Data Dictionary

Description:

The Appalachian Regional Commission designates certain counties and tracts as “distressed areas” in order to identify and monitor the economic status of areas within Appalachia. The designations are based on an index that compares economic indicators of Appalachian counties and tracts to national averages. The index employs three-year average unemployment rates, per capita market incomes, and poverty rates to rank every county in the nation. Counties are designated as distressed, at-risk, transitional, competitive, or attainment, based on this rank, where distressed counties rank as mostly economically depressed (worst 10% economic performance in the nation) and attainment the most economically strong (best 10% economic performance in the nation). In addition to all census tracts located in distressed counties, census tracts in at-risk or transitional counties are considered “distressed areas” if they have a median family income 67% or less than the national average and a poverty rate 150% of the national average or greater.

Distressed areas are updated annually using the most recently available American Community Survey five-year estimates. Distressed areas for 2012 are presented at the 2000 census tract boundaries, and 2013-2022 areas are presented at the 2010 boundaries, 2023 areas are presented at the 2020 boundaries, and 2024-2025 areas are presented at the 2022 boundaries.

Description:

The Bureau of Labor Statistics’ Local Area Unemployment Statistics (LAUS) program produces monthly and annual employment, unemployment, and labor force data for Census regions and divisions, States, counties, metropolitan areas, and many cities, by place of residence. PolicyMap contains county and state counts of people employed, unemployed, and in the labor force, as well as the unemployment rate. The annual values presented in PolicyMap are annual averages for the years listed as provided by the BLS.

The concepts and definitions used by LAUS come from the Current Population Survey (CPS), the household survey that is the official measure of the labor force for the nation. According to this definition, employed persons include people who did any paid work as employees, worked in their own business or farm, or did unpaid work of 15 or more hours in an establishment owned by a relative. Unemployed persons include people who had no employment but were available for and seeking employment. People in the labor force are all those people classified as employed or unemployed. The labor force does not include military (active duty) and institutionalized persons.

Every April, the Bureau of Labor Statistics re-releases revised data for the previous five years. PolicyMap’s data reflects these revisions.

In 2015, the BLS changed the methodology of their model, with changes relating to structural differences, real-time benchmarking, smoothed seasonal adjustment, and treatment of outliers. At the state-level data, this methodology was used to re-estimate all data back to 1976 (PolicyMap displays data for 2000 and later). At sub-state geographies (city, county, CBSA, metropolitan division), the new methodology was only used to re-estimate data going back to January 2010. For this reason, comparisons between data before and after this change are not advised. Certain sub-state areas have the revised methodology re-estimated back to 1990: New York City (back to 1976), the Los Angeles-Long Beach-Glendale metropolitan division, the Chicago-Naperville-Arlington Heights, IL metropolitan division; the Cleveland-Elyria, OH metropolitan area; the Detroit-Warren-Dearborn, MI metropolitan area; the Miami-Miami Beach-Kendall, FL metropolitan division; and the Seattle-Bellevue-Everett, WA metropolitan division. Note that the previous methodology disaggregated data from the state level; because the state methodology is being re-estimated for before 2010, sub-state level data is indirectly affected.

Data for Connecticut counties is unavailable prior to 2020 due to boundary changes introduced by BLS. Data for Connecticut is available 2020 onward at the new county-equivalent, planning region boundaries.

The annual data contains percent change calculations on the number of people in the labor force and employed, as well as a change in percent calculation of the unemployment rate. These were calculated by PolicyMap.

Description:

The Bureau of Labor Statistics’ gathers data on employment and wages from state workforce agencies and compiles it under the Quarterly Census of Employment and Wages (QCEW) program. The data are derived from quarterly tax reports submitted by employers to State workforce agencies under State Unemployment Insurance (UI) laws; and from Federal agencies under the Unemployment Compensation for Federal Employees (UCFE) program. PolicyMap displays only private sector employment and wage data.

PolicyMap displays QCEW annual location quotient data for both private and public sector employment. BLS uses location quotients, or LQs, to see where occupations or industries are concentrated in the United States. LQs measure how concentrated jobs in a particular industry is in a specific area compared to the national average. If an LQ is equal to 1, then the industry has the same share of its area employment as it does in the nation. An LQ greater than 1 indicates an industry with a greater share of the local area employment than is the case nationwide. For example, in Philadelphia County, the LQ for employment for Educational Services in the private sector is roughly 4. Educational Services makes up 8% of jobs in Philadelphia, compared to about 2% nationally. The concentration of employment in Philadelphia County in Private Educational Services is about 4 times that of the concentration of employment for Educational Services in the U.S. This industry makes up a larger share of the Philadelphia County employment total than it does for the nation as a whole. For more information on QCEW LQ calculations, visit https://www.bls.gov/cew/about-data/location-quotients-explained.htm 

