Skip to content
FourthSpace

Sacramento opportunity map: loneliness, places to gather, and who to call first

Which Sacramento County neighborhoods combine high loneliness with few places to gather within a short walk? Where do lonely adults cluster? And which venues should FourthSpace contact first?

Analysis by Andrew Ly (opens in a new tab) · Updated

Key finding

about 475,000

Estimated adults who feel lonely at least sometimes (tract sum)

About 475,000 Sacramento County adults feel lonely at least sometimes (estimate).

Tracts are split into thirds on each measure. Walk access is the share of a tract's intersections within an 800 m walk of an indoor gathering place.

  • Higher loneliness, low venue access
  • Special land-use tract (not ranked)

sacramento_tracts_final.gpkg

Findings

  1. About 475,000 Sacramento County adults feel lonely at least sometimes (estimate).

    Summed across 362 tracts, this matches the county-level estimate in Project 04. Tract rates run from 30.5% to 49.6%.

  2. Loneliness clusters.

    Global Moran's I is 0.57 (p = 0.001). 74 tracts form significant hot spots. Most sit in south Sacramento (approximate ZIPs 95828, 95823, 95824, 95822 and 95832) and north Sacramento (95838 and 95815). Together they hold an estimated 111,000 lonely adults.

  3. 36 tracts pair high loneliness with poor walk access.

    They hold an estimated 52,000 lonely adults. Their median walk access is 0%, meaning no intersection in the typical priority tract sits within a 10-minute walk of an OSM-listed indoor gathering place. Only 4 such venues are mapped inside all 36 tracts combined.

  4. Most residents lack a gathering place within a 10-minute walk.

    Weighted by population, 29.5% of a tract's intersections are within 800 meters of an indoor gathering place. 164 of 362 tracts contain none.

  5. In this county, loneliness tracks age and income more than living alone.

    Across tracts, PLACES loneliness correlates -0.73 with median age and -0.61 with median household income. It correlates -0.20 with the share of adults living alone. Hot spot tracts have a median household income of $67,600, against $90,100 countywide.

  6. The outreach list has 765 named venues.

    544 are likely independents. 302 are cafes, 52 are libraries, 53 are bars and 41 are pubs. 118 independent venues sit in priority or hot spot tracts. Top independents by nearby estimated lonely adults include Yogurt Time (Walerga Rd), Southgate Community Library, Caffeine Society (Stockton Blvd), Antelope Library, Bakers Donuts (Florin Rd), The Flame Cafe (Florin Rd), Valley Hi-North Laguna Library, Jose P. Rizal Community Center and Eagles #9 (Fruitridge Rd).

Options

  1. Option A. Launch in the hot spots

    My pick

    Start the pilot in south Sacramento (ZIPs 95823, 95824 and 95828, along the Florin Rd and Stockton Blvd corridors), where hot spot tracts hold the most lonely adults and several venues already exist.

  2. Option B. Launch where venues are dense

    Start in Midtown and Downtown, where walk access is highest. Early users get more places per screen. Loneliness is lower there.

  3. Option C. Libraries first

    Pitch Sacramento Public Library branches near the hot spots, starting with Southgate and Valley Hi-North Laguna, as anchor venues.

  4. Option D. Independents in priority tracts first

    Work the 118 independent venues in priority or hot spot tracts with a short pitch and a free listing.

  5. Option E. Fill the map gaps

    Before outreach, spend a day adding missing venues in priority tracts to OSM and to the FourthSpace database.

Default pick: A, starting with C inside it. The data says the need is greatest in south and north Sacramento. Opening there with library branches as anchors gives free space and events from day one, and the nearby cafes and community centers on the list fill out the map. B is easier but points the product at the people who need it least.

