New York City Housing and Vacancy Survey (NYCHVS), 1978-2023 (ICPSR 39267)

Version Date: Mar 17, 2025

Principal Investigator(s):
Elyzabeth Gaumer, New York (N.Y.). Department of Housing Preservation and Development; Moon Wha Lee, New York (N.Y.). Department of Housing Preservation and Development; Michael A. Stegman, New York (N.Y.). Department of Housing Preservation and Development; Peter Marcuse, New York (N.Y.). Department of Housing Preservation and Development; Abraham Engel, New York (N.Y.). Department of Housing Preservation and Development

https://doi.org/10.3886/ICPSR39267.v1

Version V1

These data are unavailable indefinitely from ICPSR.

Slide tabs to view more

Summary

The New York City Housing and Vacancy Survey (NYCHVS) is a citywide survey designed to be representative of the NYC housing stock and community-dwelling population that is conducted about every three years by the City of New York. The NYCHVS has been conducted since 1965 and is the longest running housing survey in the country. The survey is mandated by New York State and New York City laws to measure the net rental vacancy rate and describe the supply, condition, and continued need for rent control and rent stabilization which covers half of the city's rental housing, or about 1 million apartments and 2.5 million tenants.

Detailed data from the survey cover many characteristics of the New York City housing market, including characteristics of the City's population, household, housing stock, and neighborhoods. Other data include, but is not limited to, rent regulatory and home ownership status; structural conditions; unit maintenance and neighborhood conditions; crowding, rents, utility costs, type of heating fuel, rent/income ratios; owner purchase price and estimated value, mortgage status and interest rate; number of stories and units in building, cooperative/condominium status, and wheelchair accessibility.

The data in this collection are organized by occupied units, vacant units, non-interviews, individual person records, and all housing units. Person records for each individual in occupied housing units regardless of age were added beginning in 1991. Non-interview units data are available for 1978 to 2008. Starting in 2021, datasets for all housing units were added.

Citation

Gaumer, Elyzabeth, Lee, Moon Wha, Stegman, Michael A., Marcuse, Peter, and Engel, Abraham. New York City Housing and Vacancy Survey (NYCHVS), 1978-2023. Inter-university Consortium for Political and Social Research [distributor], 2025-03-17. https://doi.org/10.3886/ICPSR39267.v1

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote

Funding

New York (N.Y.). Department of Housing Preservation and Development

Subject Terms

Geographic Coverage

Smallest Geographic Unit

Sub-borough

Distributor(s)

Inter-university Consortium for Political and Social Research
Hide

Time Period(s)

1978 -- 2023

Date of Collection

1978 -- 2023

Data Collection Notes

  1. For additional details on the New York City Housing and Vacancy Survey (NYCHVS), please visit the NYC Housing Preservation and Development (HPD) Research webpage, which contains links to questionnaires, user guides, and weighting information for the 2021 and 2023 surveys.

Hide

Study Purpose

The purpose of the study is to measure the net rental vacancy rate for New York City's rental stock, as required by law, and describe the supply, condition, and continued need for rent control and rent stabilization.

Study Design

Sampled addresses were assigned to field researchers, who determined whether the housing unit was occupied, vacant, or out of scope (e.g., unit does not exit, is not a housing unit). If interviews could be completed, in-person interviews were conducted by field researchers with knowledgeable adult residents of occupied units, or other knowledgeable informants who could provide information on vacant units. In 2021, the mode of data collection was changed from paper-based survey to computer-assisted personal interview (CAPI). Due to the COVID-19 pandemic, options for phone interviews or conducting interviews in alternative locations were added.

Sample

The main sample of the survey were selected every decade as a representative sample of housing units across the five New York City boroughs. Additional new units were selected in each subsequent survey year until the next census. Units were selected based on the following sources: the Census Bureau master address file from the most recent census (1980, 1990, 2000, 2010, 2020), a sample of addresses resulting from new construction certificates of occupancy issues for each borough, a list of previously nonresidential addresses converted to residential housing units, and housing units located in structures owned by the City because the owner failed to pay taxes on the property.