Description:

Names, locations, and financial data for a large dataset of non-profit entities. Data is sourced from the Internal Revenue Service’s data on Form 990 filings. Organizations file Form 990, or Returns for Organziations Exempt from income tax under particular sections of the Internal Revenue Code. This form is used to report required information such as total expenses, total assets, number of employees, and other information. Some locations (i.e, a given address or set of latitude and lonngitude coordinates) have more than one Employer Identification Number (EIN); this likely represents organizations that have multiple chapters, each with their own EIN, that use a single organization name and address.

Description:

Equitable Recovery Program: The Community Development Financial Institutions Fund awards Community Development Financial Institutions (CDFIs) grants for economic recovery from the COVID-19 pandemic. The CDFI Fund designates census tracts as eligible for these grants based on the following criteria: (a) are census tracts that (i) demonstrate “severe impact” of the COVID-19 pandemic, and (ii) have a median income at or below 120% of the Area Median Income, and (iii) are CDFI Investment Areas; or (b) are Native Areas.

Severe COVID Impact/Low Community Resilience designation is given to census tracts that demonstrate any of the following: severe mortality (being in the highest tercile of the number of deaths per 100,000 people, according to reported cumulative mortality for the period from April 1, 2020 to March 31, 2021), severe change in unemployment or low community resilience (based on data from the US Census Bureau Community Resilience Estimates Program). CDFI Investment Areas are geographies which have a population poverty rate of at least 20% (including Persistent Poverty Counties, which are counties that have had 20% of more of their populations living in poverty over the past 30 years),or have unemployment rates of at least 1.5 times the national or other criteria listed by CDFI Fund.

The Community Development Financial Institution (CDFI) Fund, a division of the US Department of the Treasury, administers the New Markets Tax Credit (NMTC) and Bank Enterprise Award (BEA) programs, and supports and invests in Community Development Financial Institutions. For information about the NMTC, please see entry, below. The CDFI Fund maintains a list of Census Tracts and their program eligibility or designation, based on income, poverty and unemployment data provided by the Census Bureau’s 2016-2020 American Community Survey (ACS) for 2020 census tracts. For more on these programs users should consult the CDFI Fund website directly: www.cdfifund.gov.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the US Department of the Treasury, administers the New Markets Tax Credit (NMTC). PolicyMap has performed calculations on various data sources in order to map eligibility and threshold requirements established by the CDFI Fund for Part II (Community Impact) of the NMTC Allocation Application. The NMTC Allocation Application data on PolicyMap is available as follows. Note that the latest eligibility criteria use Census American Community Survey (ACS) 2016-2020 estimates.

CDFI Fund New Markets Tax Credit NMTC Eligibility NMTC Eligible Census tracts include those that have either (1) Median Family Income at or below 80% of Area Median Income (AMI) in the period of 2016-2020 or (2) Poverty Rate of 20% or greater in the period of 2016-2020. PolicyMap provides a map of those eligible Census tracts (“Eligible Tracts”), as well as the underlying data used to create that map in the (“Eligibility Criteria”). PolicyMap also provides the underlying data without the NMTC thresholds (“Tract Family Income as % of AMI” and “Poverty”).

Severe Distress, Deep Distress, High Migration Rural County, and Non-Metropolitan Meeting the NMTC Severe Distress, Deep Distress, High Migration Rural County, or Non-Metropolitan criteria is based on whether or not a given Census tract meets basic NMTC Eligibility, plus one of the following factors:

Severe Distress: having a median family income at or below 60% of AMI in the period of 2016-2020; having a poverty rate at or above 30% in the period of 2016-2020; having an unemployment rate of at least 1.5 times the national unemployment rate in the period of 2016-2020;

Deep Distress: having a median family income at or below 40% of AMI in the period of 2016-2020; having a poverty rate at or above 40% in the period of 2016-2020; having an unemployment rate of at least 2.5 times the national unemployment rate in the period of 2016-2020;

High Migration Rural County: having median family income at or below 85 percent of the applicable AMI in the period of 2016-2020; being in any county during the 20-year period ending with the year in which the most recent census was conducted, with a net out-migration of inhabitants from the county of at least 10 percent of the population of the county at the beginning of such period; high migration rural counties qualify as low-income communities;

Non-Metropolitan: being in a county that is not part of a metropolitan statistical area.

