In other words, if an agency has refused to provide public access to a data set, or perhaps only allows access to that data after you’ve paid a fee and/or signed a license agreement, or otherwise hasn’t already posted the data online — that data is exempt.

Then it gives agencies another loophole.  The next sentence says that even if an agency has a data set online, it doesn’t need to post it on the portal if they “cannot” put it on the portal.  (“Cannot” isn’t defined in the bill.  Does it mean “doesn’t want to”? Does it mean the data’s too complex for some reason?  “Cannot” seems to offer quite a bit of wiggle room.)

The bill further states:

the agency shall report to the department and to the council which public data set or sets that it is unable to make available, the reasons why it cannot do so and the date by which the agency expects that such public data set or sets will be available on the single web portal.

I’m not a lawyer, but it seems to me that if an agency doesn’t want to comply, it just needs to give a reason.  And it needs to give a date by when it will add the data to the portal.  The date could be two years from now, or it could be two decades from now.  That part of the bill doesn’t have a deadline.

Without aggressive support from the top — the Mayor and/or perhaps a new Chief Data Officer position with some teeth — agencies could just take their ball and go home and not play the open data game.  And the public will be the worse for it without much recourse.

Over-reliance on “the portal”

Let’s be optimistic and assume that all city agencies (even the current holdouts – I’m looking at you, City Planning Department & MapPLUTO) decide to post their data online.

The bill doesn’t say, or even mention as an option, that agencies can keep posting the data online at their own websites.  Instead, it has to be posted on “a single web portal that is linked to nyc.gov”.

But I’m not as enthusiastic as I once was for the portal approach (currently implemented here).

  1. Data for APIs, or people?

At first I thought the portal would be so much better than the city’s earlier Datamine site. But the site seems to focus heavily on APIs and web service access to the data, which might be great for programmers and app developers, but not so good for people, like Community Board staff, or reporters, or students, or anyone else who just wants to download the data and work with the files themselves.

  1. Some agency websites are doing a better job

Also, why not allow — even encourage — agencies to continue posting data on their own websites?  I think that, in many instances, the individual agencies are doing a better job than the data portal. The files available for downloading from agency sites such as Finance, City Planning, Buildings, and Health are more up to date, more comprehensive (though still hardly complete), and easier to understand than what I can find on the portal.

I think it would be ok if both approaches existed (portal and individual agency sites). But the way the bill is worded, I think the risk is that agencies are more likely to do only what they have to do or what they’re expected to do.  Since the bill focuses on the portal, I think we may see individual agency data sites whither away, the rationale being why bother with individual sites since they have to post to the portal.  With sites such as City Planning’s Bytes of the Big Apple (which is really great, with the exception of the PLUTO license/fee), I think that could be a big loss for the many people and organizations who have come to rely on the high quality data access that these agency sites provide.  Hopefully I’ll be proven wrong.

  1. The current portal falls far short of a forum for public discussion

The bill requires DoITT to

implement an on-line forum to solicit feedback from the public and to encourage public discussion on open data policies and public data set availability on the web portal.

But if the current portal is the model for this online forum, I’m concerned.

When I access data from the agencies themselves, I can talk with the people directly responsible for creating and maintaining the data I’m seeking. I can have conversations with them to understand the data’s limitations. I can discuss with them how I’m planning to use the data, and if they think my expectations of the data are realistic.

In contrast, the portal requires me to either go through a web form (which I’ve done, and received zero communication in return), or to contact someone who has no identification beyond their name (or some online handle).  Do they work for an agency?  Do they even work for New York City?  I have no idea; the portal provides no information.  So much for a site that’s supposed to be promoting “transparency in government.”

To me, the portal is somewhat analogous to the city’s 311 system and the recent articles about putting the city’s Green Book online.  Though 311 is great in a lot of ways, it has put a wall between the public and individual city agency staff members.  Try finding a specific staffperson’s contact information via nyc.gov, like the New York Times recently did.  It’s almost impossible; you have to communicate through 311. Similarly, the online data portal — if it ends up replacing agency websites as sources for online data access — will make it difficult to locate someone knowledgeable about the data.

This widens the “data gap” — the gap of knowledge between data creators and data users.  In order to know whether a particular data set meets my needs (if I’m creating an app, or even just writing a term paper), sometimes a written description of the data is not enough.  I may need to actually talk with someone about the data set.

But good luck finding that person through the data portal.

And even when people have used the portal to submit online comments, I don’t know if anything ever comes of it.  It looks like only 14 of the 800+ datasets at the portal have comments (sort the list by “Most Comments”).  All of the comments raise important questions about the data.  For example, two people offered comments about the HPD Registration data available through the portal.  They asked “Is there any plan to expand it?” and “Could you help us?”  Both remain unanswered.

Maybe everyone who commented was contacted “offline”, as they say.  Either way, this hardly constitutes a forum for public discussion.  No public interactivity.  No transparency.  No guidance.  It’s no wonder there’s been so little use of the portal’s  button (and I use the term “Discuss” loosely).

Public data inventory

Another section of the bill has a nugget of hope.  But the way it’s worded, I’m not too optimistic.

Section 23-506(a) says that within 18 months, DoITT shall present a “compliance plan” to the Mayor, the Council, and the public.  Among other things, the plan must “include a summary description of public data sets under the control of each agency.”

In effect, this “summary description” (if it’s done right) will be the public data inventory that advocates have been pushing for (and which has been required by the NYC Charter since 1989). That’s a good thing. At least now we’ll know what data sets each agency maintains.

Hopefully it’ll be a comprehensive list. I guess the list’s comprehensiveness will be up to DoITT to enforce. (And if the list comes up obviously short, perhaps some enterprising FOILers can point out — very publicly — where the holes are 😉 ).

But that same section of the bill also says that the plan “shall prioritize such public data sets for inclusion on the single web portal on or before December 31, 2018“.  So it still relies solely on the data portal. And it gives the city another 6 years to make the data public. As someone said on Twitter, “sheesh”!

Then there’s another loophole.  The bill allows agencies to avoid meeting even the 2018 deadline by allowing them to

state the reasons why such [public data] set or sets cannot be made available, and, to the extent practicable, the date by which the agency that owns the data believes that it will be available on the single web portal.

“[T]o the extent practicable”?  When the agency “believes” it’ll be available?  Wow.  Those are some loose terms.  If I ran an agency and didn’t want to provide online access to my department’s data, I’d probably feel pretty confident I could continue preventing public access while easily complying with the law.

Where does this all leave us?

It looks like the City Council will pass this law, despite its limitations.  In fact, DoITT was so confident the law will pass, it emailed its February 2012 newsletter on the day the Council’s technology committee voted on the bill (Feb. 28, a day ahead of the expected full Council vote).  Here’s what the newsletter said about Intro 29-A:

“Will be voted on and then passed”?  I guess the full Council vote is pretty much a foregone conclusion.

That leaves us to hope that the bill’s implementation will address the issues I’ve outlined above, and any others that advocates may have identified.  Fingers crossed?

(Disclaimer: my viewpoints on this blog are my own, not necessarily my employer’s.)

Redistricting’s partisan impacts: a GIS analysis

Our team at the Center for Urban Research is collaborating with The New York World to analyze the impacts of redistricting in New York State.  The latest effort was featured today on the front page of the Times Union; it focuses on how the majority parties in the State Senate and Assembly would likely retain — and strengthen — their control of both houses through the redrawn district lines.

Briefly, we found that the new boundaries for state Senate and Assembly districts proposed by LATFOR would increase the number of seats held by the majority parties in both chambers.  We based the analysis on 2010 election data available from LATFOR’s website.  The goal was to determine the results of state legislative elections held within the new districts if voters cast their ballots in the exact same way as they did in 2010, the most recent election year for State Senate and Assembly.

  • In the State Senate, the Republican Party’s 32-to-30 majority would expand to 34-to-29 if each voter cast his or her ballot in support of the same party as in the 2010 elections.
  • In the State Assembly, the 98-to-50 advantage the Democrats enjoyed following 2010’s elections would also increase, to 102-to-48.