Sample sizes vary based on the decade of data collection and are listed here as approximate values:

  • 1978: 14,400 units
  • 1981: 16,300 units
  • 1984: 16,500 units
  • 1987: 17,000 units
  • 1991, 1993, 1996, 2002, 2005: 18,000 units
  • 2008: 21,000 units
  • 2011, 2014, 2017: 19,000 units
  • 2021: 12,000 units
  • 2023: 15,600 units

Time Method

Longitudinal: Panel

Universe

  • Housing units in New York City, including occupied and vacant units.
  • Residents of occupied units or, in case of vacant units, knowledgeable informants (e.g., building managers, superintendents, rental or real estate agents).

Unit(s) of Observation

Housing Unit, Individual

Data Type(s)

Mode of Data Collection

Description of Variables

Across all survey years, items cover the following topics: demographics, unit characteristics, housing quality, household roster, housing costs, public assistance, income, and employment.

In 1991, the questionnaire was redesigned, replacing the version used from 1975 to 1987. In addition to demographic and income item revisions, all household members were asked demographic items instead of the main respondent only. Questions on handicapped-accessibility of housing units were added in 1996. New items on household member immigration status and rental subsidies were added in 1999. New questions on health conditions and neighborhood opinion were added beginning in 2002. In 2011, new items were added on rental assistance, energy assistance, grab bars, and whether anyone in the household (aged 65 or older) experienced a fall in the past 3 months. Starting with 2021, many items were revised and new items on child care, debts, assets, mortgage information, and COVID-19 related issues were added.

The variable LINK (1978-2008) indicates whether or not cases match between survey years within the same decade. For a case to be linked, the respective housing unit had to be either occupied or vacant in one of the four surveys (but could be occupied, vacant, or non-interview in the other years).

The variables UID (1978-2017) and CONTROL (2021 onward) are unit-level identifiers. Person records can be linked to the appropriate household record using LNO (1991 onward), a line number identifier for each person in an occupied unit. Linked units from the same decade will have the same household sequence number.

Linking data across years. 1978, 1981, 1984, and 1987 data can be linked using LINK. 1991, 1993, 1996, and 1999 data can be linked using UID. 2002, 2005, and 2008 data can be linked using UID. 2011, 2014, and 2017 data cannot be linked.

Response Rates

Response rate varies depending on the year of data collection (available 1991 onward):

  • 1991: 95 percent
  • 1993: 98 percent
  • 1996: 98 percent
  • 1999: 95 percent
  • 2002: 98 percent
  • 2005: 96 percent
  • 2008: 98 percent
  • 2011: 98 percent
  • 2014: 92 percent
  • 2017: 81 percent
  • 2021: 73 percent
  • 2023: 71 percent

Hide

Original Release Date

2025-03-17

Version History

2025-03-17 ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

  • Checked for undocumented or out-of-range codes.

Hide

Weight

Weight variables vary depending on the survey year. In general, household- and unit-level weights are available in the Occupied, Vacant, and All Units datasets. Person-level weights are available in Occupied (1981-1996 only) and Person Record (1996 onward) datasets.

1981 to 1996: Person-level weight is PWA from 1981 to 1987, and APW from 1991 to 1996. Household-level weight is FW.

In 1999, a new weighting methodology was developed. Caution should be used in making comparisons between 1999 and earlier surveys for total housing units and borough level and total population by age, race, and sex. Within-borough housing comparisons were not affected.

1999 to 2023: Survey weight variable FW is used to generate weighted estimates at a unit or household level. Person-level survey weight PW and aggregate persons weight APW (in Occupied datasets) are used to generate estimates at a person or population level. Replicate weights (FW1 through FW80, PW1 through PW80) may be used to estimate variance.

Detailed documentation on sample design, weighting, and error/variance estimation is available for the 2021 and 2023 survey years via the New York City Housing Preservation and Development (NYC HPD) research webpage.

Hide