*The median family income threshold for NMTC, more specifically, is: Census tracts with, if located within a non-Metropolitan Area, median family income at or below 60% of statewide median family income or, if located within a Metropolitan Area, median family income at or below 60% of the greater of the statewide median family income or the Metropolitan Area median family income.

Other Criteria for NMTC: Other criteria can include two of the following: meeting NMTC Heavy Distress requirements; being located within: an SBA Designated HUB Zone, a Medically Underserved Area (MUA), a Census tract within which a Brownfield is located, a HOPE VI Redevelopment Area, a Federal Native Area, an Appalachian Regional Commission or Delta Regional Authority Area, a Colonias Area, a State or Local Economic Zone (such as TIF or KOZ), a FEMA Disaster Area, or a ERS/USDA Food Desert. Please note that the data on PolicyMap do not take into account the following, due to unavailability of data: HOPE VI Redevelopment Areas, Federal Native Areas, Colonias Areas, State or Local Economic Zones, and FEMA Disaster Areas. Included in this submenu are the data for each of the available factors that constitute the Secondary Criteria for NMTC Severely Distressed.

• Median Family Income: Census ACS 2016-2020
• Area Median Income: Census ACS 2016-2020
• Poverty Rate: Census ACS 2016-2020
• Unemployment Rate: Census ACS 2016-2020
• SBA HUBZones: Small Business Administration HUBZones
• Medically Underserved Areas: US Department of Health and Human Services Health Resources and Services Administration Shortage Areas
• Delta Regional Authority Distressed Counties: Delta Regional Authority Distressed List (https://dra.gov/map-room/)
• Appalachian Regional Commission: County Economic Status and Distressed Areas in Appalachia (http://www.arc.gov/appalachian_region/CountyEconomicStatusandDistressedAreasinAppalachia.asp)
• Brownfield locations: EPA Brownfields
• ERS/USDA Food Deserts: ERS, USDA

Because any of these data sources may have been updated since the production of these calculations, users should verify eligibility directly with the CDFI Fund. Information in PolicyMap does not include HOPE VI Redevelopment Areas, Federal Native Areas, Colonias Areas, or State or Local Economic Zones. FEMA Disaster Areas are accessible under Federal Guidelines.

NMTC Eligibility and Qualified Opportunity Zones A joint dataset that includes both NMTC eligibility and designated Qualified Opportunity Zones. Census tracts labeled as “Designated OZ” are census tracts that have been nominated and designated as a Qualified Opportunity Zone (QOZ), according to the CDFI Fund and census tracts labeled as “NMTC Eligible” are census tracts that meet the CDFI Fund’s New Markets Tax Credit (NMTC) eligibility for CY 2018. For more information see the directory entry for Qualified Opportunity Zones.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the US Department of the Treasury, provides funding to CDFIs and qualified non-profit housing organizations through the Capital Magnet Fund. Areas of Economic Distress and Rural Areas are among the selection criteria used to determine eligibility. Additionally, Areas of High Housing Need and Metropolitan Areas were among the past selection criteria used to determine eligibility and are also available on PolicyMap. The CDFI Fund considers a tract to be an Area of Economic Distress in 2024 if it meets at least one of the following criteria: (i) At least 20% of Very-Low Income households spend more than half of their income on housing; or (ii) The tract is a Low-Income Housing Tax Credit qualified census tract; or (iv) Greater than 20 percent of households have incomes below the poverty rate with a rental vacancy rate of at least 10 percent; or (iv) Greater than 20 percent of households have incomes below the poverty rate with a homeownership vacancy rate of at least 10 percent; or (v) Tract is an underserved rural area as defined in the CMF Interim Rule. FHFA’s Duty to Serve regulation defines “rural area” as: (i) A census tract outside of a metropolitan statistical area, as designated by the Office of Management and Budget; or (ii) A census tract in a metropolitan statistical area, as designated by the OMB, that is: (A) Outside of the MSA’s Urbanized Areas as designated by the U.S. Department of Agriculture’s (USDA) Rural-Urban Commuting Area Code #1, and outside of tracts with a housing density of over 64 housing units per square mile for USDA’s RUCA Code #2. FHFA’s Duty to Serve regulation defines “high opportunity areas” as: (1) tracts designated by the Department of Housing and Urban Development (HUD) as a “Difficult Development Area” (DDA) during any year covered by an Enterprise’s Underserved Markets Plan or in the year before a Plan’s effective date, whose poverty rate falls below 10% (for metropolitan areas) or 15% (for non-metropolitan areas). Or (2) an area designated by a state or local Qualified Allocation Plan (QAP) as a high opportunity area whose poverty falls below 10% (for metropolitan areas) or 15% (for non-metropolitan areas). High opportunity areas are used to determine eligibility for extra credit under Duty to Serve. Low-Income Area means a census tract in which the median income does not exceed 80 percent of the median income for the area in which such census tract or block numbering area is located. For a census tract or block numbering area located within a Metropolitan Area, the median family income shall be at or below 80 percent of the Metropolitan Area median family income or the national Metropolitan Area median family income, whichever is greater. In the case of a census tract located outside of a Metropolitan Area, the median family income shall be at or below 80 percent of the statewide Non-Metropolitan Area median family income or the national Non-Metropolitan Area median family income, whichever is greater.

