When analyzing the ratio of homelessness to state population, New York, Vermont, and Oregon had the highest rates in 2023. However, Washington, D.C. had an estimated ** homeless individuals per 10,000 people, which was significantly higher than any of the 50 states. Homeless people by race The U.S. Department of Housing and Urban Development performs homeless counts at the end of January each year, which includes people in both sheltered and unsheltered locations. The estimated number of homeless people increased to ******* in 2023 – the highest level since 2007. However, the true figure is likely to be much higher, as some individuals prefer to stay with family or friends - making it challenging to count the actual number of homeless people living in the country. In 2023, nearly half of the people experiencing homelessness were white, while the number of Black homeless people exceeded *******. How many veterans are homeless in America? The  number of homeless veterans in the United States has halved since 2010. The state of California, which is currently suffering a homeless crisis, accounted for the highest number of homeless veterans in 2022. There are many causes of homelessness among veterans of the U.S. military, including post-traumatic stress disorder (PTSD), substance abuse problems, and a lack of affordable housing.

The graph displays the top 15 states by an estimated number of homeless people in the United States for the year 2025. The x-axis represents U.S. states, while the y-axis shows the number of homeless individuals in each state. California has the highest homeless population with 187,084 individuals, followed by New York with 158,019, while Hawaii places last in this dataset with 11,637. This bar graph highlights significant differences across states, with some states like California and New York showing notably higher counts compared to others, indicating regional disparities in homelessness levels across the country.

Context

This database contains the data reported in the Annual Homeless Assessment Report to Congress (AHAR). It represents a point-In-time count (PIT) of homeless individuals, as well as a housing inventory count (HIC) conducted annually.

The data represent the most comprehensive national-level assessment of homelessness in America, including PIT and HIC estimates of homelessness, as well as estimates of chronically homeless persons, homeless veterans, and homeless children and youth.

These data can be trended over time and correlated with other metrics of housing availability and affordability, in order to better understand the particular type of housing resources that may be needed from a social determinants of health perspective.

HUD captures these data annually through the Continuum of Care (CoC) program. CoC-level reporting data have been crosswalked to county levels for purposes of analysis of this dataset.

Querying BigQuery tables

You can use the BigQuery Python client library to query tables in this dataset in Kernels. Note that methods available in Kernels are limited to querying data. Tables are at bigquery-public-data.sdoh_hud_pit_homelessness

Sample Query

What has been the change in the number of homeless veterans in the state of New York’s CoC Regions since 2012? Determine how the patterns of homeless veterans have changes across the state of New York

homeless_2018 AS ( SELECT Homeless_Veterans AS Vet18, CoC_Name
FROM bigquery-public-data.sdoh_hud_pit_homelessness.hud_pit_by_coc WHERE SUBSTR(CoC_Number,0,2) = "NY" AND Count_Year = 2018 ),

veterans_change AS ( SELECT homeless_2012.COC_Name, Vet12, Vet18, Vet18 - Vet12 AS VetChange FROM homeless_2018 JOIN homeless_2012 ON homeless_2018.CoC_Name = homeless_2012.CoC_Name )

SELECT * FROM veterans_change

Yearly statewide and by-Continuum of Care total counts of individuals receiving homeless response services by age group, race, gender, veteran status, and disability status.

This data comes from the Homelessness Data Integration System (HDIS), a statewide data warehouse which compiles and processes data from all 44 California Continuums of Care (CoC)—regional homelessness service coordination and planning bodies. Each CoC collects data about the people it serves through its programs, such as homelessness prevention services, street outreach services, permanent housing interventions and a range of other strategies aligned with California’s Housing First objectives.

The dataset uploaded reflects the 2024 HUD Data Standard Changes. Previously, Race and Ethnicity are separate files but are now combined.

Information updated as of 2/06/2025.

This dataset provides information on individuals experiencing sheltered homelessness in the Austin/Travis County Continuum of Care (CoC) in a given fiscal year. "Sheltered" homelessness refers to individuals residing in emergency shelter, safe haven, or transitional housing project types. This measure overlaps, but is different from, the Point in Time (PIT) Count (SD23 Measure EOA.E.1a), which is a snapshot of both sheltered and unsheltered homelessness on one night in January.

Data Source: The data for this measure was reported to the City of Austin by the Ending Community Homelessness Coalition (ECHO). Each year, ECHO, as the homeless Continuum of Care Lead Agency (CoC Lead), aggregates and reports community wide data (including this measure) to the Department of Housing and Urban Development (HUD). This data is referred to as System Performance Measures as they are designed to examine how well a community is responding to homelessness at a system level.

