Data

Where King County homeless survey respondents were last housed, 2017-2024

Shares of homeless people in King County, Washington, who told the county's point-in-time surveys they were last housed in the county, elsewhere in Washington, in another state or abroad, with why newcomers said they came, how long locals had lived here, and comparable figures from other West Coast places.

Coverage: Count Us In surveys of January 2017, 2018 and 2019; KCRHA survey of January-February 2024

What this dataset measures

The county's own homeless count surveys are the only regular source that asks homeless people where they lived before. We use every year that published results, keep each year's wording, and add the method checks a reader needs to judge the 2019-to-2024 change.

Curated data tables

These related tables form this versioned dataset release. Each table has its own downloadable file and field definitions.

Where respondents were last housed, King County surveys, 2017-2024

Answers to each year's residence question, as counts and shares of people who answered. 2017-2019 asked where people lived when they most recently became homeless (sheltered and unsheltered); 2024 asked where they last had stable housing (unsheltered only).

Unit of observation: One survey year

View this table

king-county-by-year.csv text/csv · 1.1 KB (1,126 bytes)
19 field definitions
FieldTypeMeaningUnitsValue type
year integer Year of the count — observed
survey string Survey name — observed
universe string Who was surveyed, with the share of surveys taken in shelter or transitional housing — observed
question string Exact question wording — observed
answered integer People who answered the residence question people observed
surveyed integer All completed surveys that year people observed
answer_rate_pct number answered / surveyed percent derived
king_county integer Answered King County (2024: any King County subregion) people observed
elsewhere_wa integer Answered another Washington county (2017-2019: Pierce, Snohomish, Thurston or another county) or Washington State outside King County (2024) people observed
other_state integer Answered out of state (2017-2019) or United States outside Washington (2024) people observed
abroad integer International (2024 only; not an answer choice in 2017-2019) people observed
unknown integer Answers recorded as Unknown (2024: 7) people observed
king_county_pct number king_county / answered percent derived
elsewhere_wa_pct number elsewhere_wa / answered percent derived
other_state_pct number other_state / answered percent derived
abroad_pct number abroad / answered; empty where not asked percent derived
outside_county integer elsewhere_wa + other_state + abroad people derived
outside_county_pct number outside_county / answered percent derived
outside_county_pct_of_known number outside_county / (answered - unknown) percent derived

Unsheltered people by last region of stable housing, 2024 (KCRHA Table 13)

Rows of Table 13 as printed, with the group each row falls in.

Unit of observation: One region

View this table

4 field definitions
FieldTypeMeaningUnitsValue type
last_region string Region as printed — observed
group string King County, Elsewhere in Washington, Another U.S. state, Another country or Not known — derived
responses integer Responses people observed
pct number responses / 821 (the table prints the same to two decimals) percent derived

Why people from outside King County moved here, 2017 and 2018

Answers to 'If from outside the county, why did you move to King County?' Each person gave one answer; the shares add to 100%.

Unit of observation: One answer in one year

View this table

reasons-moved-2017-2018.csv text/csv · 806 bytes
6 field definitions
FieldTypeMeaningUnitsValue type
year integer Survey year — observed
reason string Answer choice as printed — observed
people integer People giving this answer people observed
answered integer People who answered the question people observed
pct number people / answered percent derived
pct_of_residence_answers number people / everyone who answered the residence question that year (866 in 2017, 888 in 2018) percent derived

Time in King County before becoming homeless, 2018 and 2019

Answers to 'If in King County, how long were you living in King County before you became homeless?'

Unit of observation: One answer in one year

View this table

6 field definitions
FieldTypeMeaningUnitsValue type
year integer Survey year — observed
time_in_king_county string Answer choice — observed
people integer People giving this answer people observed
answered integer People who answered people observed
pct_printed number Percent as printed (whole percents in 2019) percent observed
pct number people / answered, to one decimal percent derived

How far method differences could move the share

Our arithmetic on the published counts: the most the 2019 figure could rise if sheltered respondents are set aside, and the range the 2024 figure could take if the 645 non-answers had answered.

Unit of observation: One check

View this table

method-bounds.csv text/csv · 632 bytes
4 field definitions
FieldTypeMeaningUnitsValue type
check string What is assumed — derived
basis string The arithmetic — derived
outside_county_pct number Resulting share last housed outside King County percent estimated
meaning string What the figure means — derived

Share last housed outside the area, West Coast surveys

The latest published figure from each place's homeless survey for the share who were last housed (or became homeless) outside the county or city. Questions and who was surveyed differ; see each row.

