search

NYCPlanning / db-equitable-development-tool

Data Repo for the equitable development tool (EDDT)
MIT License
0 stars 0 forks source link

Update source data and logic for creating units_rad in Housing Security and Quality #265

Closed AmandaDoyle closed 2 years ago

AmandaDoyle commented 2 years ago

units_rad Comes All Counted Units field in Housing New York Units by Building We have logic to assign PUMA and Borough for each record. SUM All Counted Units by each geographic level.

mbh329 commented 2 years ago

@levysamu I don't quite understand why we are using the housing new york dataset to getunits_rad. I thought the housing new york dataset documents the total number of affordable units that HPD is responsible for while RAD units relates to the specific number of nycha units that have gone through the federally backed RAD conversion program

levysamu commented 2 years ago

We are not using HNY for RAD units. We aren't using RAD units at all for anything.

The problem now is that there are two fields in the Income Restricted Units table shown below, but the bottom one is pulling the wrong information. Units With Regulatory Agreements Since 2014 is currently displaying RAD units when it should be showing HNY units. We don't need to display RAD units anywhere in the EDDE. We need all NYCHA units (of which RAD is a subset) and all HNY units.

image

mbh329 commented 2 years ago

Okay, great. I think that issue was lost in translation but it is clear now

mbh329 commented 2 years ago

@AmandaDoyle @levysamu I am working through the new logic to create this new data point, this is quite different than our previous process. For the housing new york data, there are a good chunk of records that HPD classifies as confidential where no lat/long or address is given therefore no PUMA can be assigned via our current process (a borough column is included in the source data). Citywide the confidential projects account for 1561 units out of 194481 total units. Do we want to keep these confidential units in the data or should they be filtered out?

I should be able to convert the NTA column (which is included) to puma geography but there will be records that can't be geocoded down to puma. Should I include these in the larger geographic aggregations?

AmandaDoyle commented 2 years ago

@mbh329 The logic for filtering and assigning records to a geographic area should be the same as this indicator as source data is the same. Therefore: 1) Do not filter out confidential projects 2) If the project has borough information include the records at the Brough or Citywide level