View more details and insights related to this data set on the story page: https://data.austintexas.gov/stories/s/2ejn-hrh2

Homelessness Report April 2025. Published by Department of Housing, Local Government, and Heritage. Available under the license Creative Commons Attribution Share-Alike 4.0 (CC-BY-SA-4.0).Homelessness data Official homelessness data is produced by local authorities through the Pathway Accommodation and Support System (PASS). PASS was rolled-out nationally during the course of 2013. The Department’s official homelessness statistics are published on a monthly basis and refer to the number of homeless persons accommodated in emergency accommodation funded and overseen by housing authorities during a specific count week, typically the last full week of the month. The reports are produced through the Pathway Accommodation & Support System (PASS), collated on a regional basis and compiled and published by the Department. Homelessness reporting commenced in this format in 2014. The format of the data may change or vary over time due to administrative and/or technology changes and improvements. The administration of homeless services is organised across nine administrative regions, with one local authority in each of the regions, “the lead authority”, having overall responsibility for the disbursement of Exchequer funding. In each region a Joint Homelessness Consultative Forum exists which includes representation from the relevant State and non-governmental organisations involved in the delivery of homeless services in a particular region. Delegated arrangements are governed by an annually agreed protocol between the Department and the lead authority in each region. These protocols set out the arrangements, responsibilities and financial/performance data reporting requirements for the delegation of funding from the Department. Under Sections 38 and 39 of the Housing (Miscellaneous Provisions) Act 2009 a statutory Management Group exists for each regional forum. This is comprised of representatives from the relevant housing authorities and the Health Service Executive, and it is the responsibility of the Management Group to consider issues around the need for homeless services and to plan for the implementation, funding and co-ordination of such services. In relation to the terms used in the report for the accommodation types see explanation below: PEA - Private Emergency Accommodation: this may include hotels, B&Bs and other residential facilities that are used on an emergency basis. Supports are provided to services users on a visiting supports basis. STA - Supported Temporary Accommodation: accommodation, including family hubs, hostels, with onsite professional support. TEA - Temporary Emergency Accommodation: emergency accommodation with no (or minimal) support....

A. SUMMARY This archived dataset includes data for population characteristics that are no longer being reported publicly. The date on which each population characteristic type was archived can be found in the field “data_loaded_at”.

B. HOW THE DATASET IS CREATED Data on the population characteristics of COVID-19 cases are from:  * Case interviews  * Laboratories  * Medical providers    These multiple streams of data are merged, deduplicated, and undergo data verification processes.  

Race/ethnicity * We include all race/ethnicity categories that are collected for COVID-19 cases. * The population estimates for the "Other" or “Multi-racial” groups should be considered with caution. The Census definition is likely not exactly aligned with how the City collects this data. For that reason, we do not recommend calculating population rates for these groups.

Gender * The City collects information on gender identity using these guidelines.

Skilled Nursing Facility (SNF) occupancy * A Skilled Nursing Facility (SNF) is a type of long-term care facility that provides care to individuals, generally in their 60s and older, who need functional assistance in their daily lives.  * This dataset includes data for COVID-19 cases reported in Skilled Nursing Facilities (SNFs) through 12/31/2022, archived on 1/5/2023. These data were identified where “Characteristic_Type” = ‘Skilled Nursing Facility Occupancy’.

Sexual orientation * The City began asking adults 18 years old or older for their sexual orientation identification during case interviews as of April 28, 2020. Sexual orientation data prior to this date is unavailable. * The City doesn’t collect or report information about sexual orientation for persons under 12 years of age. * Case investigation interviews transitioned to the California Department of Public Health, Virtual Assistant information gathering beginning December 2021. The Virtual Assistant is only sent to adults who are 18+ years old. https://www.sfdph.org/dph/files/PoliciesProcedures/COM9_SexualOrientationGuidelines.pdf">Learn more about our data collection guidelines pertaining to sexual orientation.

Comorbidities * Underlying conditions are reported when a person has one or more underlying health conditions at the time of diagnosis or death.

Homelessness Persons are identified as homeless based on several data sources: * self-reported living situation * the location at the time of testing * Department of Public Health homelessness and health databases * Residents in Single-Room Occupancy hotels are not included in these figures. These methods serve as an estimate of persons experiencing homelessness. They may not meet other homelessness definitions.

Single Room Occupancy (SRO) tenancy * SRO buildings are defined by the San Francisco Housing Code as having six or more "residential guest rooms" which may be attached to shared bathrooms, kitchens, and living spaces. * The details of a person's living arrangements are verified during case interviews.