Unit of observation: One place in one survey year

View this table

west-coast-comparison.csv text/csv · 1.8 KB (1,866 bytes)
9 field definitions
FieldTypeMeaningUnitsValue type
label string Short chart label: place and survey year — derived
place string Place surveyed — observed
year integer Survey year (California study: fieldwork ended 2022) — observed
outside_pct number Share last housed outside the county or city percent derived
local_pct number Share last housed in the county or city, as printed (Multnomah: our count) percent observed
who string Who was surveyed — observed
basis string What the figure rests on — observed
question string Question or the report's wording of it — observed
source_id string Source id in this dataset — observed

San Francisco: where people were housed before becoming homeless, 2019-2024

San Francisco's surveys by the same firm, sheltered and unsheltered, in three counts.

Unit of observation: One survey year

View this table

san-francisco-2019-2024.csv text/csv · 235 bytes
8 field definitions
FieldTypeMeaningUnitsValue type
place string San Francisco — observed
year integer Survey year — observed
local_pct number Housed in San Francisco percent observed
outside_pct number 100 - local_pct percent derived
rest_of_state_pct number Another California county percent observed
other_state_pct number Out of state percent observed
answered integer People who answered, where printed people observed
source_id string Source id — observed

Supporting files

Source snapshots, manifests, quality reports, documentation, and other materials used to produce or review this release.

  • dataset-metadata.json Documentation · application/json · 55.2 KB (56,494 bytes)

    This dataset's sources, table definitions and saved analyses as one JSON file.

Saved analyses

Where King County homeless survey respondents were last housed, 2017-2024

Question: What share of respondents were last housed outside King County, and has it changed?

Result: In 2024, 39.3% of unsheltered respondents (323 of 821) said they were last stably housed outside King County: 19.9% elsewhere in Washington, 18.3% in another state and 1.2% abroad. In 2017-2019, 23%, 17% and 16% of respondents said they were living outside the county when they became homeless.

Operation: Counts as printed, grouped into King County, elsewhere in Washington, another U.S. state, another country and not known; each share = count / people who answered that year

Scope: Count Us In surveys 2017-2019 (sheltered and unsheltered) and KCRHA's 2024 survey of unsheltered people

View this analysis

  • The 2024 survey differs from 2017-2019 in who was surveyed, how they were found and how the question was worded, so the change from 2019 to 2024 mixes real change with method change.
  • Only 56% of 2024 respondents answered the question (821 of 1,466).
  • KCRHA says the 2024 answer has no date, so for some people the last stable housing may be years before the survey.

Share last housed outside King County, by year

Question: What share were last housed outside the county each year?

Result: 22.9% (2017), 16.9% (2018), 16.4% (2019), 39.3% (2024; 39.7% of answers with a known place).

Operation: outside_county / answered

Scope: As in last-housed-by-year

View this analysis

  • Same as last-housed-by-year.

Unsheltered people by last region of stable housing, 2024

Question: Where in or outside King County were unsheltered people last stably housed?

Result: Seattle Metro 26.4% and South King County 22.7% were the largest King County answers; 19.9% named elsewhere in Washington, 18.3% another state and 1.2% another country.

Operation: Table 13 rows as printed

Scope: Unsheltered people surveyed for the 2024 count who answered (821)

View this analysis

  • Raw answers from hub-based surveys, not weighted to the whole unsheltered population.

Why people from outside the county moved here, 2017 and 2018

Question: Did people who came from outside King County say they came for homeless services?

Result: 'To access homeless services and/or benefits' was the top reason in 2017 (25.4%, 50 of 197) and second in 2018 (20.4%, 28 of 137), after work (31.4%). Family or friends was about 18% both years. Those who named services were 5.8% (2017) and 3.2% (2018) of everyone who answered the residence question.

Operation: Counts as printed; shares of people who answered

Scope: Count Us In respondents who were from outside King County and answered, 2017 (197) and 2018 (137)

View this analysis

  • Small numbers (137-197 people).
  • Self-reported reasons; people could pick only one.
  • The answer combines homeless services with benefits, so it does not separate shelter from other public aid.

Time in King County before becoming homeless, 2018 and 2019

Question: Among those last housed in King County, how long had they lived there?

