RRA
Data Products Distress & Urban Decay Assessment Value Spatial Pattern Dataset

Distress & Urban Decay

Assessment Value Spatial Pattern Dataset

Parcel-level Local Indicators of Spatial Association (LISA) computed on assessor fair market values across 230,000+ tri-parish parcels — identifying statistically significant high-value and low-value spatial clusters at the parcel level.

230K+
Parcels
Tri-Parish
Coverage
LISA
Method
Quarterly
Refresh

About this product

This dataset applies Local Indicators of Spatial Association (LISA), a well-established spatial statistics method developed by Luc Anselin (1995), to assessor fair market values across all parcels in Caddo, Bossier, and DeSoto parishes. The result is a parcel-level classification of spatial clustering in assessed values — identifying areas where high-value properties cluster together (HH), low-value properties cluster together (LL), and transition zones where valuations diverge from their immediate neighborhood context (HL, LH).

Each parcel is assigned a LISA quadrant for three value dimensions: land value, improvement value, and the differential between the two. Z-scores quantify how far each parcel's value deviates from the parish mean, and p-values indicate whether the observed clustering is statistically significant. Spatial gradients describe the directional pattern of value change in each parcel's local neighborhood.

The dataset is produced at parcel grain and delivered as a flat file. It contains no derived scores or editorial conclusions — the LISA quadrant, z-score, and raw FMV figures are the output. Interpretation is left to the buyer.

Buyers

  • Tax and real estate attorneys researching assessed value patterns for appeals or litigation support
  • Academic and policy researchers studying spatial patterns in property valuation
  • CDFIs and community development organizations mapping investment climate by neighborhood
  • Journalists and data professionals analyzing assessor records

Delivery

  • ·CSV — one row per parcel with all LISA fields and raw FMV values
  • ·GeoJSON — centroid-point file for direct mapping import, delivered by parish
  • ·Quarterly refresh aligned to assessor data updates
  • ·Tri-parish coverage: Caddo, Bossier, DeSoto

Data Preview

Improvement value spatial clusters — central Shreveport. Green = high-value clusters (HH); red = low-value clusters (LL); yellow/blue = transition zones.

LISA cluster map — central Shreveport

Significant cluster counts across 215K+ tri-parish parcels. 62,748 parcels in statistically significant low-value clusters vs. 10,873 in high-value clusters.

LISA quadrant distribution

Dataset Fields

parcel_id
Parcel identifier (GEOGNO for Caddo; parcel_id for Bossier and DeSoto)
parish
caddo / bossier / desoto
address / owner_name
Assessor-reported situs address and owner of record
land_fmv / improved_fmv / total_fmv
Assessor fair market value — land, improvements, and combined total
differential_fmv
Improvement value minus land value — indicates structure-dominated vs. land-dominated parcels
total_assessed_value
Total assessed value on record: land + improved (Caddo); assessor-reported assessed value (Bossier/DeSoto)
assessment_ratio
assessed_value / market_value — available for Bossier and DeSoto; NULL for Caddo (single assessor valuation figure)
land_lisa / improved_lisa / diff_lisa
LISA spatial cluster classification — HH (high-high), LL (low-low), HL (high surrounded by low), LH (low surrounded by high), NS (not statistically significant)
land_z / improved_z / diff_z
Z-score of parcel value relative to the parish population mean and standard deviation
land_lisa_p / improved_lisa_p
Moran's I p-value — statistical significance of the observed spatial cluster
land_gradient / improved_gradient
Directional spatial gradient of assessed values in the parcel's local neighborhood
lat / lon
Parcel centroid coordinates (WGS84)
Request Sample →