Transmission Type * Information on transmission of COVID-19 is based on case interviews with individuals who have a confirmed positive test. Individuals are asked if they have been in close contact with a known COVID-19 case. If they answer yes, transmission category is recorded as contact with a known case. If they report no contact with a known case, transmission category is recorded as community transmission. If the case is not interviewed or was not asked the question, they are counted as unknown.

C. UPDATE PROCESS This dataset has been archived and will no longer update as of 9/11/2023.

D. HOW TO USE THIS DATASET Population estimates are only available for age groups and race/ethnicity categories. San Francisco population estimates for race/ethnicity and age groups can be found in a view based on the San Francisco Population and Demographic Census dataset. These population estimates are from the 2016-2020 5-year American Community Survey (ACS).

This dataset includes many different types of characteristics. Filter the “Characteristic Type” column to explore a topic area. Then, the “Characteristic Group” column shows each group or category within that topic area and the number of cases on each date.

New cases are the count of cases within that characteristic group where the positive tests were collected on that specific specimen collection date. Cumulative cases are the running total of all San Francisco cases in that characteristic group up to the specimen collection date listed.

This data may not be immediately available for recently reported cases. Data updates as more information becomes available.

To explore data on the total number of cases, use the ARCHIVED: COVID-19 Cases Over Time dataset.

E. CHANGE LOG

• 9/11/2023 - data on COVID-19 cases by population characteristics over time are no longer being updated. The date on which each population characteristic type was archived can be found in the field “data_loaded_at”.
• 6/6/2023 - data on cases by transmission type have been removed. See section ARCHIVED DATA for more detail.
• 5/16/2023 - data on cases by sexual orientation, comorbidities, homelessness, and single room occupancy have been removed. See section ARCHIVED DATA for more detail.
• 4/6/2023 - the State implemented system updates to improve the integrity of historical data.
• 2/21/2023 - system updates to improve reliability and accuracy of cases data were implemented.
• 1/31/2023 - updated “population_estimate” column to reflect the 2020 Census Bureau American Community Survey (ACS) San Francisco Population estimates.
• 1/5/2023 - data on SNF cases removed. See section ARCHIVED DATA for more detail.
• 3/23/2022 - ‘Native American’ changed to ‘American Indian or Alaska Native’ to align with the census.
• 1/22/2022 - system updates to improve timeliness and accuracy of cases and deaths data were implemented.
• 7/15/2022 - reinfections added to cases dataset. See section SUMMARY for more information on how reinfections are identified.

This dataset contains two tables on the percent of household overcrowding (> 1.0 persons per room) and severe overcrowding (> 1.5 persons per room) for California, its regions, counties, and cities/towns. Data is from the U.S. Department of Housing and Urban Development (HUD), Comprehensive Housing Affordability Strategy (CHAS) and U.S. Census American Community Survey (ACS). The table is part of a series of indicators in the Healthy Communities Data and Indicators Project (HCI) of the Office of Health Equity: Healthy Communities Data and Indicators Project of the Office of Health Equity. Residential crowding has been linked to an increased risk of infection from communicable diseases, a higher prevalence of respiratory ailments, and greater vulnerability to homelessness among the poor. Residential crowding reflects demographic and socioeconomic conditions. Older-adult immigrant and recent immigrant communities, families with low income and renter-occupied households are more likely to experience household crowding. A form of residential overcrowding known as "doubling up"—co-residence with family members or friends for economic reasons—is the most commonly reported prior living situation for families and individuals before the onset of homelessness. More information about the data table and a data dictionary can be found in the About/Attachments section.The household crowding table is part of a series of indicators in the Healthy Communities Data and Indicators Project (HCI) of the Office of Health Equity. The goal of HCI is to enhance public health by providing data, a standardized set of statistical measures, and tools that a broad array of sectors can use for planning healthy communities and evaluating the impact of plans, projects, policy, and environmental changes on community health. The creation of healthy social, economic, and physical environments that promote healthy behaviors and healthy outcomes requires coordination and collaboration across multiple sectors, including transportation, housing, education, agriculture and others. Statistical metrics, or indicators, are needed to help local, regional, and state public health and partner agencies assess community environments and plan for healthy communities that optimize public health. More information on HCI can be found here: https://www.cdph.ca.gov/Programs/OHE/CDPH%20Document%20Library/Accessible%202%20CDPH_Healthy_Community_Indicators1pager5-16-12.pdf
The format of the household overcrowding tables is based on the standardized data format for all HCI indicators. As a result, this data table contains certain variables used in the HCI project (e.g., indicator ID, and indicator definition). Some of these variables may contain the same value for all observations.