The project was a good example of the power of GIS.  The analysis didn’t necessarily need a map to display the results (though Michael Keller at the NY World put together a nice one). But the analysis effectively wouldn’t have been possible without GIS.

Converting Polygons to Points

We analyzed election results at the level of voter tabulation districts, or VTDs, which are several blocks in size and typically cast no more than a few hundred votes in state legislative elections.  We mapped the VTDs onto the new lines proposed by LATFOR, then added up the votes of all VTDs that fell within a proposed district to determine its outcome.

In order to allocate the VTD-level vote counts to LATFOR’s proposed districts, CUR matched VTDs spatially with the current and proposed legislative district using ESRI’s ArcGIS Desktop software. The current and proposed Senate and Assembly districts are coterminous with Census blocks (in fact, the districts are “built” using Census blocks).  Unfortunately, neither LATFOR nor the state’s Board of Elections provides election results at the block level.  The Board of Elections records data by election district, which sometimes are smaller than VTDs, but for this project we did not have access to the election district data.

The challenge was that where the VTDs were larger than Census blocks in some places, the VTD boundaries crisscrossed the district lines (see example at right from Queens; click to enlarge).  In order to assign Senate and Assembly district IDs to each VTD, CUR converted the VTD boundaries to centroids (the geographic center-point of each VTD).  We used the lat/lon centroid values provided by the Census Bureau’s TIGER data.  Then we used a spatial join using ArcGIS to add legislative district identifiers to each VTD based on the legislative district its centroid was inside.  See the image below for the locations of the VTD centroids in this area of Queens.

In the instances where VTDs crisscross legislative districts, this technique will allocate all of a VTD’s votes to a single legislative district rather than splitting them across multiple districts.  This will over- and underestimate vote totals in some districts. But the process avoids the cumbersome effort involved in the alternative: splitting VTD vote counts.  The splitting process uses one of two methods:

  • using block-level population to “spread” the VTD votes across the VTD (multiplying the VTD vote count by the percentage of the VTD population occupied by each block and assigning the result to each block), or
  • weighting the VTD vote count based on the area of the portion of the VTD in each district.

Either of these approaches will result in fractions of people being assigned to one legislative district or another.  In fact, LATFOR appears to have used some sort of weighting method to assign election district vote counts to VTDs, since some of LATFOR’s VTD vote totals included fractions.

The centroid-approach and the weighted population / area approach both make assumptions about how to allocate vote counts.  But we tested the centroid process with current legislative districts and found that our VTD-allocated vote totals either exactly matched the results from the Board of Elections or were within a few hundred votes (which did not change the 2010 outcome).

Whether we used the centroid-approach and the weighted population / area technique, it otherwise would’ve been difficult if not impossible to determine how to allocate the VTD-level vote counts to legislative districts without GIS.  There are almost 15,000 VTDs across New York State, and there are (currently) 62 Senate districts and 150 Assembly districts.  With GIS, the process was relatively straightforward and efficient.

Aggregating by District

At the VTD-level, LATFOR provides the total number of votes cast by party in each election, not by candidate.  One challenge that we confronted was assigning the votes cast for fusion candidates who were backed by a major party but also received support on smaller parties’ ballot lines.  For example, many Democratic candidates received significant numbers of votes on the Working Families Party ballot line, and many Republicans got substantial support on the Conservative Party line.  Cross-party endorsements were even more variable for the Independence Party: in some districts, the Democrat received support on the Independence Party line; in others, its endorsement went to the Republican.

We decided that the most accurate way to re-map the election results was to assign the votes for each VTD based on the actual vote patterns for the Senate or Assembly district that contained that VTD in 2010.  In other words, if the Democratic candidate in an Assembly district ran on the Democratic, Conservative, and Independence lines, we assigned the Democratic, Conservative, and Independence votes in all the VTDs in that district to the Democratic candidate.  When we allocated the VTDs to the proposed Senate and Assembly districts, we added up the votes based on these patterns.  This ensured that the local voting patterns from 2010 were allocated accurately to the proposed districts.

The Results: Maps vs. Plain Old Numbers

The result is that we were able to calculate the number of proposed districts that, all other things being equal, would have had a Democratic winner in the Assembly and a Republican winner in the State Senate.  The important finding is that both parties would have increased their majority — which is especially interesting in the Senate, where the Republicans currently only have a 1-seat majority.  In Albany, the majority in each house is extremely powerful, so holding on to (or improving) those margins is all-important.

Of course, as the New York World/Times Union article points out,

To be sure, no district votes the exactly the same way in consecutive elections: the quality of candidates, changes in the population and the national political climate (which in 2010 favored Republicans) all play important roles. But voting behavior in previous elections offers the best available indication as to how a district is likely to perform.

The map that the New York World published along with the article uses red/blue color-shading to visualize the impact of the voting patterns on the proposed districts.  In the state Senate, the analysis shows the majority party increasing the number of seats by two.  On the map, that result is almost lost in the sea of red districts (most of the Republican seats are in upstate New York and Long Island, where the districts cover much larger areas than the more densely populated and largely Democratic districts in New York City).  The real power of our finding is the change in number: from 32 to 34.  In some ways, that says it all.

Nonetheless, the map (along with CUR’s interactive map comparing current and proposed district boundaries) provides a strong graphic and interactive element to the story, and provides context as you move your mouse over the districts to see the vote totals change from one to the next.

Watch for more analysis as LATFOR publishes its proposed Congressional district lines, and when the final Senate and Assembly districts are drawn.

Interactive NY redistricting map with cartoDB and more

UPDATE Nov. 5, 2012

In preparation for the Nov. 2012 election, many news organizations and others are linking to our interactive State Legislature and Congressional redistricting maps. We’ve posted examples at the Center for Urban Research website.


UPDATE Sept. 7, 2012

We’ve updated our map of redistricted State Senate and Assembly districts, highlighting the differences in race/ethnicity characteristics between total population and voter-eligible population – in other words, comparing the characteristics of all those who live in the new districts versus the smaller group who will be eligible to vote for each district’s representatives.  In some cases the differences are striking.

Our examination of the district-by-district data is available here.


New York State, like all other states, is in the midst of redrawing its legislative district lines. To help you follow along, our team at the Center for Urban Research has launched an interactive redistricting map for New York.  We collaborated with The New York World to develop the maps (though we encourage anyone and everyone to use them!).

The World’s reporters and editors are using our maps to go between the lines and explain how redistricting really works in the Empire State. (Here’s their first piece: The art of redistricting war.)  And we hope you’ll be able to use the maps too, to help answer questions such as:

  • Will you still be represented by the same State Senate or Assembly district you live in now?
  • Will you live in the newly proposed (and controversial) 63rd Senate district?
  • Is your neighborhood, town, or county going to be “carved up” by a new legislative seat?
  • Will your community’s historical voting power be diluted by the new districts?

We have some examples of gerrymandering at our Center’s website. In the meantime, here’s how you can use the maps.

Map features

The maps compare the current and proposed district lines (which our team mapped based on Census block lists published by the state’s redistricting task force, known as LATFOR). Here’s how they work:

  • Enter your address to find out what district currently represents you, and which proposed district you’d live in.
  • The current districts are on the left, and the proposed districts on the right.
  • You can also click on either map to highlight the current and proposed districts. As you move one map, the other moves in sync.
  • When you enter an address or click on the map, an info window pops up listing the current and proposed districts. You can click the link for the current district to go to that Senator or Assemblymember’s website.
  • Switch between State Senate and Assembly districts. Congressional districts will be posted once the data is available from LATFOR.
  • You can zoom in to street level, or zoom out to a statewide view. Switch between a street basemap or an aerial view from Microsoft’s Bing maps to see geographic details.

