The Cook County Assessor's Office is committed to transparency. To fulfill that commitment, the Data Department creates and maintains public assessment-related data sets. This document outlines these data, their uses, and pertinent source code. Data is published primarily on the Cook County Open Data Portal and through GitHub. The AssessR package leverages this open data in its documentation.
Releasing open data entails building reproducible data, documenting it, determining how and how often it should be updated, and receiving clearance from all necessary parties. These steps are outlined in further detail below.
Open data should, whenever possible, be automatically updated (views or tables in Athena), well-maintained, or immutable (political boundaries). Datasets that are released to the public must be scripted and the script must be version-controlled. To avoid creating discrepancies between internal and public use of the Department's data, data should not be built in a manner that is inconsistent with how it is used within the Department. Keep data as simple as possible - don't add columns that won't be useful for end-users and don't release redundant data.
The Data Department releases open data through the Cook County Open Data Portal, a service maintained by Socrata. For data that lives in the Department's warehouse, the preferred method for upload is through a provisioned Athena or S3 Socrata Gateway. When creating a new dataset there will be an opportunity to 'Connect to an External Data Source (Socrata Gateway)' where available data can be viewed. Make sure to add data to the portal using the Department's account (rather than a personal account) in order to properly establish ownership of the asset.
During the upload process set a schedule for updates (when applicable), add dataset metadata, and document, code/recode, and format columns. Recoding or 'transforming' columns entails writing SoQL blurbs while formatting columns is interactive within the 'Review & Configure Data' view. Choose a schedule that ensures the data is current but that doesn't needlessly pull data from AWS and run up monthly operating expenses - Socrata only offers daily to monthly update schedules, so if the data changes less frequently it should likely be updated manually (i.e. shapefiles). Adhere to the Department's asset naming convention by prefixing the asset with 'Assessor' and tag it with property tax and cook county assessor. Add documentation for the data to this wiki page.
It is highly recommended to create a 'Story' on Socrata or add to an existing one in order to describe and vignette the data for its intended audience.
Open data upload can also be automated using Socrata's SODA API and the RSocrata package.
Clearance must be received from the following parties before private assets on the open data portal can be made public:
- Chief Data Officer
- Executive Committee, FOIA, Comms (optionally, TPI)
- Bureau of Technology
BoT may have their own clearance requirements such as adding a 'Story' on Socrata or formatting common county-related columns in a particular way. Defer to their schemas.
The Data Department creates and maintains the following open data sets.
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | All | PIN, Year | Parcel | Monthly |
Notes: Contains a cornucopia of locational and spatial data for all parcels in Cook County.
Use cases: Joining parcel-level data to this dataset allows analysis and reporting across a number of different political, tax, Census, and other boundaries.
Code: default-vw_pin_universe.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | Regression-class | PIN, Card, Year | Residential Improvement | Bi-Weekly |
Notes: Residential PINs with multiple improvements (living structures) will have one card for each improvement.
Use cases: This data describes the location and physical characteristics of all single and multi-family improvements in the county. It can be:
- Used on its own to characterize the housing stock in a specific location
- Joined to assessments for analysis of assessments across geographies and housing types
- Joined to sales for the construction of hedonic home value estimates
Code: default-vw_card_res_char.sql
| Time frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | 299, 399 | PIN, Year | Condominium Unit | Bi-Weekly |
Notes:
Use cases: This data describes the location and physical characteristics of all condominium units in the county. Condominium units are associated with substantially less characteristic data than single and multi-family improvements. It can be:
- Used on its own to characterize the housing stock in a specific location
- Joined to assessments for analysis of assessments across geographies and housing types
- Joined to sales for the construction of hedonic home value estimates
Code: default-vw_pin_condo_char.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | All | Sale Deed Number | Parcel Sale | Monthly |
Notes: Refreshed monthly, though data may only change roughly quarterly depending on how often new sales are added to iasWorld.
Use cases: Alone, sales data can be used to characterize real estate markets. Sales paired with characteristics can be used to find comparable properties or as an input to an automated modeling application. Sales paired with assessments can be used to calculate sales ratio statistics. Outliers can be easily removed using filters constructed from class, township, and year variables.
Code: default-vw_pin_sale.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | All | PIN, Year | Parcel | Monthly |
Notes: Refreshed monthly, data is updated as towns are mailed/certified by Valuations and the Board of Review.
Use cases: Alone, can characterize assessments in a given area. Can be combined with characteristic data to make more nuanced generalizations about assessments. Can be combined with sales data to conduct ratio studies.
Code: default-vw_pin_history.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | All | PIN, Year | Parcel | Monthly |
Notes: Refreshed monthly, data is updated as towns are mailed/certified by Valuations.
Use cases: Alone, can be used to investigate appeal trends. Can be combined with geographies to see how AV shifts around the county between mailing and assessor certified stages.
Code: default-vw_pin_appeal.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 1999-Present | All | PIN, Year | Parcel | Monthly |
Notes: Refreshed monthly, data is updated as towns are mailed/certified by Valuations.
Use cases: Can be used for geocoding or joining address-level data to other datasets.
Code: default-vw_pin_address.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 2000-Present | All | PIN10, Year | Parcel | Annually |
Notes: Refreshed monthly, data is updated yearly as spatial files are made available.
Use cases: Can be used to isolate parcels by distance to specific spatial features.
Code: proximity-vw_pin10_proximity.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 2022-Present | All | PIN, Year | Parcel | Monthly |
Notes: Refreshed monthly, data is updated when necessary as PINs are re-classified.
Use cases: Can be used to study parcels that are exempted from paying property taxes.
Code: default-vw_pin_exempt.sql
| Time Frame | Property Classes | Unique By | Row | Updated |
|---|---|---|---|---|
| 2021 | — | Neighborhood Code | Neighborhood Polygon | Annually |
Notes: Refreshed yearly, but only changes with new neighborhood definitions. None are pending.
Use cases: Thematic mapping and location references.