The Local Employment Dynamics (LED) Partnership is a voluntary federal-state enterprise created for the purpose of merging employee, and employer data to provide a set of enhanced labor market statistics known collectively as Quarterly Workforce Indicators (QWI). The QWI are a set of economic indicators including employment, job creation, earnings, and other measures of employment flows. For the purposes of this dataset, LED data for 2018 is aggregated to Census Summary Level 070 (State + County + County Subdivision + Place/Remainder), and joined with the Emergency Solutions Grantee (ESG) areas spatial dataset for FY2018. The Emergency Solutions Grants (ESG), formally the Emergency Shelter Grants, program is designed to identify sheltered and unsheltered homeless persons, as well as those at risk of homelessness, and provide the services necessary to help those persons quickly regain stability in permanent housing after experiencing a housing crisis and/or homelessness. The ESG is a non-competitive formula grant awarded to recipients which are state governments, large cities, urban counties, and U.S. territories. Recipients make these funds available to eligible sub-recipients, which can be either local government agencies or private nonprofit organizations. The recipient agencies and organizations, which actually run the homeless assistance projects, apply for ESG funds to the governmental grantee, and not directly to HUD. Please note that this version of the data does not include Community Planning and Development (CPD) entitlement grantees. LED data for CPD entitlement areas can be obtained from the LED for CDBG Grantee Areas feature service. To learn more about the Local Employment Dynamics (LED) Partnership visit: https://lehd.ces.census.gov/, for questions about the spatial attribution of this dataset, please reach out to us at GISHelpdesk@hud.gov. Data Dictionary: DD_LED for ESG Grantee Areas

Date of Coverage: ESG-2021/LED-2018

Content

Included in this data set are data elements that will help the public identify agencies that are certified to operate programs for runaway and homeless youth. These programs are available to assist runaway and homeless youth in emergency situation and provide independent living skills for youth in transition. Data elements include the agency name, agency business address, phone number, website and type of program offered.

Context

This is a dataset hosted by the State of New York. The state has an open data platform found here and they update their information according the amount of data that is brought in. Explore New York State using Kaggle and all of the data sources available through the State of New York organization page!

• Update Frequency: This dataset is updated annually.

Acknowledgements

This dataset is maintained using Socrata's API and Kaggle's API. Socrata has assisted countless organizations with hosting their open data and has been an integral part of the process of bringing more data to the public.

Cover photo by Zac Ong on Unsplash
Unsplash Images are distributed under a unique Unsplash License.

Analysis of ‘COVID-19 Deaths by Population Characteristics Over Time’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://catalog.data.gov/dataset/60f5842f-a359-4b03-ad21-1bcfc3bf7fe6 on 13 February 2022.

--- Dataset description provided by original source is as follows ---

Note: On January 22, 2022, system updates to improve the timeliness and accuracy of San Francisco COVID-19 cases and deaths data were implemented. You might see some fluctuations in historic data as a result of this change.

A. SUMMARY This dataset shows San Francisco COVID-19 deaths by population characteristics and by date. Deaths are included on the date the individual died.

Population characteristics are subgroups, or demographic cross-sections, like age, race, or gender. The City tracks how deaths have been distributed among different subgroups. This information can reveal trends and disparities among groups.

Data is lagged by five days, meaning the most date included is 5 days prior to today. All data update daily as more information becomes available.

B. HOW THE DATASET IS CREATED COVID-19 deaths are suspected to be associated with COVID-19. This means COVID-19 is listed as a cause of death or significant condition on the death certificate.

Data on the population characteristics of COVID-19 deaths are from: * Case interviews * Laboratories * Medical providers

These multiple streams of data are merged, deduplicated, and undergo data verification processes. It takes time to process this data. Because of this, data is lagged by 5 days and death totals for previous days may increase or decrease. More recent data is less reliable.

Data are continually updated to maximize completeness of information and reporting on San Francisco COVID-19 deaths.

Data notes on each population characteristic type is listed below.

Race/ethnicity * We include all race/ethnicity categories that are collected for COVID-19 cases.

Sexual orientation * Sexual orientation data is collected from individuals who are 18 years old or older. These individuals can choose whether to provide this information during case interviews. Learn more about our data collection guidelines. * The City began asking for this information on April 28, 2020. Gender * The City collects information on gender identity using these guidelines.

Comorbidities * Underlying conditions are reported when a person has one or more underlying health conditions at the time of diagnosis or death.