If you’re using the “Overlay” view, you can move the transparency slider to the right to display proposed districts, and to the left to fade back to current districts. The video below shows how:

If you want to share the map you’ve made, click the “Link” in the upper right of the map page to get a direct link to the area of the map you’re viewing. It will look like this:

http://www.urbanresearchmaps.org/nyredistricting/map.html?
lat=40.72852&lon=-73.99655&zoom=13&maptype=SIDEBYSIDE
&districttype=SENATE
  • You can share this on Twitter, Facebook, etc and email it to friends and colleagues.
  • You can also embed the map at your site. Use this link …
http://www.urbanresearchmaps.org/nyredistricting/map.html?output=embed
  • … or add < &output=embed > to any of the direct links you create, like this:
http://www.urbanresearchmaps.org/nyredistricting/map.html?
lat=40.72852&lon=-73.99655&zoom=13&maptype=SIDEBYSIDE
&districttype=SENATE&output=embed
  • … or wrap the snippet below in an iframe tag (I’d wrap it myself, but wordpress.com strips out iframe tags):
src="http://www.urbanresearchmaps.org/nyredistricting/map.html?output=embed" 
frameborder="0" marginwidth="0" marginheight="0" 
scrolling="no" width="600" height="700"

Side-by-side maps with OpenLayers

We borrowed from our “Census Comparinator” mapping site that Dave Burgoon artfully developed, in order to provide three ways to compare the current and proposed legislative districts:

  • a side-by-side view — two maps that are synced and move as one;
  • an overlay — a single map where you can fade between current and proposed districts; and
  • the vertical “before-and-after” slider approach.

I blogged about the Comparinator approach here and here. John Reiser also gave the technique a shoutout at his “Learning Web Mapping” blog for Rowan University.

With Census data, our “Comparinator” approach helped visualize changing spatial patterns of race/ethnicity trends – in cartographic terms, between two choropleth maps. With legislative districts, the comparison is between two sets of boundary files with no inner fill. So here we’ve set the side-by-side view as the default — we think the side by side maps give the easiest way of visualizing how the districts may change. But we also give you the option of viewing the districts with our vertical slider bar if you’d like, or the overlay.

Behind the scenes

For the proposed districts, we used ArcGIS to create the legislative district shapefiles based on LATFOR’s Census block assignment lists.  The current district boundaries are from the Census Bureau’s TIGER files (here’s the FTP page if you’d like to download the “lower” house districts — in New York, that’s the Assembly — or the “upper” house shapefiles — the State Senate).

We use OpenLayers for the map display and navigation with this application, as we’ve done with most of our other interactive maps. OpenLayers is easy to use, enables us to access Bing map tiles directly (so the basemap performance is smooth), and provides a robust JavaScript library for online maps.

That said, newer approaches such as Leaflet.js enable more interaction such as mouseovers, so we’ve started experimenting with some impressive new tools. More to follow!

One of those new tools is the powerful backend geospatial database engine from the team at Vizzuality: cartoDB. Hosting the legislative district shapefiles on cartoDB provided lots of advantages over hosting the data ourselves or setting up an Amazon cloud instance on our own. cartoDB provides:

  • great performance — not only for the district boundaries, but soon we’ll be adding election district maps to show voting patterns within each Senate and Assembly district. We don’t want to bother with creating pre-rendered tiles for this data. cartoDB will render it speedily on the fly.
  • cartographic flexibility: cartoDB uses cartoCSS for map symbology and labeling. Though there are still some quirks with cartoCSS, it was easy to grasp and it’s basically just CSS, so it makes styling easy if you’re familiar with modern web design. And cartoCSS incorporates scale-dependent rendering as well as attribute-based symbology, which makes it powerful and flexible. CartoCSS can be implemented using the cartoDB management interface, or programmatically.
  • easy data management: if you know SQL — and even better, if you’re familiar with SQL commands with PostGIS — you can quickly and easily modify tables, filter data, and perform spatial operations. (The screenshots at the cartoDB github page offer some examples.) Very cool.
  • scaling: cartoDB uses PostGIS and makes use of Amazon’s platform. So if our maps go viral, we’re ready for the usage spike!
  • open source: if you want to manage your own instance of cartoDB, just download the code and go! Big props to Vizzuality for an amazing geospatial toolkit.

Other thanks go to:

  • LATFOR, the state’s redistricting task force. Whatever you think about their redistricting process, they’ve done a great job with open data. They’ve not only posted the list of Census blocks that make up each proposed legislative district. But they also posted a wealth of data at the Census block level and also at the election district level (with a crosswalk between EDs and Census “voter tabulation districts”). This data is indispensable for visualizing, analyzing, and (hopefully) making sense of the new districts.
  • Dave Burgoon and the CUR team. Dave put together the redistricting mapping site in record time. Although it’s based on work he had already done with the Census Comparinator maps, it still involved substantial modifications and enhancements. But he made it happen as professionally and elegantly as always.
  • The New York World. We had been planning to create an interactive mapping application to build on our Census Comparinator site and to help people visualize the impacts of the redistricting process and demographic changes more broadly.  But the World team – Alyssa Katz, Michael Keller, and Sasha Chavkin – met with us a few weeks ago to discuss how we could collaborate on analyzing and mapping the upcoming district proposals from LATFOR.  The discussion inspired us to roll out a mapping site specific to New York State and focused on comparing the current and proposed districts. We’re thrilled to be able to work closely with them on this project (watch for more maps and data in the near future!).
  • The Hagedorn Foundation. The Foundation has provided funding support for our efforts to map and analyze Census data for a variety of civic engagement purposes, especially for Hagedorn’s Long Island-based grantees but also nationwide. Their support has been essential for us to develop innovative mapping applications like the NYS redistricting maps – not to advocate specific district plans one way or another, but to give local residents and others the tools they need to understand the impact of redistricting and hopefully get involved in the process.

Proposed NYS Senate & Assembly districts available in GIS format

UPDATE Nov. 5, 2012

In preparation for the Nov. 2012 election, many news organizations and others are linking to our interactive State Legislature and Congressional redistricting maps. We’ve posted examples at the Center for Urban Research website.


UPDATE Sept. 7, 2012

We’ve updated our map of redistricted State Senate and Assembly districts, highlighting the differences in race/ethnicity characteristics between total population and voter-eligible population – in other words, comparing the characteristics of all those who live in the new districts versus the smaller group who will be eligible to vote for each district’s representatives.  In some cases the differences are striking.

Our examination of the district-by-district data is available here.  The New York Times gave our analysis a shout-out in their CityRoom primary election day column.

You can also visit our original NYS redistricting “comparinator” map described below, at www.urbanresearchmaps.org/nyredistricting/map.html


UPDATE February 5, 2012

You can visualize these proposed districts in relation to the current New York State Senate and Assembly districts with our new interactive redistricting map.  We developed the interactive map in collaboration with The New York World, and here’s an article using the maps to describe the redistricting process in the Empire State.  For more background on the interactive map, visit this blog post.


Original Post

If you’re hoping to use GIS or any of the online mapping tools to map the legislative district lines in New York State that were proposed today by the state’s redistricting task force, you’ll have some work to do.  The Task Force released PDF maps as well as “block assignment lists” for the proposed districts.

Unless you’d like to use the shapefiles and/or KML files that our team at the CUNY Graduate Center created!  Here’s our web page with the info: http://www.urbanresearch.org/news/proposed-nys-districts-in-gis-format

Happy redistricting mapping!

Access to local GIS data

Rob Goodspeed has an interesting post about his survey of the policies and practices of local governments in Massachusetts regarding GIS data. It looks like a good read. In my experience (in New York State), local governments can have more interesting GIS data (for example, tax parcels and real property records) than the state or Feds, but their data access policies and/or practices can be more limiting. There are major exceptions (NYC, for example), but even New York City requires a fee and restrictive license to access its property data.

I look forward to reading Rob’s paper. Among other things, Rob is a PhD student at MIT. Nick Grossman of Civic Commons first alerted me to the paper via Twitter.

Some NYC OpenData improvements – small but important victory!

I noticed today that NYC’s new OpenData site (on the Socrata platform) has made some modest improvements since I blogged about it earlier this month, and since several people have responded to comments from Socrata’s CEO.