Result: In 2019, 31% were born or grew up in King County and 15% had lived there 10 years or more; 19% had lived there less than a year (16% in 2018).

Operation: Counts as printed

Scope: Count Us In respondents who were living in King County when they became homeless and answered, 2018 (712) and 2019 (752)

View this analysis

  • Asked only of people who said King County; people from outside are not in these rows.

How far method differences could move the share

Question: Could the switch to unsheltered-only surveys, or the missing 2024 answers, explain the change?

Result: Even if every sheltered 2019 respondent was a local, the 2019 unsheltered share could be at most about 24.1%, below 2024's 39.3%. But if the 645 people who skipped the 2024 question were all locals, 2024's share of all surveyed would be 22.0%.

Operation: Arithmetic on published counts

Scope: 2019 and 2024 surveys

View this analysis

  • Does not test the change in question wording, which could move the share either way.
  • Non-answers were not random; KCRHA's answer list includes 'Data not collected', so some people were never asked.

Share last housed outside the area, West Coast surveys

Question: How does King County compare with other West Coast places that ask a similar question?

Result: King County's 39% is close to Portland's Multnomah County (42%, 2023), San Francisco (41%, 2024) and Los Angeles (36%, 2020). Alameda County, Sacramento and San Diego report about 19-22%, and a statewide California study found 25%.

Operation: Latest published figure per place; outside share = 100 minus the local share as printed (King County and Multnomah County from counts)

Scope: Eight West Coast places or studies, surveys from 2020 to 2026

View this analysis

  • Not a like-for-like ranking: question wording, survey method and year differ.
  • Sacramento and San Diego figures come from summaries without the question or sample.

San Francisco, 2019-2024

Question: Did a city that kept the same survey firm see a similar rise?

Result: In San Francisco the share housed outside the city before becoming homeless rose from 30% (2019) and 29% (2022) to 41% (2024).

Operation: As printed

Scope: San Francisco surveys, sheltered and unsheltered, 2019, 2022, 2024

View this analysis

  • San Francisco's 2026 survey (new firm and method) found 44% outside the city; it is not in this table.

Sources

  1. All Home and Applied Survey Research, 2017 Seattle/King County Count Us In (Point-in-Time Count) comprehensive report Where survey respondents were living when they most recently became homeless, and why people from outside the county moved here (p.29, Appendix 3 p.97-98)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the county's annual homeless count and survey, written by Applied Survey Research for All Home, the county's homelessness planning body at the time. Survey answers are unweighted shares of people who answered; peer surveyors used quotas and approached every third person they met, which the 2019 report says was not a random sample.

  2. All Home and Applied Survey Research, 2018 Seattle/King County Count Us In report Residence when people became homeless, time in King County before becoming homeless, and reasons for moving (p.30, Appendix 3 p.118-119)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the county's annual homeless count and survey, written by Applied Survey Research for All Home, the county's homelessness planning body at the time. Survey answers are unweighted shares of people who answered; peer surveyors used quotas and approached every third person they met, which the 2019 report says was not a random sample.

  3. All Home and Applied Survey Research, 2019 Seattle/King County Count Us In report Residence when people became homeless and time in King County before becoming homeless (p.29 Figures 22-23, Appendix C p.100)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the county's annual homeless count and survey, written by Applied Survey Research for All Home, the county's homelessness planning body at the time. Survey answers are unweighted shares of people who answered; peer surveyors used quotas and approached every third person they met, which the 2019 report says was not a random sample.

  4. All Home, King County and Applied Survey Research, Count Us In 2020 report The 2020 survey asked where people last had stable housing (Appendix E, question 18) but the report publishes no result

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the county's annual homeless count and survey, written by Applied Survey Research for All Home, the county's homelessness planning body at the time. Survey answers are unweighted shares of people who answered; peer surveyors used quotas and approached every third person they met, which the 2019 report says was not a random sample.

  5. King County Regional Homelessness Authority (KCRHA), 2024 Point-in-Time Count, King County, revised report (corrected March 2026) Unsheltered respondents by the region where they were last stably housed (Table 13, p.79), with the question wording (Appendix I, p.126) and caveats (p.78-79)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the regional homelessness authority, the agency that runs the county's homeless services and the count. Unsheltered people were surveyed at 20 hub sites using respondent-driven sampling (each respondent refers others) designed with University of Washington researchers; Table 13 gives raw answer counts, not weighted estimates.

