By Seth Crider

Question: 

Do unemployed New Yorkers have adequate access to Workforce training, employment, and career improvement centers? 

Materials:

Voronoi Analysis: 

For this analysis I wanted to start at a more granular level by using census block data combined with the unique unemployment rates found through nyc’s open source labor statistics repository. For this first portion of the project I was more interested in the potential disbursement of the unemployed as to try and address more nuance and deviations between counties.

After creating the point file with my locations, I created the Voronoi Polygon with a 25% buffer. This was the closest I could get to a full and even polygon to cover the counties I had chosen in their entirety and ensure the divisions were sound. I was able to find the unemployment rate for each county (which differs from the blanket 4.2% for the entire city. The data by county fluctuates anywhere from 3-6 percent). With this I made a weight comprised of the adjusted rate percentage over the number of actual eligible workforce members per census block, and statistically how many of those people could be unemployed. Using the new columns with the information I then connected the Voronoi and my shape census/labor file via an intersection in QGIS, clipped both to each other and had a generally good idea about areas that seemed under represented. Due to the Polygons having assigned I.D’s— I could interpolate up from my granular rates using the new Voronoi features and determine not just general catchment, but an idea where the facilities are the most burdened. The polygons were then graduated based off of the number of unemployed within the catchment area. 

Note:

I took the current labor force numbers divided by the projected 2020 current population to produce a global percentage for each county. I applied the rate to the 2010 census block population data (categorized by county) to make up my labor force division among all census blocks

From there I took current unemployment rates (individual to each county) and applied them to eligible workforce. 

Euclidean and Isochrone Analysis:  

I decided to stick with 1.5 miles as my general radius distance from centers. I contemplated 1 mile (as most tend to stick within those limits) but I am also well aware of alternative transportation and wanted to give a bit more leeway within the numbers for those who may rideshare, skooter, bike, skateboard, etc. For both the euclidean, and isochrone maps I dissolved both shapefiles, clipped my census block file to each and calculated the unemployed within each area. 

My location data had addresses so I used them to generate Isochrones from the third party website with 40 minutes as the specified timespan. I believe this is being very generous–but as someone who commutes about that time every day, I know it’s not unreasonable or unheard of. It is important to note that I kept the metrics within the realms of walking mostly because taxis, ubers, and public transport can be super expensive and an unrealistic expectation for someone on a fixed income (or no income). After dissolving the isochrones into one shape I then performed the same functions as I did with the euclidean shapefile to get a number of potential unemployed within the catchment area. 

Road Network Analysis:

Service Area of Staten Island Southshore Career Workforce Center

I specified the road network service area as the same distance as both the isochrone and the buffer area (1.5 miles). In the comparison, it seems the buffer in this case may actually be closer in total area to the road network service area. I feel like the closest approximation resides somewhere within the isochrone analysis. We are dealing with very fluid and dynamic movements of people so there certainly is no perfect solution – however – the Isochrone included more of the smaller offshoot streets and neighborhoods that I think logically would not be too far from a complete service area– especially since those areas are surrounded by natural boundaries (Arden Heights Woods, Ocean View Cemetery, Nature Reserve etc.). The boundaries would most likely prevent people from traveling anywhere but in the direction of the service center.  

Of course, in other circumstances like centers surrounded by a high volume of residential streets and less natural boundaries other analysis like the road network, and buffer would be ideal. In this case, future research could be invested in adjusting the distance requirements on all three of the area analysis (Buffer, Isochrone, and Road Network) to see how much the area fluctuates between each other with smaller or larger parameters. Doing so would provide a better understanding of the unique limitations of each of these methods and the circumstances for they should be considered.  

Discussion:

Performing all three analysis was a good exercise to remind yourself of how arbitrary constructed statistical boundaries can be. Especially if you have to develop the methods yourself without the consult of others who may be more expert on the topic of research. In this way, the process becomes more democratic in defining catchment and allows for more variances within the numbers which is great. For this analysis I was keen on the results from the isochrone because we live in a city and the boundaries are far more fluid as far as shaping pathways and routes and also more attuned to the reality of the unemployed (fewer cars and more personal mobility). The Voronoi Polygon analysis was a nice as a first pass to determine general idea around access to locations and which areas might be worth considering and paying attention to throughout the analysis. (Like Queens) 

Next Steps: 

I would like to conduct more research specifically on the details of unemployed populations in NYC. For instance where do they look for jobs and opportunities? Are they willing to travel, or commute long distances for employment or help. How many have consulted a career center before? How about marginalized people with felony records or history of drug abuse? Do they have options? 

It would also be helpful to consult the centers themselves. 

How many people do they serve? Do they advertise? Are they successful in job placement?

By incorporating other figures like net household income, and how many unemployed are providing for families could help illustrate where city resources could be focused. Alternative aid like getting people into centers through transport, remote consult. etc. could be a potential future. `

Why is this important?

Most have us have lived through periods of unemployment. How did it feel, and where did you go? Although temp agencies, and quick jobs have been a go-to for decades –they hardly ever provide real sustainable development. Vocational, and professional centers are important in combating those first steps into poverty, and financial insecurity. They give hope, and a plan of action. They also operate as an important gatekeeper between companies and the pool of the selectable workforce by  finding good fits, and ushering in mutually beneficial agreements and flows.