In particular, many of the files listed in the Socrata/OpenData site as “GIS” files or “shapefiles” are now actually available for download as shapefiles.  You have to dig a bit to find the download option — it’s not available via the  button. You have to click the  button, and then scroll down to the “Attachments” section of the About page.  But in many cases, you’ll now find a zipped file containing a GIS shapefile.  Small — but important — victory!

The back story

When the OpenData site first launched, I was very concerned because there was no option to actually download most geospatial data sets — you could only access them as spreadsheets or web services via an API.  That’s not very helpful for people who want to work with the actual data using geographic information systems.  And it was a step backward, since many agencies already provide the GIS data for download, and earlier versions of the OpenData site had made the data available for direct download.

It also seemed like it was extra work for the agencies and for us — extra work to convert the data from GIS format into spreadsheets, for example, and then extra work for the public to try to convert the data back into GIS format once they had downloaded a spreadsheet from the OpenData site.  Seems pretty silly.

It also seemed like it was an example of DoITT not understanding the needs of the public — which includes Community Boards, urban planning students, journalists, and many others who routinely use GIS to analyze and visualize data.  Spreadsheets and APIs are nice for app developers — and the “tech community” broadly speaking — but what about the rest of us?

More public access to data, not less

If the city adds the shapefiles as a download option, that’s providing more open access to data, not less.  But by not offering GIS data along with the other formats, the Socrata system seems to be limiting access.  I’d hope that NYC would be as open and flexible and accommodating as possible when it comes to accessing public data.  Socrata’s CEO seems to argue that with the Socrata platform it’s too hard to do that.  If he’s right, maybe we should just stick with a tried and true approach — NYC agency websites already provide direct download of GIS data along with many other formats.

But I know that we can do better.  In fact, Chicago’s open data portal (also powered by Socrata) has offered many GIS datasets for direct download from Day 1.  Actually, Chicago has 159 datasets tagged as “GIS” files, while New York only has 69what’s up with that, NYC? I thought NYC was the best in everything when it comes to open data?

Still more to be done

Alas, even though we’re talking about a victory here, we can’t pop open the champagne quite yet.  Several of NYC’s data sets via the Socrata site aren’t as current as what you can already get from agency websites.  For example:

  • zoning is current as of August 2011, but you can download more current data (September 2011) from the ever-improving Planning Department’s Bytes of the Big Apple website;
  • building footprints are older (September 2010) than what you can download from DoITT’s GIS site itself (click through DoITT’s online agreement and you’ll get a buildings database from March 2011); and

Also, some data sets described on the Socrata/OpenData site as “shapefiles” are still not available in GIS format.  Some examples:

  • NYC’s landmarks data.  The OpenData site describes this data as a “point shapefile … for use in Geographic Information Systems (GIS).”  But it’s only available from the OpenData site as a spreadsheet (or similar format) or via an API.
  • Waterfront Access Plans.  The OpenData site describes this file as a “polygon shapefile of parklands on the water’s edge in New York City … for mapping all open spaces on the water’s edge in New York City.”  But like the landmarks data, it’s only available as a spreadsheet or via an API.  False advertising, if you ask me.  But if you go to the source (the City Planning Department), the shapefile is there for all to access.  So why is the Socrata/OpenData site any better? I’m still wondering that myself.

And the Socrata/OpenData site still doesn’t provide the kind of meaningful data descriptions (or metadata) that you’ll get from agency websites such as Bytes of the Big Apple or Dept of Finance — data descriptions that are absolutely essential for the public to understand whether the information from NYC OpenData is worth accessing.

But hope springs eternal — someone listened to our concerns about lack of actual geospatial data downloads, maybe they’ll also listen when it comes to everything else. Fingers crossed!

Pretty NYC WiFi map, but not useful beyond that

@nycgov posted a tweet on Friday touting the map of WiFi hotspots on the new NYC OpenData site.  I was impressed the city was trying to get the word out about some of the interesting data sets they’ve made public. It was retweeted, blogged about, etc many many times over during the day.

The map is nice (with little wifi symbols  marking the location of each hotspot).  And it certainly seems to show that there are lots of hotspots throughout the city, especially in Manhattan.

But when I took a close look, I was less than impressed.  Here’s why:

  • No metadata.  The NYC Socrata site has zero information on who created the data, why it was created, when it was created, source(s) for the wifi hotspots, etc.  So if I wanted to use this data in an app, or for analysis, or just to repost on my own website, I’d have no way of confirming the validity of the data or whether it met my needs.  Not very good for a site that’s supposed to be promoting transparency in government.
  • No contact info.  The wifi data profile says that “Cam Caldwell” created the data on Oct. 7, 2011 and uploaded it Oct 10.  But who is Cam?  Does this person work for a city agency?  It says the data was provided by DoITT, but does Cam work at DoITT?
    • If I click the “Contact Data Owner” link I just get a generic message form.  I used the “Contact Data Owner” link for a different data set last week, and still haven’t heard back.  Not even confirmation that my message was received, let alone who received it.  Doesn’t really inspire confidence that I can reach out to someone who knows about the data in order to ask questions about the wifi locations.
  • No links for more information. The “About” page provides a couple of links that seem like they might describe the data, but they don’t.

If I were to use the wifi data for a media story, or to analyze whether my Community Board has more or less hotspots than other Boards, or if I wanted to know if the number of hotspots in my area has changed over time, the NYC Socrata site isn’t helpful.

Even looking at the map on its own, it’s not very helpful.  Without knowing if the list of hotspots is comprehensive (does it include the latest hotspots in NYC parks? does it include the new hotspots at MTA subway stations? etc) or up to date (the Socrata site says the list of wifi sites is “updated as needed” – what does that mean?), I have zero confidence in using the data beyond just a pretty picture.

I’m sure if I clicked the “Contact Data Owner” link, eventually I’d get answers to these questions. But that’s not the point.  The point is that the new NYC OpenData site bills itself as a platform to facilitate how “public information can be used in meaningful ways.”  But if the wifi data is any guide, the OpenData site makes it almost impossible to meaningfully do anything with the data.

The wifi data is another example of how I think NYC’s implementation of the new Socrata platform is a step backwards.  Other NYC websites that provide access to public data — the City Planning Department’s Bytes of the Big Apple site as well as agency-specific sites from Finance, Buildings, HPD, and others — all provide detailed metadata, data “dictionaries”, and other descriptive information about available data files.  This contextual and descriptive information actually makes these data sets useful and meaningful, inviting the public to become informed consumers and repurposers of the city’s data.

The Socrata platform, in and of itself, seems great.  But NYC hasn’t done a very good job at all of putting it to use.  #opendata #fail

NYC’s new OpenData website: soars and falters all at once

UPDATE (10/13/11)

This evening I received a call from NYC DoITT.   They were mainly calling to tell me that they changed the official rules for BigApps 3.0.  Yesterday the rules said that no new data would be added to the OpenData site until after the BigApps competition.  As I said in my blog, why wait?  But DoITT saw that and agreed.  So now that clause has been removed from the rules (see section D.1).  DoITT says that they agree they data should be accessible whether there’s a competition in effect or not.  That’s great news!  I’m looking forward to more dialogue on the other issues I’ve raised below.

___________________________________________

ORIGINAL POST

New York City yesterday announced its new version of what had been called its “Datamine” website, a single online point of entry to access the city’s digital data holdings.

I’ve critiqued the Datamine project before, but I was heartened by the city’s choice to use the Socrata platform to upgrade Datamine. As I wrote a couple of months ago:

NYC’s Datamine was an improvement in some ways over earlier opendata efforts in New York. Now that it’s been around for two years, I think it’s fair to say that Datamine is clunky at best. For me, I can’t wait for it to be replaced by something better. I’m looking forward to the NYC/Socrata roll out.

Yesterday’s announcement came with great fanfare: 230 new data sets! (so they say), BigApps 3.0!, cash prizes!, etc.

But is “NYC OpenData” any better than Datamine?

After digging into the site for several hours last night and today, I’d have to say yes and no. It has some great stuff with great promise, but it still falls flat in some key areas. I look forward to using it for the APIs, but for the raw data I’ll go back to the individual agencies that in many cases are doing a better job of providing access to the data.  Overall the city has come a long way with open data, but I still think the city’s concept of data-as-economic-engine is misguided.  More on that below.