Method

  1. 01_build_tracts.py joins tract shapes, PLACES estimates and ACS measures. It also computes estimated lonely adults per tract (PLACES crude prevalence times adults). That figure is an estimate.
  2. 02_venues_access.py sorts OSM features into three groups: indoor gathering places (652), parks (816) and fitness or sports venues (159). It builds a walk network from 229,174 road nodes, leaving out interstates and ramps. It then finds the network distance from every intersection to the nearest indoor gathering place. A tract's walk access is the share of its intersections within 800 meters, about a 10-minute walk.
  3. 03_priority_and_outreach.py runs local Moran's I on tract loneliness to find hot spots. It then flags priority tracts, meaning the top third for loneliness and the bottom third for walk access. Tract 9883, a special land-use tract, is left out of rankings. Finally it ranks every named indoor or fitness venue by the estimated lonely adults living in tracts centered within 1.6 km and marks likely chains (names that appear three or more times).
  4. 04_maps.py draws the PNG maps and the interactive HTML map.

Tools

Python 3.13, geopandas, shapely, networkx (walk network and multi-source shortest paths), SciPy (nearest-node lookup), libpysal and esda (local Moran's I), matplotlib (PNG maps), Plotly with MapLibre (interactive map, Carto basemap). OSMnx was planned but needs live Overpass access from the analysis machine, so the walk network is built from TIGER roads.

Caveats

  1. PLACES tract values are model-based estimates. They lean heavily on each tract's age, sex, race and income makeup, which is why loneliness tracks age and income so closely. Spatial smoothing in the model also raises Moran's I. Treat hot spots as places where risk factors cluster.
  2. OpenStreetMap is incomplete. It lists 313 cafes where County Business Patterns counts 501 coffee shops and snack bars, and full-service restaurants were not queried. Walk access is understated, most in suburbs that mappers cover less.
  3. TIGER roads miss many footpaths, crossings and park trails, so network distances run a bit long. Walk access counts intersections. It does not count homes.
  4. The lonely-adults reach number per venue is an estimate for ranking. Tracts within 1.6 km are counted in full, so neighboring venues share the same people.
  5. OSM addresses are missing for many venues. 45% have a website and 37% a phone number. The list needs a manual pass before outreach.
  6. Chain detection is a simple name count. A few independents with common names may be flagged, and a few chains with one location may not.

Sources

  1. CDC PLACES 2025 release, census tract GIS file (loneliness, lack of support, frequent mental distress, lack of transportation)

    Year
    BRFSS 2023 based
    Geography
    363 Sacramento County tracts

    https://www.cdc.gov/places (opens in a new tab)

    Public domain

  2. ACS 2020-2024 5-year, table-based summary files (B09021 living arrangements, B25044 vehicles, C17002 poverty, B19013 income, B01002 age, B25003 tenure, B08301 commute)

    Year
    2020-2024
    Geography
    Sacramento County tracts

    https://www2.census.gov/programs-surveys/acs/summary_file/2024/table-based-SF/ (opens in a new tab)

    Public domain

  3. OpenStreetMap third places via the Overpass API (1,632 features: cafes, libraries, community centers, bars, pubs, bakeries, bookstores, parks, gyms and more)

    Year
    Downloaded Nov 14, 2026
    Geography
    Sacramento County

    https://www.openstreetmap.org/copyright (opens in a new tab)

    Publisher's terms

  4. Census cartographic boundary tracts (cb_2024_06_tract_500k) and TIGER/Line 2024 roads

    Year
    2024
    Geography
    Sacramento County

    https://www.census.gov/geographies/mapping-files.html (opens in a new tab)

    Public domain

  5. City of Sacramento open data: parks and community centers (context layers)

    Year
    2026
    Geography
    City of Sacramento

    https://data.cityofsacramento.org (opens in a new tab)

    Publisher's terms

Interactive charts

  • Sacramento County: loneliness by tract and places to gather

Keep reading

The second place went quiet: remote work and daily company

How much of American work moved home, and how much did it stay there? What does a remote workday do to the time people spend around others? And where does Sacramento stand, given its large state workforce and the July 2026 return-to-office order?

Next study
On a remote workday, people spend about an hour with anyone outside their household

Be first through the door.

We're opening in Sacramento first. Join the list and we'll email you when we open.

Fields marked * are required.

Places you'd go (optional)
How we use your details

Study06/07Join the waitlist