For more information about the Capital Magnet Fund data, see the CDFI Fund’s website here.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the US Department of the Treasury, administers the Capital Magnet Fund (CMF). CMF, established in 2008 and appropriated in 2010, is a competitive grant program to attract private capital for affordable housing development. CMF dollars are available to CDFIs and nonprofit housing developers who are active in affordable housing development. In FY 2012-2013, CMF awardees used the funds to finance 8,049 affordable rental units and 922 homeowner-occupied homes. CMF dollars may be used for the following purposes: loan loss reserves, revolving loan funds, affordable housing funds, or risk-sharing loans; economic development activities or community service facilities (day-care centers, workforce development centers, health care clinics) that support affordable housing as part of an overall community revitalization strategy. CMF grants must be matched at least 10:1 with other funding sources. One hundred percent of housing-eligible project costs must be used to finance units for households with income below 120% of area median income (AMI); and 51% of costs must be used for households with income below 80% of AMI. Rental housing projects have more specific requirements; please see the CDFI Fund website for more information.

CMF grantees track the use of funds though periodic financial and project reports submitted to the CDFI Fund.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the US Department of the Treasury, supports and invests in Community Development Financial Institutions through the CDFI Program and Native American CDFI Assistance Program. CDFIs are financial institutions that provide products and services in economically distressed target markets. The CDFI Fund certifies CDFIs through an application process on a rolling basis, depending on the type of institution. Not all CDFIs are certified, but certification is a requirement for some federal program funding. PolicyMap was able to locate 83% of CDFI addresses on a map. Data on certified CDFI locations are updated twice annually.

For CDFI transactions that span multiple census tracts or counties, medians are calculated using the total project cost while aggregations are calculated by dividing the total transaction cost by the number of census tracts or counties involved. Transaction or project counts at smaller geographies may not match larger geography counts given the double counting of split transactions and projects across census tracts and counties. Subcategories of transactions, such as CDFI Investments by Borrower Type, may not sum to the total amount of CDFI investments in a given area.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the U.S. Department of the Treasury, collects data from Community Development Entities (CDEs) based on information submitted through the New Markets Tax Credit (NMTC) program. NMTC awards are allocated to CDEs investing in operating businesses and real estate projects located in Low-Income Communities (LICs). This dataset is an aggregated collection of these projects, totaling the number, project type, and dollar value of investments reported from 2010 through 2019. Calculations were conducted by PolicyMap to create summary values based on geography and by project type. Total values were aggregated to state and zip code based on the address provided for the transaction. The CDFI Fund provided 2000 or 2010 census tracts associated with each transaction. Dollar values for transactions in multiple tracts were averaged across all tracts associated with the project. The total number of transactions for census tracts in an area may not be equivalent to totals by state and zip code. NMTC investments are also aggregated to CDEs, using the list of certified CDEs made available by the CDFI Fund. Only entities certified as CDEs may receive NMTC allocations. PolicyMap geocoded 276 CDEs with New Markets transactions as reported from 2005 – 2012, and was able to locate 100% of the addresses on a map.

CDE Locations are no longer updated by the source. PolicyMap will remove this data in 2020 unless the dataset is updated or another source is found.

Description:

The Community Development Financial Institutions (CDFI) Fund is a division of the U.S. Department of Treasury. Per the Consolidated Appropriations Act of 2012, funding was provided for several CDFI Fund programs (Bank Enterprise Award Program, CDFI Program, Healthy Food Financing Initiative, and Native American CDFI Assistance Program) on condition that a minimum of 10% of the projects served must be in persistent poverty counties. The legislation defines a persistent poverty county as any county that has had 20 percent or more of its population living in poverty for the past 30 years as measured by the U.S. Census Bureau. Based on this criteria, the CDFI Fund used data from the 1990 and 2000 decennial censuses, and the 2016-2020 American Community Survey to determine qualifying counties.