The good

Socrata’s platform is impressive. I’ve blogged about it before, but it’s worth summarizing some of the high points:

  • You can immediately preview the data in your browser (no downloading needed just to see what it contains). And you can view more details about each row in the file — very helpful if you’re interested in one particular aspect of the data.
  • You can visualize  the data in multiple ways — using an interactive map option built into the platform or using one of 9 different chart options.
  • If you want to download/export  a data set, they give you at least 8 formats for extracting/exporting.
  • Short links and “perma” links are available to each data set.
  • There’s a “Discuss”  option where anyone can attach notes and commentary for each data set.  It’s user-generated metadata — you can immediately see, for example, if anyone else has commented about the data’s quality, or completeness, or how up-to-date it is.

The big news with this new approach is the availability of an API for programmatic access for each data set in the Socrata system.  On its face, the APIs look great, and the city deserves kudos for implementing them.  Socrata has developed a template for developers to hook into the data — either row by row, selected queries, or to view metadata — and the template also provides data publishers with guidance on how to structure their data for automated consumption.  And, it seems that DoITT has created web services for the mapped data sets, which is a big step forward.

There are other improvements with specific data sets, such as:

  • It looks like the map data for NYC park boundaries is fixed — I posted a detailed review last year about how the parks data via Datamine was basically impossible to use.  I had to scrape the NYC Parks website to convert it to a useful format. But now the park names are included with the park IDs in the same file. (However, this improvement is tempered by the fact that I can view the map of parks on NYC’s Socrata website, but I can’t download the data in a mapped format. I discuss that in more detail below.)

There are some interesting new data sets.  Two things that caught my eye are:

  • School zones are included in the data, which is something I had urged the city to include [PDF] when the BigApps competition was first announced in 2009.  (School zones are the key determinant as to where your child can attend public elementary school, rather than the administrative school districts.)  But the earlier version of Datamine included school zone boundaries, so this isn’t really new.
  • HPD Registrations.  Unfortunately the data dictionary accompanying this file can be cryptic, so I couldn’t easily decipher exactly what the file includes. But it seems to be a list of almost 140,000 buildings in the city registered as “multiple dwellings” along with each building’s landlord/owner, managing agent(s), and building details.  Should come in pretty handy for anyone interested in the landlord landscape in New York.

Here’s an example of why the data dictionary is not very helpful – the excerpt below is trying to tell us what the “REG-INDV-HM-UNIT-NO” field means:

Um, what?

I thought it was also intriguing (in an insider baseball kind of way) that the interactive maps used at the NYC Socrata site to show mapped views of the data are from ESRI.  And the API/web services provided for the mapped data files are ESRI-based.  DoITT’s GIS unit has made a point of using non-ESRI technology for its interactive maps (Citymap, Scout, ZoLA, etc). But the GIS web services for Socrata all come from DoITT.  Wonder what’s happening there.

The not-so-good

The Mayor’s news release about the new Socrata site proclaims that more than 230 new data sets are included. We don’t get any details about which ones; the release simply says that:

Examples of this new data include a directory of HHC Facilities; electricity, gas and steam consumption available by zip code; and school attendance and report statistics.

But I looked pretty closely at what new data sets I could find, and I was hard pressed to identify more than a few dozen.

Examples of old data masquerading as new simply because it’s available through the new Socrata site include many of the files from NYC’s Dept of Finance, such as:

  • Condominium comparable rental income listings (38 individual datasets);
  • Cooperative comparable rental income listings  (40 datasets); and
  • Summary of Neighborhood (Property) Sales (21 datasets).

That’s almost 100 data sets right there, close to half the number the city says are newly available.  But each of these have been online, for free download, at Finance’s website for several years.  This page notes that coop sales information has been available since 2006, and Finance started making the data available for batch download a couple of years after that.  The Neighborhood Sales data was put online a couple of years ago.  And Finance’s website has more thorough information about the data sets and how to use them than the Socrata site.

Other not-so-new examples include:

  • Street centerlines.  These are from DoITT circa 2009. In contrast, the City Planning Department “LION” file at DCP’s website is from September 2010, and is updated regularly.
  • Building perimeters. From DoITT circa 2010.  But DoITT has a more recent file at their website for direct download (click through the online agreement and you’ll find building footprints from March 2011).
  • Coastal boundaries. From City Planning, but this was posted on the Bytes of the Big Apple site last month.  Great data set, but not new.
  • Campaign contributions. From the NYC Campaign Finance Board.  The data is current (covering the 2013 election cycle), but the files are already available in batch format and via a searchable website from CFB.
  • Landmarks data. There are multiple, conflicting data sets at the NYC Socrata site regarding landmarks.  For example, one data set of “NYC Landmarks” is from 2009, another (called “LPC Landmark Points”) is from 2010.  Either way, there have been several new landmarks and historic districts designated since then by the Landmarks Commission.

Even if there was only one new data set in the new Socrata site, that’s better than nothing. But there’s so much data maintained by city agencies that is still not easily, publicly accessible.  My blog post when BigApps was first announced in 2009 has a listing of some key data files that still haven’t seen the light of day.

The city should be doing a better job — especially since there’s been so much pressure on them to improve their open data policies, they have an avowed policy of doing so, and they’re also under a state law (FOIL) to require them to do so. Frustrating.

One of my biggest and longest standing gripes is about property data.  There are a number of property-related files the NYC Socrata website.  But nothing that allows us to come close to the City Planning Department’s “MapPLUTO” dataset.  The city still charges a fee (up to $3,000 per year) with a restrictive license agreement in order to access the PLUTO data — a mapped file of all properties in NYC with a wealth of information about each one (zoning, ownership, building heighs, land use categories, assessed value, etc).  It’s an essential data set for anyone trying to understand real estate, urban planning, neighborhood change, and more in the city.

When will City Planning get it? They’ve done such a great job of making other data sets available — files they used to charge for but now provide for free, and in better formats, with great metadata, and updated frequently.  The agency obviously spends a lot of time preparing these other data sets that are freely available, so I don’t buy the argument that the PLUTO fee covers their “costs” of doing extra work to put PLUTO together.  I just don’t understand.  And property data is so incredibly useful in NYC — certainly to the big real estate players, but I’m not concerned about them.  If it were free for everyone, at least we’d have a chance at a level playing field — helping “the little guy” do property analysis and mapping so he/she can analyze land use, understand policy implications, etc.

Data for people, not just machines

Data access — at least in this first iteration of the new Socrata site — seems to be weighted toward APIs, and therefore app developers. I understand the value of the API approach — I’ve developed apps myself, and at CUNY we have online sites that can definitely make use of the APIs. And I was kind of amazed that DoITT opened these up.  So the APIs are good, and perhaps they’re worth the effort to create and maintain a one-stop-shop like NYC Socrata.

But for the average user — someone at a Community Board, or a local media outlet, or a City Councilmember’s office — the city’s implementation of the Socrata system seems against them.

For example, with one or two exceptions I wasn’t able to download any mapped data sets from NYC Socrata.  Many files (45 by my count) are described as “GIS datasets”, and they’re obviously in ESRI’s “shapefile” format to begin with, but the “Export” option only provides flat files (CSV, JSON, XLS, XML for example), and not even the now-ubiquitous KML format (used by Google and many others).

If I click the API link for these data sets, this enables me to view the data as map layers in my desktop GIS application.  But I can’t extract any actual data from these links in order to work with it on my own.  The screenshot below (from ESRI’s ArcCatalog application) seems promising, but the inability to download the mapped data itself is very limiting.

It’d be easy enough (I’m assuming) to just add shapefiles to the list of Socrata’s data export formats. The shapefile format (.SHP) is already basically an open one (all the major open source GIS packages read it), so why force GIS users to do extra work to access GIS data?  And why have DoITT go through extra work converting from SHP to something else, just to have the user convert it back again. For “point” locations this isn’t a big deal — it’s easy enough to convert latitude/longitude coordinates into a mapped data set.  But this isn’t straightforward at all for polygons (district boundaries, for example) or lines (streets, transit routes, etc). I’m not saying don’t provide the data in the other formats, just add SHP to the list where appropriate.  (Some GIS datasets are available as GIS downloads: school zones, for example. But this is an exception, as far as I can tell.)