Transmission type * Information on transmission of COVID-19 is based on case interviews with individuals who have a confirmed positive test. Individuals are asked if they have been in close contact with a known COVID-19 case. If they answer yes, transmission category is recorded as contact with a known case. If they report no contact with a known case, transmission category is recorded as community transmission. If the case is not interviewed or was not asked the question, they are counted as unknown.

Homelessness Persons are identified as homeless based on several data sources: * self-reported living situation
* the location at the time of testing * Department of Public Health homelessness and health databases * Residents in Single-Room Occupancy hotels are not included in these figures.
These methods serve as an estimate of persons experiencing homelessness. They may not meet other homelessness definitions.

Skilled Nursing Facility (SNF) occupancy * A Skilled Nursing Facility (SNF) is a type of long-term care facility that provides care to individuals, generally in their 60s and older, who need functional assistance in their daily lives. * Facilities are mandated to report COVID-19 cases or deaths among their residents. The City follows up with these facilities to confirm.
* There may be differences between the City’s SNF data and the California Department of Public Health (CDPH) dashboard. The difference may be because the City and the State use dif

--- Original source retains full ownership of the source dataset ---

The most recent rate of homelessness is calculated using ACS population estimates from the previous year, unless otherwise noted.

Data Source: HUD's Annual Homeless Assessment Report (AHAR) Point-in-Time (PIT) Estimates by State and American Community Survey (ACS) 1-Year Estimates

Why this MattersSafe, adequate, and stable housing is a human right and essential for the health and well-being of individuals, families, and communities.People who experience homelessness also struggle to maintain access to healthcare, employment, education, healthy relationships, and other basic necessities in life, according to the DC Interagency Council on Homelessness Strategic Plan.BIPOC populations are disproportionately affected by homelessness due to housing discrimination, mass incarceration, and other policies that have limited socioeconomic opportunities for Black, Latino, and other people of color.

The District's Response Strategic investments in proven strategies for driving down homelessness, including the Career Mobility Action Plan (Career MAP) program, operation of non-congregate housing, and expansion of the District’s shelter capacity.Homelessness prevention programs for at-risk individuals and families, such as emergency rental assistance, targeted affordable housing, and permanent supporting housing.Programs and services to enhance resident’s economic and employment security and ensure access to affordable housing.

Homeless and battered women's shelters compiled from Reference USA. Reference USA is an internet-based reference service from the Government Division of InfoGroup. This site was designed as a reference to government agencies. ReferenceUSAGov database contains more than 57 million US businesses, 320 million residents, and 855,000 healthcare providers. InfoGroup compiles information from public sources, including yellow pages and business white pages telephone directories, annual reports, federal government data, leading business magazines trade newsletters, major newspapers, industry and specialty directories, and postal service information. Over 350 database specialists make phone calls to verify information on business and healthcare providers in the database, placing in excess of 24 million phone calls annually.

This dataset contains estimates of the prevalence of homelessness on Census night 2016, derived from the Census of Population and Housing using the Australian Bureau of Statistics (ABS) definition of homelessness. Prevalence is an estimate of how many people experienced homelessness at a particular point-in-time. Data is by LGA 2016 boundaries. Periodicity: 5 yearly. For more information visit the Australian Bureau of Statistics.

Dataset for the maps accompanying the Housing in Aotearoa New Zealand: 2025 report. This dataset contains data for severe housing deprivation from the 2018 and 2023 Censuses.

​

Data is available by health district.

​

Severe housing deprivation has data for the census usually resident population from the 2018 and 2023 Censuses, including:

• estimated prevalence rate of severe housing deprivation (per 10,000 people)
• estimated rate for those; without shelter, in temporary accommodation, sharing someone else’s private dwelling, in uninhabitable housing, for whom it could not be determined whether they were severely housing deprived or not.

​

Map shows the estimated prevalence rate of severe housing deprivation (per 10,000 people) for the census usually resident population for the 2023 Census.

​

Download lookup file from Stats NZ ArcGIS Online or embedded attachment in Stats NZ geographic data service. Download data table (excluding the geometry column for CSV files) using the instructions in the Koordinates help guide.

​

Footnotes

Geographical boundaries

Statistical standard for geographic areas 2023 (updated December 2023) has information about geographic boundaries as of 1 January 2023. Address data from 2013 and 2018 Censuses was updated to be consistent with the 2023 areas. Due to the changes in area boundaries and coding methodologies, 2013 and 2018 counts published in 2023 may be slightly different to those published in 2013 or 2018.