Description:

The Community Development Financial Institutions (CDFI) Fund, a division of the US Department of the Treasury, is now using a priority point system for scoring its applications. Applicants are awarded up to 5 “priority points” for their commitment to serve communities facing the highest levels of distress.

The score is tabulated using various distress indicators, which are also mapped on PolicyMap. These indicators are: poverty rates, median household income, unemployment rates, home foreclosures and high-cost mortgages.

Description:

Opportunity Insights, a team of researchers and policy analysts based at Harvard University, conducted a longitudinal study of economic and social conditions of adults based on where they were raised. One of the products of this study was a dataset published at https://opportunityinsights.org/. This dataset includes average household earnings and incarceration rates for adults who were raised in low income households in a given tract, county, or commuting zone.

The researchers used demographic data from the 2000 and 2010 Census short forms, combined with data from the 2000 Census long form and 2015 American Community Survey. They linked this Census data with tax returns from 1989, 1994, 1995, and 1998 to 2015. By combining all this data at the person level, they were able to match people who were born from 1978 to 1983 with the census tracts where they were born and raised, and with the household earnings of their parents. They used this longitudinal dataset to calculate the incarceration rate (per 100 people) for people raised in households with incomes less than the 25th percentile based on whether they were in jail or prison on April 1, 2010, the reference date of the 2010 Census for two genders—men and women—and three racial and ethnic groups—Black, Hispanic, and White. Average household income was also calculated for this age cohort and parental income group using the income data from the tax returns. Incarceration rates and average incomes were prorated based on how much time the person spent in a given tract or county in their youth, and information was suppressed for areas with fewer than 20 children. Some noise was also infused into the source data to preserve privacy.

Description:

PolicyMap calculated the diversity index and predominant race/ethnicity data layers using Census’ American Community Survey estimates. For both the diversity index and predominant race/ethnicity, PolicyMap used a total of 8 non-overlapping racial and ethnic categories provided by the US Census Bureau. These included the ethnic category Hispanic and the following 7 Non-Hispanic racial categories: White, African American, American Indian or Alaska native, Asian, Native Hawaiian or Pacific Islander, some other race, and two or more races. For predominant race, PolicyMap used 7 non-overlapping racial categories: White, African American, American Indian or Alaska native, Native Hawaiian or Pacific Islander, some other race, and two or more races.

The diversity index reflects the probability that any two people chosen at random from a given study area (e.g., block group) are of different races or ethnicities. An index value of 0 indicates complete homogeneity (i.e., an area’s entire population belonging to one racial or ethnic group), while the maximum index value represents complete heterogeneity (i.e., each racial or ethnic group constituting an equal proportion of an area’s population). The maximum value is calculated as one minus the reciprocal of the number of racial or ethnic groups. For example, with 3 racial or ethnic groups, the index value reflecting complete diversity would be 1-(1/3) or 67%. Articulated further, with 3 racial groups of equal proportions (i.e., complete diversity), the index equation becomes 1 – (0.33^2 + 0.33^2 + 0.33^2) = 67%. Given 8 racial or ethnic categories, the maximum value of the index displayed on PolicyMap is 87.5%.

With the diversity index data layer, lower index values between 0 and 20 suggest more homogeneity and higher index values above 50 suggest more heterogeneity. Racial and ethnic diversity can be indicative of economic and behavioral patterns. For example, racially and ethnically homogenous areas are sometimes representative of concentrated poverty or concentrated wealth. They could also be indicative of discriminatory housing policies or other related barriers.

The predominant racial or ethnic group is calculated as the racial or ethnic group constituting the highest proportion of the population in a given geography. A “tie” was noted in cases where the more than one racial or ethnic group shared the highest percent of population, rounded to the nearest 0.1 percent.

Description:

PolicyMap calculated Theil’s H index of segregation using the U.S. Census Bureau’s American Commmunity Survey. For this index, PolicyMap used a total of 8 non-overlapping racial and ethnic categories provided by the US Census Bureau. These categories included the ethnic category Hispanic and the following 7 Non-Hispanic racial categories: White, African American, American Indian or Alaska native, Asian, Native Hawaiian or Pacific Islander, some other race, and two or more races. Theil’s H is an index ranging from 0 to 1 that estimates the extent to which racial and ethnic groups are evenly distributed in a sub-area as compared to a larger area. Values approaching 0 suggest that sub-areas have a composition similar to the larger area (i.e., even distribution, less segregation) and values approaching 1 suggest that the racial and ethnic composition of sub-areas within a larger area deviates from the larger area (i.e., non-uniform distribution, more segregation). On PolicyMap, sub-areas are defined at the level of the Census block and are compared to the following larger areas: block groups, tracts, counties and Core-Based Statistical Areas (CBSAs), which are an approximation of metropolitan areas. The calculation of Theil’s H is based on the methodology presented in the report, “The Multigroup Entropy Index” by John Iceland (2004). This methodology involves calculating the entropy, a measure of diversity, for each sub-area and larger area and calculating the population-weighted deviation in entropy values across all sub-areas within each larger area.