Indeed, not having GIS-ready formats is a step backward. If I visit the City Planning Department’s “Bytes of the Big Apple” website, I can download a wealth of files in GIS format, and several of them are updated regularly. It’s great. Hopefully the NYC OpenData site doesn’t supplant the individual agency sites. For now, they’re better for me, and I’d imagine they’re better for many other users.

And having the raw data, rather than just API access, gives users more flexibility.  For example, during the preparation for Hurricane Irene, several organizations downloaded NYC Datamine files in GIS format to create interactive maps of evacuation zones and evacuation sites.  (And these groups helped the city in a big way because the city’s own maps and website were down, making it difficult if not impossible to get essential information from NYC.gov.)  But the city changed several of the evacuation sites just a day or two before the storm was going to hit.  If the outside organizations didn’t have the raw data that we could update ourselves, our presentation of the evacuation sites would’ve been incorrect and misleading.  I wouldn’t want to rely on the city updating its API in a crisis situation like that, given how rocky the city’s digital response was to the storm itself.

Tying open data to app competitions & economic growth is the wrong approach

(Note: my concern here still stands, but the city has modified its position a bit, which is great.  See the 10/13 Update above.)

I think the real issue here is that the city’s open data efforts are being driven more by the desire to use data access as a way to leverage economic development, and less about true government transparency.

For example, as with the first two BigApps competitions, no new data files will be added to the Socrata site until the latest BigApps competition is over (see section D.1 at the official rules).  Why wait?  Why should app developers get preference?  What about the rest of us? Is NYC providing data just so app developers can do free work for the city, and so the city can make a news splash about open data? Open data should be open 24/7 — and should be updated on a regular basis — not just when it’s convenient for the city and for developers.

Next steps

I understand that the new NYC Socrata site is a work in progress, and will almost certainly be improved going forward.  But for now, although it includes lots of data, much of this has already been available elsewhere.  The APIs are intriguing, but I hope they don’t preclude other ways for people rather than machines or apps to access the data.

At this point, with few exceptions I still would prefer to go to the individual agency websites (or even talk to agency staff and request the files via email, or even via disks & snail mail!) to get the data — from what I’ve seen so far, chances are it’ll be more timely, in better quality, and I’ll have better access to metadata/explanations of the files.

I’m even wondering if instead of a Socrata-like site, it might not be better to encourage the agencies directly responsible for creating the data to continue efforts to provide public access, and having them engage with people using the data so they’d see the benefits of open data (and/or realize that it’s not so bad to provide access to their files to the broad public in easily accessible ways).  At the least, the new NYC Socrata site shouldn’t preclude this agency-specific work to be done.

I’ve already had a good, late-night exchange on Twitter with DoITT on some of these issues. I’ll be submitting feedback directly at the Socrata website.  And hopefully the dialogue will continue.

NYC bikeshare maps & spatial analysis: an exploration of techniques

UPDATE (Feb. 2012)

  1. Reader Steve Vance suggests in the comments below that I could use Google Refine to parse the JSON file and convert it to Excel without relying on the tedious Microsoft Word editing process I summarize below.  He’s right.  Google Refine is amazing. It converted the JSON file to rows/columns in about a second.  And it has powerful editing/cleaning capabilities built-in.  Thanks Google!
  2. Alas, I had hoped to test Google Refine on the latest list of user-suggested bikeshare stations.  But when I checked in mid-February, the link at http://a841-tfpweb.nyc.gov/bikeshare/get_bikeshare_points no longer returns all the detailed info about each suggested site.  It only returns an ID and lat/lon for each site.  There’s another link I found that returns the details (http://a841-tfpweb.nyc.gov/bikeshare/get_point_info?point=1), but it seems to be just one at a time (change the “point=1” value).  Sigh.  If someone wanted to replicate what I’ve done with the latest data, perhaps either NYC DOT or OpenPlans could provide the file directly.

Original Post (Sept. 2011)

Two weeks ago New York City announced an ambitious bikeshare program, designed to provide 10,000 bikes at 600 bike-sharing stations in Manhattan and parts of Brooklyn by next summer.  I had two immediate thoughts:

  1. I wondered if all 10,000 new bikers will ride like delivery staff and further terrorize me and my pedestrian 5-year olds; and
  2. safe or not, the bike stations would be put somewhere, and maps can likely help figure out where.

I’m a cartographer, so I’ll focus on the second issue for the purpose of this blog post. My maps and analysis below don’t provide any definitive answers — they’re more of an exploration of spatial analysis techniques using the bikeshare data as an example.  I don’t know if this will be helpful to DOT, but if it is, then that’s great.  If not, hopefully at least they’ll be of interest to GIS and biking geeks alike.

NYC’s bikeshare stations: crowdsourcing suggestions

To help figure out where the bikeshare stations might be located, the city’s Dept of Transportation partnered with OpenPlans to provide an interactive map where anyone could suggest a location and provide a reason why they thought it was a good spot. If someone has already picked your favorite spot on the map, you can select that marker and click a “♥ Support Station!” button to register your approval.  Added up, these supporting clicks can provide a “rating” of how many people like each location.

It’s a great, easy to use app. Within just a few days several thousand people had posted their suggestions.  According to DOT,

As of September 20 at 3:30pm [just 6 days after the suggest-a-site went live], we have received 5,566 individual station nominations and 32,887 support clicks.

(via OpenPlans)

But the map looked overwhelmed! Manhattan was covered, as was most of downtown Brooklyn.  It seemed like almost everyone wanted a bikeshare station on their block.  New York Magazine put it this way:

As you can see in the map above, New Yorkers have spoken: The best spots for bike stations are … everywhere/wherever is right next to them.

I wondered how useful this crowdsourced data actually would be for identifying the best sites for bikesharing stations.  NYC DOT says it will be conducting “an intensive community process” to involve multiple stakeholders in helping decide where the 600 stations will go.  Presumably several factors will determine station locations, but it seemed like the crowdsourced data could play a key role — hopefully the website was more than just a PR ploy.

Given all those “dots on a map,” it seemed like a good opportunity to examine how spatial analysis tools could be used — first to see if the crowdsourced location patterns meant anything, but then to see if there’s any value to using them in siting analysis.  Had “the crowd” told us something new and useful, or was it something we already knew and would be better determined through DOT’s public process?

Spatial patterns

Luckily OpenPlans (and DOT) designed the suggest-a-station website so all those dots on the map could be scooped up via a simple HTTP request and converted to GIS format.  At the end of this post I describe how we got the data and put it into a mappable format.  Once we did, we were able to analyze it spatially.  I’ll post the shapefile, as well as a version at Google’s Fusion Tables, shortly.

A few days after the program was announced, DOT produced a “heat map” that “illustrated the number of suggestions and supports per square mile as of September 19” (map at right).

Our version of a “heat map” using the September 19 data (based on the results as of 9am that day) is shown below.  (Our map uses the same rating scale as the DOT map, but its slightly different patterns could be due to different model specifications to create the map.  We used ArcGIS’s “Kernel Density” function to develop our map — DOT may have used a different method. Even if we both used kernel estimation, this technique can result in different surface patterns based on different inputs such as cell size and search radius.)

But do these maps really tell us anything useful? Some people tweeted that the concentration of suggested bikeshare sites matched New York’s “hipster” population.  Others said that the patterns were “almost perfectly congruent with race/class/culture divides” in the city.

I disagree — I don’t think the suggested bikeshare patterns match any obvious demographic characteristics, whether it’s race/ethnicity or “hipsterism”.  (This may be worth pursuing further, but for now I leave that to others.)

I think a more likely relationship is based on where people work.  The orange-to-red areas on both maps — indicating a high concentration of suggested bikeshare sites with high ratings from website visitors — match the locations of the city’s commercial areas: Manhattan below 59th Street and downtown Brooklyn.