  6. KCRHA, 2026 Point-in-Time Count executive report (June 2026) Has no last-housed figures; says a comprehensive report will follow later in the year

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Same agency as the 2024 report.

  7. Portland State University for Multnomah, Clackamas and Washington counties, 2023 Tri-County Point-in-Time Count Findings Report (April 2024) Unsheltered people by where they last lived in housing, by county (Table 4.9, p.97-98), and the main reason people from outside moved to the area (Table 4.10)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Written by university researchers for the three Portland-area counties, which fund homeless services. The Multnomah rows of Table 4.9 print percentages that don't match their counts; we used the counts.

  8. San Francisco Department of Homelessness and Supportive Housing, 2024 Point-in-Time Count slide deck Share housed in San Francisco before becoming homeless, 2024, with earlier years (slide 49)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the city's homeless count and survey, written by Applied Survey Research (the same firm that ran King County's Count Us In surveys) for the city's homelessness department.

  9. San Francisco, 2019 Homeless Count and Survey comprehensive report Place of residence at the time of housing loss, 2019 (p.18)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the city's homeless count and survey, written by Applied Survey Research (the same firm that ran King County's Count Us In surveys) for the city's homelessness department.

  10. San Francisco, 2022 Point-in-Time Count report Place of residence at the time of housing loss, 2022 (p.30, Figure 17)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. Report of the city's homeless count and survey, written by Applied Survey Research (the same firm that ran King County's Count Us In surveys) for the city's homelessness department.

  11. Los Angeles Homeless Services Authority (LAHSA), 2020 Greater Los Angeles Homeless Count presentation Place of residence before becoming homeless, unsheltered adults 25 and older (chart, PDF p.24)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. The county-city homelessness agency's own presentation; figures are read from a chart. LAHSA's later presentations publish no residence figures, though its survey still asks.

  12. UCSF Benioff Homelessness and Housing Initiative, California Statewide Study of People Experiencing Homelessness, executive summary (June 2023) Share of homeless adults in California who lived in the same county as their last housing

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Academic. A university study funded by the state's health and human services agency, with a representative sample of about 3,200 adults in eight counties, October 2021 to November 2022.

  13. Alameda County, 2024 Point-in-Time Count Report Share of surveyed unsheltered people whose previous permanent address was in the county (p.20-21)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Government. The county's own count report; the survey was given to unsheltered people on the morning of the count.

  14. Sacramento Steps Forward, 2026 Point-in-Time Count data summary Share who first became homeless while living in Sacramento County (p.1)

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Other. The nonprofit that runs the county's homeless system and count. The summary does not say who answered or the exact question.

  15. Regional Task Force on Homelessness (San Diego), 2025 Point-in-Time Count release Share who said they began experiencing homelessness in San Diego County

    Retrieved November 13, 2026. Public record (local government report). Public record; figures restated, document linked

    Reliability: Other. The body that runs the region's homeless system and count; a release, not a full report, and it does not give the question or who answered.

License: FactHub's selection, cleaning and combined tables are licensed under CC BY 4.0. Source data keeps the terms listed above (terms).

Limitations and reproducibility

Reproducible from public sources: Every count is printed in the cited reports (page numbers in each source's role); shares and checks are simple arithmetic shown in each table's derivation.

  • The 2024 survey changed who was asked (unsheltered people only), how they were found (referral chains from hub sites instead of street and shelter quotas) and the question (last stable housing instead of where people lived when they became homeless).
  • Only 56% of 2024 respondents answered the residence question; none of the figures are weighted.
  • Self-reported. KCRHA notes the 2024 answer carries no date, so last stable housing could be years before the survey.
  • No King County survey since 2018 asks why people came, and none asks whether people became homeless before or after arriving.
  • 2020's survey asked the question but published no result; 2022 and 2026 have no published residence figures as of November 13, 2026.

Publications using this dataset

Version details

The exact citing versions, and the version of this dataset each cites.

  • Article king-county-homeless-last-housed-outside — version 6 (published) cites version 3
  • Article king-county-homeless-last-housed-outside — version 5 (published) cites version 3
  • Chart king-county-homeless-last-housed-by-year — version 1 (published) cites version 3
  • Chart west-coast-homeless-last-housed-outside — version 1 (published) cites version 3

Node: dataset:king-county-homeless-last-housed