Geographies for which no data or limited data were provided or for which the population was less than 10 are represented as having “Insufficient Data.”

Description:

Business Dynamics Statistics (BDS) provides data over time on openings and closings of businesses, as well as detailed information about those businesses. Data comes from the Longitudinal Business Database. Data provided on PolicyMap includes number and percent of firms of various sizes and ages. This can be used to see the activity of small businesses and startups, as well as larger and established businesses. Firms included in our calculations include all which have at least one active physical establishment in a given area. (If one firm has two establishments in two different areas, it will be counted once in each location.) Firm age is calculated by assigning an initial age according to the age of the oldest establishment that is part of the firm when it is created. (If a firm is created out of a merger or acquisition, its age is based on the age of the oldest establishment in the original firm or acquisition.) Firm size is calculated by the average employment in the current and previous year.

BDS data is available at the state and CBSA level. To view CBSA data, select “Metro Area” from the “Shaded by” menu in the map legend.

Description:

County Business Pattern Data (CBP) is an annual series that provides economic data by industry. The data describe the number and type of jobs that are located in any given place. This is different from describing the occupations of people living in the same area.
CBP covers most of the country’s economic activity. The series excludes data on self-employed individuals, employees of private households, railroad employees, agricultural production employees, and most government employees.
CBP data are extracted from the Business Register, the Census Bureau’s file of all known single and multi-establishment companies. The Company Organization Survey (annual) and Economic Censuses (every five years) provide individual establishment data for multi-location firms. Data for single-location firms are obtained from various surveys conducted by the Census Bureau, such as the Economic Censuses, the Annual Survey of Manufacturers, and Current Business Surveys, as well as from administrative records of the Internal Revenue Service, the Social Security Administration, and the Bureau of Labor Statistics.
Jobs in the CBP data are reported by North American Industry Classification System (NAICS, pronounced “Nakes”) categories. NAICS is the standard for use by Federal statistical agencies in classifying business establishments for the collection, analysis, and publication of statistical data related to the national business economy. NAICS is run through the Office of Management and Budget (OMB), and, in 1997, replaced the Standard Industrial Classification (SIC) system. Business establishments self-assign their NAICS code based on the primary economic activities in which they engage.
CBP data is given using two different methodologies. At the county, CBSA, state, and nation level, values are given for the number of employees in a given industry. These values are infused with noise added by the Census in order to avoid disclosing data that would be identifiable to a specific employer. In addition, some values are withheld by the Census to avoid disclosing data for individual companies. In these cases, the Census provides a range within which the value falls, but these are not included on PolicyMap, and are shown as Insufficient Data.

On PolicyMap, the data is also given as an “estimate”. For each industry, at every geography, (including Zip code) CBP provides values for the number of establishments that fall in various ranges of number of employees. These ranges are 1-4 employees, 5-9, 10-19, 20-49, 50-99, 100-249, 250-499, 500-999, and 1000 or more. There are additional higher ranges at the county and CBSA level. PolicyMap takes the midpoint of each range (so, for 10-19 it would be 14.5) and multiplies that by the number of establishments. Establishments with 1,000 or more employees are given a value of 1,750. This number for each range is added together to get the estimate. Though this number is less precise, its advantages are that there are no longer suppressions (so there is better coverage on the map), and ZIP code data is included.

Description:

The LEHD Origin-Destination Employment Statistics (LODES) datasets are released at the Census block level in a series of state-based files available for download here: http://lehd.ces.census.gov/data/#lodes. This is the same data as what is available on Census’ OnTheMap application: http://onthemap.ces.census.gov/. The Longitudinal Employer-Household Dynamics (LEHD) program is part of the Center for Economic Studies at the U.S. Census Bureau. The LEHD program combines federal, state and Census Bureau data on employers and employees under the Local Employment Dynamics (LED) Partnership. Under the LED partnership, states share Unemployment Insurance earnings data and the Quarterly Census of Employment and Wages (QCEW) data with the Census. Census combines these data with additional federal administrative data, Census data, and surveys. The LEHD program also creates a partially synthetic dataset on workers’ residential patterns, offering a dynamic link showing where people live and where they work. LED was built state by state, and a handful of state-year combinations are not available. These include Alaska (2017-2021), Arizona (2002, 2003), Arkansas (2002, 2019-2021), the District of Columbia (2002-2009), Massachusetts (2002-2010), Mississippi (2002 , 2003, 2019-2021), New Hampshire (2002), Puerto Rico (all years), and the U.S. Virgin Islands (all years). Data on the resident workforce exists for these locations but are not included on PolicyMap due to their incompleteness. The resident workforce values for these missing state-year combinations only show residents who live in these states but work elsewhere. For more on the LED Partnership see: http://lehd.ces.census.gov/state_partners/.