Another possibility, though, is that people who suggested bikeshare locations were just following DOT’s preferences – a spatial version of survey response bias.  In its bikeshare FAQ, DOT says that phase 1 of the program will focus on the following areas:

Manhattan’s Central Business District and nearby residential areas, including Brooklyn neighborhoods of DUMBO, Downtown, Fort Greene, Bedford-Stuyvesant, Williamsburg, Greenpoint and Park Slope

The NYC Planning Department produced a map of these areas in a Spring 2009 report [PDF] as follows:

Superimposing the Phase 1 area on the rating density map above shows that there’s almost an exact match between Phase 1 (outlined in dark pink) and the highest concentrations of suggested/supported sites (the dark orange and red areas on the map):

So based on these density maps (“heat maps”), it’s not clear if the overall patterns from these maps tell us anything interesting about the wisdom of the crowd, or useful about where to put bikesharing stations.

Digging deeper

But whether the overall patterns mean anything or not, maybe the suggested locations could be analyzed to see if they have value as criteria for local siting decisions. In other words, within the patterns, maybe we can use the crowd’s suggestions as a key piece of analytic information, providing quantifiable indicators about where the stations should go.

More than 2,700 bikeshare locations (as of Sept. 25) were suggested within the Phase 1 area — four and a half times the 600 sites that will eventually be sited.  Perhaps they covered every possible bikeshare site. But perhaps there’s also a pattern (or patterns) to the suggestions that will help with the decision to whittle 2.700 down to 600.

For simplicity’s sake I evaluated the suggested station locations against one criteria — proximity to subway station entrances.  Obviously there are other factors to examine (threshold bikeshare station density, proximity to specific residential or employment centers, terrain, etc).  But several people have noted that a bikeshare program can extend the reach of subways — transit riders could ride to a distant subway more easily, cheaply, and quickly than a bus or a cab, or when they reach their subway stop they could pick up a bike and ride to their final destination without the hassles of a cab, etc.  So my assumption is that proximity to subway stations will be a key factor in determining bikeshare station locations.

But how do the suggested locations from the DOT/OpenPlans map compare with that hypothesis?  Are the highly rated bikeshare sites near subway stops?  About 10% of the suggested sites included reasons that mentioned subways.  Did website visitors suggest enough bikeshare sites near subways to make it easier for DOT to pick and choose which ones are best?

(Btw, this same type of analysis can be applied to bike routes, for example.  I just wanted to focus on one component for now.)

Spatial analytics

I used several spatial analysis techniques available through ArcGIS’s toolbox to shed some light on these questions.  The tools are powerful, and ESRI has made them easy to use and interpret.  The tools also underscore the power of GIS beyond making maps — extracting information based on the spatial relationships of multiple geo-referenced data sets.

In order to compare suggested bikeshare sites with subway stations, I used the file of subway entrances/exits available from MTA (current as of July 19, 2011).  The file provides the latitude/longitude of 1,866 entrances and exits, identifies the station name for each one, and lists the routes that serve these stations.  It provides a more precise spatial measure of access to the subways than a single point representing the center of each station (which is how stations are shown on most interactive and print subway maps).

With this file, we can determine how close each suggested bikeshare site is to the actual spots where people exit and enter the subway system.

To calculate proximity, I used the “Near” feature in the ArcGIS Toolbox, which “[d]etermines the distance from each feature in the input features to the nearest feature in the near features.”  I analyzed 5,587 suggested bikeshare sites based on the DOT/OpenPlans map as of Sept. 25 (see data discussion at the end of this post). Here are some statistics:

  • 92 sites were within 25 feet of a subway entrance;
  • fully one-third (1,954 suggested sites) were between 25 and 500 feet of a subway entrance (the length from one Manhattan avenue to the next is usually about 600 feet);
  • another quarter of the sites (1,677) were within 500 and 1,250 feet (1250 ft being roughly a quarter mile, the rule-of-thumb distance that people will walk for public transportation); and
  • the remaining 2,134 were more than a quarter mile from a subway entrance.

Seems like lots of bikeshare stations were suggested in close proximity to subway entrances. If the actual bikeshare sites will be near subways entrances, which entrances should we pick?

(An aside: since just over a third of suggested bikeshare sites were located relatively far away from subway entrances, we can also evaluate these patterns.  The hypothesis would be that if people are picking up bikes at subway stations, they’re using them to travel to destinations further away from subway stops.  Therefore some of the bikeshare sites will need to be located in these “destination” areas, and DOT will need some spatial criteria for locating them.  I’ll save this for a follow up blog post.  Thanks to Kristen Grady for suggesting it.)

One way to visualize the bikeshare/subway entrance relationships is with the following map, showing the subway stations in blue (just a center-point representing the middle of the station) and the bikeshare sites color-coded by proximity (I’ve limited the display of bikeshare sites to only those within 500 feet of a subway entrance so the map wasn’t too cluttered):

This map might be helpful, but you have to visually decide which clusters of close-by bikeshare sites are the most concentrated in order to prioritize which subway stations to focus on.  The map also omits the rating values.

If we incorporate ratings, the map below is an example of the result.  It only shows bikeshare sites very close to subway entrances — within 50 feet — and ranks the symbol size based on rating.  (We could just as easily pick another distance threshold, or display several maps each using a different distance threshold.)

This helps us focus on which subway stations might be best for a nearby bikeshare station, based on suggested bikeshare sites nearby that are ranked highest.

But we can use GIS to be more precise.  Another approach would be to visualize the pattern of the subway entrances themselves, based on average rating of each entrance’s closest bikeshare sites.  In other words, I’d like to use the ratings given to each suggested bikeshare site and assign those ratings to their closest subway entrances.  This will have the effect of combining subway proximity with bikeshare rating, and the resulting map will integrate these patterns.

Here’s an example of the result, with the rated subway entrances juxtaposed with the density map of rated bikeshare sites from earlier in this post:

This map says, “If you want to put bikeshare stations near subway entrances, these are the entrances you’d pick based on the average rating of the closest stations suggested by ‘the crowd’.”  It’s a way of prioritizing the bikeshare station siting process.  These subway entrances are the ones you’d likely start with, based on the preferences of the (bike)riding public who contributed to the DOT/OpenPlans map.

It looks like many subway entrances follow the overall pattern of bikeshare sites with the highest ratings. But there are some interesting differences in the above map. A couple of sites are completely outside the Phase 1 area (an outlier each in the Bronx and Queens), and only two subway entrances with average high ratings are in Brooklyn. The rest are in lower Manhattan. But only one of the Manhattan sites is near the highest rated area centered around NYU:

Here’s another view of this area, with the rated subway entrances overlain on a Bing street map:

In order to create the rated subway entrance map, I used the Voronoi polygon technique, also know as Thiessen polygons (Voronoi was a Russian mathemetician, Thiessen was an American meterologist.)  Voronoi polygons are enclosed areas surrounding each point (subway entrance) so all the other locations (in this case, bikeshare sites) within the polygon are closest to the enclosed subway entrance than any other entrance.  The subway entrance Voronoi polygons look like this:

Here’s a close up, with the subway entrances displayed as pink stars, and the suggested bikeshare stations as blue dots:

The blue dots (bikeshare sites) within a polygon are closer to that particular polygon’s subway entrance than any other entrance in the city. Other GIS techniques, such as creating a buffer around each subway entrance, or even using the “Near” calculations I described earlier in this post, wouldn’t precisely determine the closest criteria for all the points automatically and at once.

The other nice thing about creating Voronoi polygons is that the attributes of the reference points are transferred to the polygons (the polygons end up with more than just a random ID number; in this case, they include all the corresponding subway entrance attributes).  From there I did a spatial join in ArcGIS, joining the bikeshare sites to the polygons.  This automatically calculates the count of all points in each polygon, as well as statistics such as average and sum for any numeric attributes in the point file.  In this case, each subway entrance Voronoi polygon gets a count of the bikeshare sites within it (i.e., the ones that are closest to that entrance) as well as the summed rating and average rating.