​

Subnational census usually resident population

The census usually resident population count of an area (subnational count) is a count of all people who usually live in that area and were present in New Zealand on census night. It excludes visitors from overseas, visitors from elsewhere in New Zealand, and residents temporarily overseas on census night. For example, a person who usually lives in Christchurch city and is visiting Wellington city on census night will be included in the census usually resident population count of Christchurch city. 

​

Population counts

Stats NZ publishes a number of different population counts, each using a different definition and methodology. Population statistics – user guide has more information about different counts. 

​

Caution using time series

Time series data should be interpreted with care due to changes in census methodology and differences in response rates between censuses. The 2023 and 2018 Censuses used a combined census methodology (using census responses and administrative data), while the 2013 Census used a full-field enumeration methodology (with no use of administrative data).

​

Severe housing deprivation time series

The 2018 estimates of severe housing deprivation have been updated using the 2023 methodology for estimating severe housing deprivation. Severe housing deprivation (homelessness) estimates – updated methodology: 2023 Census has more information.

​

Severe housing deprivation

Figures in this map and geospatial file exclude Women’s refuge data, as well as estimates for children living in non-private dwellings. Severe housing deprivation (homelessness) estimates – updated methodology: 2023 Census has more information.

​

About the 2023 Census dataset

For information on the 2023 Census dataset see Using a combined census model for the 2023 Census. We combined data from the census forms with administrative data to create the 2023 Census dataset, which meets Stats NZ's quality criteria for population structure information. We added real data about real people to the dataset where we were confident the people who hadn’t completed a census form (which is known as admin enumeration) will be counted. We also used data from the 2018 and 2013 Censuses, administrative data sources, and statistical imputation methods to fill in some missing characteristics of people and dwellings.

​

Data quality

The quality of data in the 2023 Census is assessed using the quality rating scale and the quality assurance framework to determine whether data is fit for purpose and suitable for release. Data quality assurance in the 2023 Census has more information.

​

Quality rating of a variable

The quality rating of a variable provides an overall evaluation of data quality for that variable, usually at the highest levels of classification. The quality ratings shown are for the 2023 Census unless stated. There is variability in the quality of data at smaller geographies. Data quality may also vary between censuses, for subpopulations, or when cross tabulated with other variables or at lower levels of the classification. Data quality ratings for 2023 Census variables has more information on quality ratings by variable.

​

Census usually resident population count concept quality rating

The census usually resident population count is rated as very high quality.

Census usually resident population count – 2023 Census: Information by concept has more information, for example, definitions and data quality.

​

Quality of severe housing deprivation data

Severe housing deprivation (homelessness) estimates – updated methodology: 2023 Census has more information on the data quality of this variable.

​

Using data for good

Stats NZ expects that, when working with census data, it is done so with a positive purpose, as outlined in the Māori Data Governance Model (Data Iwi Leaders Group, 2023). This model states that "data should support transformative outcomes and should uplift and strengthen our relationships with each other and with our environments. The avoidance of harm is the minimum expectation for data use. Māori data should also contribute to iwi and hapū tino rangatiratanga”.

​

Confidentiality

The 2023 Census confidentiality rules have been applied to 2013, 2018, and 2023 data. These rules protect the confidentiality of individuals, families, households, dwellings, and undertakings in 2023 Census data. Counts are calculated using fixed random rounding to base 3 (FRR3) and suppression of ‘sensitive’ counts less than six, where tables report multiple geographic variables and/or small populations. Individual figures may not always sum to stated totals. Applying confidentiality rules to 2023 Census data and summary of changes since 2018 and 2013 Censuses has more information about 2023 Census confidentiality rules.

​

Inconsistencies in definitions

Please note that there may be differences in definitions between census classifications and those used for other data collections.

The peer-reviewed publication for this dataset has been published in Data & Policy, and can be accessed here: https://arxiv.org/abs/2406.16527 Please cite this when using the dataset.

This dataset has been produced as a result of the “Systematic Review of Outcomes Contracts using Machine Learning” (SyROCCo) project. The goal of the project was to apply machine learning techniques to a systematic review process of outcomes-based contracting (OBC). The purpose of the systematic review was to gather and curate, for the first time, all of the existing evidence on OBC. We aimed to map the current state of the evidence, synthesise key findings from across the published studies, and provide accessible insights to our policymaker and practitioner audiences.

OBC is a model for the provision of public services wherein a service provider receives payment, in-part or in-full, only upon the achievement of pre-agreed outcomes.

The data used to conduct the review consists of 1,952 individual studies of OBC. They include peer reviewed journal articles, book chapters, doctoral dissertations, and assorted ‘grey literature’ - that is, reports and evaluations produced outside of traditional academic publications. Those studies were manually filtered by experts on the topic from an initial search of over 11,000 results.