Federal employment is not counted in state Unemployment Insurance data, and as a result federal employment was not included in LEHD until 2010. Data shown by default on PolicyMap does not contain federal employment. However, additional variables in the legend allow users to see data with federal employment included for years 2010 onwards. In order to calculate percent changes in employment and workforce numbers, PolicyMap subtracted federal employment from certain indicators (as noted) from 2010 to the most recent years available. Some demographic variables were introduced in 2009, including race, ethnicity, educational attainment, and sex. Because of the incomparability of the data between 2009 and 2010 with the introduction of Federal employment, PolicyMap chose to begin mapping these additional indicators in 2010.

LODES data are released at the 2020 Census tabulation block geographies and PolicyMap aggregated up to the larger geographies using Census provided relationship tables. PolicyMap displays all LEHD data at the TIGER 2010 boundary geographies, except CBSAs, which are shown at 2019 geographies, and congressional districts, which are shown at the 116^th congressional districts.

Description:

The U.S. Census Bureau’s Manufacturing, Mining, and Construction Statistics Division provides annual and monthly estimates of housing units authorized by building permit officials. Data are available at the county, state, CBSA, and national geographies. These estimates are aggregated and imputed from reports submitted by local permit-issuing offices. Most permit-issuing offices are municipalities, and the rest are counties, townships or towns. 9,000 out of the 20,000 permit-issuing places submit the Form C-404 report “Report of Building or Zoning Permits Issued and Local Public Construction” on a monthly basis. The other places are surveyed only annually. The 9,000 surveyed monthly include: all permit-issuing places in the 75 Metropolitan Areas (MAs) with the largest number of permits (as of 2002); all permit-issuing places in states with limited numbers of permit-issuing places; permit-issuing places with special data reporting arrangements. The rest of the sample is stratified by state. Prior to 2005, monthly counts of building permits were based on a different sample of 8,500 out of 19,000 permit-issuing agencies. As a result, comparisons of building permit issuances between 2004 and 2005 should be made very cautiously. If a building permit report is not received for a given month or year, the missing data are either obtained from the Survey of Use of Permits (SUP) or imputed. The SUP is an annual survey of a smaller sample of permit-issuing areas that gathers data on housing construction, completion, sales, and characteristics of new housing. Monthly state and national data are estimates based on the sample data collected from the 9,000 permit-issuing agencies. For information on standard errors associated with these estimates see: http://www.census.gov/construction/bps/. Monthly county data are counts rather than estimates, and are therefore reported only for those counties where every permit office issues monthly reports. Annual data are obtained by summing monthly data reporters. If permit-issuing agencies submit both monthly and annual reports, the annual count is used. The annual building permit data on PolicyMap is unadjusted data.

Building permit data will not accurately reflect construction activity in those areas where building permits are not issued. Nationally, only roughly 2 percent of housing starts are issued in areas not requiring permits, however this varies greatly state to state and region to region.

Description:

The United States Census Bureau’s Planning Database (PDB) was designed as an aid for survey and census planning purposes. It contains information on the mail return rate for the 2010 census, self-response rate for the 2013-2017 ACS, and Low Response Score, all of which may be helpful in targeting areas for a higher response rate in future counts and surveys. In the 2010 census, households that did not return their census form were contacted through a non-response follow-up (NRFU), which utilized various methods to count these households. Non-response follow-ups require more resources than mail returns, so this data can be used to strategically increase mail returns in certain areas.

Description:

The Bureau of Labor Statistics’ Local Area Unemployment Statistics (LAUS) program produces monthly and annual employment, unemployment, and labor force data for Census regions and divisions, States, counties, metropolitan areas, and many cities, by place of residence. PolicyMap contains county and state counts of people employed, unemployed, and in the labor force, as well as the unemployment rate. The annual values presented in PolicyMap are annual averages for the years listed as provided by the BLS.