From there we could create a choropleth map of the Voronoi polygons. But since we’re interested in the entrance locations rather than an aggregated area around them, I chose to create a graduated symbol map of the actual subway entrances. So I did an attribute join between the Voronoi polygons and the entrances using the shapefile ID field.  That enabled me to make the “Average rating by subway entrance” map above.

Limitations

One limitation to the Voronoi approach is that closeness is measured “as the crow flies.” There are other techniques that measure proximity using “Manhattan distance” (i.e., distance along streets rather than a straight line), such as ESRI’s Network Analyst extension for ArcGIS, but I’ll leave that to the DOT analysts who are going to decide on the actual bike share sites.

Other limitations of this approach have to do with the data themselves.  The bikeshare data from the DOT/OpenPlans website has issues such as:

  • entries accompanied by fictitious names (some examples from the Sept. 25 data include “Andy Warhol”, “George Costanza”, “Holden Caulfield”, “Lady Liberty”, and “United States”.  One or more people using the “United States” pseudonym submitted 51 entries throughout Manhattan, Brooklyn, and Queens, plus a single entry in the Bronx); and
  • multiple entries submitted by a single person. Someone – or some people – named Ryan submitted 143 entries.  Someone named Andrew Watanabe submitted 85 entries. Ryan and Andrew were the top two submitters.  After them and “United States”, there were 4 others who submitted 40 or more entries. It’s possible that these were all sincere. But some seem to be pretty goofy. Of Watanabe’s 85 suggested sites, for example, several included the following reasons:
    • “When whales accidentally swim into the Gowanus, they will be able to ride bike share bikes back out to sea.” (site on the Gowanus Canal)
    • “This will keep drunk booksellers from passing out on the sidewalk.” (site near the Bedford Ave L train stop in Williamsburg)
    • “When the zombie apocalypse comes, they will be riding bicycles. BRAAAAAINS!” (site in the middle of Mt. Laurel Cemetery in Queens)

Multiple entries might be fine, but if someone started plunking down markers on the map just for fun, this doesn’t really help us with meaningful location criteria.

There’s another concern about the crowdsourced data – the squeaky wheel problem.  The first map below shows the bikeshare suggestion pattern as of September 19; the second map below shows the patterns as of September 25.  The more recent map shows a new concentration of sites at the northern tip of Roosevelt Island (as well as a greater concentration in lower Manhattan and downtown Brooklyn, areas that already were very dense):

 

Sept. 19 patterns

 

Sept. 25 patterns

Why did northern Roosevelt Island all of a sudden become such a bikeshare hotspot?  I can’t say for certain.  But in a blog post on September 14 at the Roosevelt Islander, residents were urged to add sites to the DOT/OpenPlans map.  The post ended with the pitch:

So here’s what you can do to bring bike sharing to Roosevelt Island. Click on this link and say you want a bike sharing station on Roosevelt Island – do it now – please [emphasis added]

I don’t think making a pitch like this is a bad thing. (Far from it! It seems to have succeeded in getting attention on bikeshare sites on the island).  But whoever will be analyzing the sites from the DOT/OpenPlans map will need to decide if (and how) they should discount these crowdsourced lobbying efforts so the squeaky wheels don’t skew the map.

Making sense of it all

My analysis in this post is more for illustration than for actually determining best locations for bikeshare stations. A more rigorous analysis would need to deal with the data limitations I mentioned above, and also factor in other criteria.

But it was a fun exploration of the data and the techniques, and hopefully provides some useful ideas if readers are thinking of other spatial analysis projects involving proximity (especially the “closest” criteria).  I’m indebted to DOT & OpenPlans for enabling the creation of an interesting data set — the suggested bikeshare sites — for me to brush up on my spatial analysis skills.

Does my initial exploration shed any light on the wisdom of the crowd? It’s probably too early to tell (or my analysis was too limited to meaningfully evaluate the suggested sites).  But even so, I think the techniques I’ve described are helpful for prioritizing sites and for quantifying the results.  In that respect, the crowd’s input is a good thing.

Data issues, as always

Here are the steps we used to download the suggested bikeshare sites from the DOT/OpenPlans website in order to map and analyze the data:

  1. We used Fiddler to figure out that the suggested station locations were being maintained in a text file (in JSON format) available via http://a841-tfpweb.nyc.gov/bikeshare/get_bikeshare_points (Dave Burgoon ferreted this out).
  2. The JSON data looks like this:
{
"id":"4830",
"lat":"40.742031",
"lon":"-73.777397",
"neighborhood":"Fresh Meadows",
"user_name":"David",
"user_avatar_url":"",
"user_zip":"11355",
"reason":"There is no public transportation from Brooklyn-Queens greenway (Underhill Ave) to Flushing Meadows. By placing bike stations from Cunningham Park thru Kissena Park to Flushing Meadows will allow residents enjoy the parks more.",
"ck_rating_up":"1",
"voted":false
}

I don’t know of a straightforward way to read a JSON file into a desktop GIS package, so I needed to restructure the file into rows & columns.  I chose to do that with a series of Find/Replace statements in MS Word (perhaps there’s a better/more efficient way, but this approach worked), then added a row of field names, and saved the result as a .TXT file (one row of which is shown below):

id,lat,lon,neighborho,username,avatar,zipcode,reason,rating,voted
4830,40.742031,-73.777397,Fresh Meadows,David,,11355,There is no public transportation from Brooklyn-Queens greenway (Underhill Ave) to Flushing Meadows. By placing bike stations from Cunningham Park thru Kissena Park to Flushing Meadows will allow residents enjoy the parks more.,1,FALSE
  1. We’re primarily an ESRI shop at the CUNY Center for Urban Research (with periodic forays into open source, as well as a longstanding reliance on MapInfo for some key tasks).  So my next step was to convert this to a shapefile — which I did by using ArcGIS’s “Display X/Y Points” tool to create a point file based on the lat/lon values.
  2. Just in case there were multiple points at the same location, I ran the ArcGIS script called “Collect Events“, which aggregates point data based on location, and creates a new shapefile of each unique location with a count of all the points at each location.
    • I downloaded the JSON file a couple of times between Sept 19 and 25.  In the latest one (September 25, downloaded at 11pm) there were 55 points at latitude 40.7259, longitude -73.99 (a location at the intersection of E. 3rd Street and Second Ave in Manhattan).  But the user-supplied ZIP Codes and comments for most of these points indicated that they should have been all over the city.
    • Turns out this location is the center point of the Google map that’s displayed at the DOT bikeshare website.  If you zoom in on the DOT/OpenPlans map you’ll bring the map center into close view — and you can see the heavy map marker shadow due to all the points placed at that spot:

    • Presumably what happened here is that when you click the “Suggest Station” button, a marker is put at this spot by default. The marker is accompanied by a note that says DRAG ME! Then click ‘Confirm Station.’  But I’m assuming that 55 people didn’t drag the marker, but just left it there after they had entered their information. (I guess that’s not too bad — only 1 percent of the people using the site didn’t follow directions.)
    • Earlier this week (9/27) it looks like these sites were removed from the live map.  For my purposes, I removed those points from the shapefile, otherwise it would skew the analysis.  I could have put them somewhere in the ZIP Code that was entered with each spot, but I couldn’t be sure of the precise location (the reasons were vague regarding location), and I didn’t want to skew the analysis the other way.
  1. Other data notes:
    • There were 8 locations outside the immediate New York City area – some as far away as Montreal and Portland, Oregon.
    • The reason provided for Portland location was: “Even though it’s a whole continent from NYC it always seems to me like our cultures admire one another. I think NYC would enjoy all the benefits of positioning one of their Bike Share stations in Portland as sign of goodwill and mutual admiration.”
  1. There were also 53 points with lat/lon = 0, which I assumed was just a data entry/processing error.

Out of the 5,973 points as of 11pm September 25, after I removed the 55 locations and zoomed in on the points in or immediately near New York City (and omitting the 8 outside the city and the 53 with lat/lon=0), I ended up with 5,857 points.