The full text of the articles was obtained from their PDF versions and preprocessed. This involved text format normalisation, removing acknowledgements and bibliographic references.

The corpus was then connected to the INDIGO Impact Bond Dataset. Projects and organisations mentioned in this latter dataset were searched for in the article’s corpus to relate both datasets.

Other types of information that were identified in the texts were 1) financial mechanisms (type of outcomes-based instrument); using a list of terms related to those financial mechanisms based on prior discussions with a policy advisory group (Picker et al., 2021); 2) references to the 17 Sustainable Development Goals (SDGs) defined by the United Nations General Assembly in the 2030 Agenda; 3) country names mentioned in each article and income levels related to the countries; according to the World Classification of Income Levels 2022 by the World Bank.

Three machine learning techniques were applied to the corpus:

• Policy areas identification. A query-driven topic model (QDTM) (Fang et al., 2021) was used to determine the probability of an article belonging to different policy areas (health, education, homelessness, criminal justice, employment and training, child and family welfare, and agriculture and environment), using all text of the article as input. The QDTM is a semi-supervised machine learning algorithm that allows users to specify their prior knowledge in the form of simple queries in words or phrases and return query-related topics.

• Named Entity Recognition. Three named entity recognition models were applied: “en_core_web_lg” and “en_core_web_trf” models from the python package ‘spaCy’ and the “ner-ontonotes-large” English model from ‘Flair’. “en_core_web_trf” is based on the RoBERTa-base transformer model. ‘Flair’ is a bi-LSTM character-based model. All models were trained on the “OntoNotes 5” data source (Marcus et al., 2011) and are able to identify geographical locations, organisation names, and laws and regulations. An ensemble method was adopted, considering the entities that appear simultaneously in the results of any two models as the correct entities.

• Semantic text similarity. We calculated the similarity score between articles. The 10,000 most frequently mentioned words were first extracted from all the articles’ titles and abstracts and the text vectorization technique TF*IDF was applied to convert each article’s abstract into an importance score vector based on these words. Using these numerical vectors, the cosine similarity between different articles was calculated.

The SyROCCo Dataset includes references to the 1952 studies of OBCs mentioned above and the results of the previous processing steps and techniques. Each entry of the dataset contains the following information.

The basic information of each document is its title, abstract, authors, published years, DOI and Article ID:

• Title: Title of the document.

• Abstract: Text of the abstract.

• Authors: Authors of a study.

• Published Years: Published Years of a study.

• DOI: DOI link of a study.

• Article ID: ID of the document selected during the screening process.

The probability of a study belonging to each policy area:

• policy_sector_health: The probability of a study belongs to the policy sector “health”.

• policy_sector_education: The probability of a study belongs to the policy sector “education”.

• policy_sector_homelessness: The probability of a study belongs to the policy sector “homelessness”.

• policy_sector_criminal: The probability of a study belongs to the policy sector “criminal”

• policy_sector_employment: The probability of a study belongs to the policy sector “employment”

• policy_sector_child: The probability of a study belongs to the policy sector “child”.

• policy_sector_environment: The probability of a study belongs to the policy sector “environment”.

Other types of information such as financial mechanisms, Sustainable Development Goals, and different types of named entities:

• financial_mechanisms: Financial mechanisms mentioned in a study.

• top_financial_mechanisms: The financial mechanisms mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

• top_sgds: Sustainable Development Goals mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

• top_countries: Country names mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions. This entry is also used to determine the income level of the mentioned counties.

• top_Project: Indigo projects mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

• top_GPE: Geographical locations mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

• top_LAW: Relevant laws and regulations mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

• top_ORG: Organisations mentioned in a study are listed in descending order according to the number of times they are mentioned, and include the corresponding context of the mentions.

Homelessness Report May 2025. Published by Department of Housing, Local Government and Heritage. Available under the license Creative Commons Attribution Share-Alike 4.0 (CC-BY-SA-4.0).Homelessness data Official homelessness data is produced by local authorities through the Pathway Accommodation and Support System (PASS). PASS was rolled-out nationally during the course of 2013. The Department’s official homelessness statistics are published on a monthly basis and refer to the number of homeless persons accommodated in emergency accommodation funded and overseen by housing authorities during a specific count week, typically the last full week of the month. The reports are produced through the Pathway Accommodation & Support System (PASS), collated on a regional basis and compiled and published by the Department. Homelessness reporting commenced in this format in 2014. The format of the data may change or vary over time due to administrative and/or technology changes and improvements. The administration of homeless services is organised across nine administrative regions, with one local authority in each of the regions, “the lead authority”, having overall responsibility for the disbursement of Exchequer funding. In each region a Joint Homelessness Consultative Forum exists which includes representation from the relevant State and non-governmental organisations involved in the delivery of homeless services in a particular region. Delegated arrangements are governed by an annually agreed protocol between the Department and the lead authority in each region. These protocols set out the arrangements, responsibilities and financial/performance data reporting requirements for the delegation of funding from the Department. Under Sections 38 and 39 of the Housing (Miscellaneous Provisions) Act 2009 a statutory Management Group exists for each regional forum. This is comprised of representatives from the relevant housing authorities and the Health Service Executive, and it is the responsibility of the Management Group to consider issues around the need for homeless services and to plan for the implementation, funding and co-ordination of such services. In relation to the terms used in the report for the accommodation types see explanation below: PEA - Private Emergency Accommodation: this may include hotels, B&Bs and other residential facilities that are used on an emergency basis. Supports are provided to services users on a visiting supports basis. STA - Supported Temporary Accommodation: accommodation, including family hubs, hostels, with onsite professional support. TEA - Temporary Emergency Accommodation: emergency accommodation with no (or minimal) support....

The Local Employment Dynamics (LED) Partnership is a voluntary federal-state enterprise created for the purpose of merging employee, and employer data to provide a set of enhanced labor market statistics known collectively as Quarterly Workforce Indicators (QWI). The QWI are a set of economic indicators including employment, job creation, earnings, and other measures of employment flows. For the purposes of this dataset, LED data for 2018 is aggregated to Census Summary Level 070 (State + County + County Subdivision + Place/Remainder), and joined with the Continuum of Care Program grantee areas spatial dataset for FY2017. The Continuum of Care (CoC) Homeless Assistance Programs administered by HUD award funds competitively and require the development of a Continuum of Care system in the community where assistance is being sought. A continuum of care system is designed to address the critical problem of homelessness through a coordinated community-based process of identifying needs and building a system to address those needs. The approach is predicated on the understanding that homelessness is not caused merely by a lack of shelter, but involves a variety of underlying, unmet needs - physical, economic, and social. Funds are granted based on the competition following the Notice of Funding Availability (NOFA). Please note that this version of the data does not include Community Planning and Development (CPD) entitlement grantees. LED data for CPD entitlement areas can be obtained from the LED for CDBG Grantee Areas feature service. To learn more about the Local Employment Dynamics (LED) Partnership visit: https://lehd.ces.census.gov/, for questions about the spatial attribution of this dataset, please reach out to us at GISHelpdesk@hud.gov. Data Dictionary: DD_LED for CoC Grantee Areas

Date of Coverage: CoC-2021/LED-2018

BackgroundOpioid use disorder (OUD) is a growing public health crisis, with opioids involved in an overwhelming majority of drug overdose deaths in the United States in recent years. While medications for opioid use disorder (MOUD) effectively reduce overdose mortality, only a minority of patients are able to access MOUD; additionally, those with unstable housing receive MOUD at even lower rates.ObjectiveBecause MOUD access is a multifactorial issue, we leverage machine learning techniques to assess and rank the variables most important in predicting whether any individual receives MOUD. We also seek to explain why persons experiencing homelessness have lower MOUD access and identify potential targets for action.MethodsWe utilize a gradient boosted decision tree algorithm (specifically, XGBoost) to train our model on SAMHSA’s Treatment Episode Data Set-Admissions, using anonymized demographic and clinical information for over half a million opioid admissions to treatment facilities across the United States. We use Shapley values to quantify and interpret the predictive power and influencing direction of individual features (i.e., variables).ResultsOur model is effective in predicting access to MOUD with an accuracy of 85.97% and area under the ROC curve of 0.9411. Notably, roughly half of the model’s predictive power emerges from facility type (23.34%) and geographic location (18.71%); other influential factors include referral source (6.74%), history of prior treatment (4.41%), and frequency of opioid use (3.44%). We also find that unhoused patients go to facilities that overall have lower MOUD treatment rates; furthermore, relative to housed (i.e., independent living) patients at these facilities, unhoused patients receive MOUD at even lower rates. However, we hypothesize that if unhoused patients instead went to the facilities that housed patients enter at an equal percent (but still received MOUD at the lower unhoused rates), 89.50% of the disparity in MOUD access would be eliminated.ConclusionThis study demonstrates the utility of a model that predicts MOUD access and both ranks the influencing variables and compares their individual positive or negative contribution to access. Furthermore, we examine the lack of MOUD treatment among persons with unstable housing and consider approaches for improving access.