The concepts and definitions used by LAUS come from the Current Population Survey (CPS), the household survey that is the official measure of the labor force for the nation. According to this definition, employed persons include people who did any paid work as employees, worked in their own business or farm, or did unpaid work of 15 or more hours in an establishment owned by a relative. Unemployed persons include people who had no employment but were available for and seeking employment. People in the labor force are all those people classified as employed or unemployed. The labor force does not include military (active duty) and institutionalized persons.

Every April, the Bureau of Labor Statistics re-releases revised data for the previous five years. PolicyMap’s data reflects these revisions.

In 2015, the BLS changed the methodology of their model, with changes relating to structural differences, real-time benchmarking, smoothed seasonal adjustment, and treatment of outliers. At the state-level data, this methodology was used to re-estimate all data back to 1976 (PolicyMap displays data for 2000 and later). At sub-state geographies (city, county, CBSA, metropolitan division), the new methodology was only used to re-estimate data going back to January 2010. For this reason, comparisons between data before and after this change are not advised. Certain sub-state areas have the revised methodology re-estimated back to 1990: New York City (back to 1976), the Los Angeles-Long Beach-Glendale metropolitan division, the Chicago-Naperville-Arlington Heights, IL metropolitan division; the Cleveland-Elyria, OH metropolitan area; the Detroit-Warren-Dearborn, MI metropolitan area; the Miami-Miami Beach-Kendall, FL metropolitan division; and the Seattle-Bellevue-Everett, WA metropolitan division. Note that the previous methodology disaggregated data from the state level; because the state methodology is being re-estimated for before 2010, sub-state level data is indirectly affected.

Data for Connecticut counties is unavailable prior to 2020 due to boundary changes introduced by BLS. Data for Connecticut is available 2020 onward at the new county-equivalent, planning region boundaries.

The annual data contains percent change calculations on the number of people in the labor force and employed, as well as a change in percent calculation of the unemployment rate. These were calculated by PolicyMap.

Description:

The U.S. Census provides annual survey data on public school finances. Data is available at school district geographies. It includes the following indicators: student enrollment, total elementary-secondary revenue, total revenue from federal sources, total revenue for Title I, total revenue for children with disabilities, total revenue for child nutrition act, total revenue from state sources, total revenue from local sources, total elementary-secondary expenditures. Rates calculated by PolicyMap (such as revenue per student) were suppressed for school districts with 0 students in the given year.

Description:

The Census’ Small Area Income & Poverty Estimates (SAIPE) dataset provides more current estimates of selected income and poverty statistics than the most recent decennial census. Estimates are created for states, counties, and school districts, depending on the data. This dataset mainly serves administrators of federal programs who need current statistics on the demonstrated need of places.

Description:

The Census Bureau identifies urban areas (UAs) following every decennial census. For the 2020 decennial census, UAs are defined as a densely settled core of census blocks that encompasses at least 2,000 housing units or has a population of at least 5,000. UAs also include adjacent territory containing non-residential urban land uses. The Rural designation encompasses all areas not included within an UA.

The Urban and Rural Classifications from the Census Bureau are released at the Census Block level. The indicators in PolicyMap are represented at the Census Block Group and Census Tract Level. Areas are identified as urban if a specified threshold of land area sits within the Census designated UA – 50 percent threshold or 80 percent threshold. These indicators were developed by PolicyMap through a spatial join of Block Group and Census Tract boundaries to Census UA boundaries.

Description:

The data presents cancer incidence rates and number of new cases of cancer by type per year in state and county level geographies. The data is available by race/ethnicity and cancer type. This data is collected by the CDC from public health surveillance systems by using either their published reports or public use files. Many state departments of health publish state-specific cancer data. This data may be more recent or may provide more detail than the data published nationally. For all data years, rates and counts are suppressed if fewer than 16 cases were reported in a specific category, such as cancer type, race and/or ethnicity, age, and state. This suppression is to ensure confidentiality and reliability of rate estimates.

For 2017-2021 data, Indiana did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas because state legislation and regulations prohibit the release of county-level data to outside entities.

For 2016-2020 data, Indiana and Nevada did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas and Minnesota because state legislation and regulations prohibit the release of county-level data to outside entities. County data from Virginia are suppressed due to incomplete data. Due to a coding issue reported by North Dakota and Wisconsin, state- and county-specific counts and rates by race and ethnicity are not presented for North Dakota. For Wisconsin, state- and county-specific counts and rates are not presented for Hispanic persons. These data for all races and ethnicities are included in national rates.

For 2015-2019 data, Nevada did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas and Minnesota because of state legislation and regulations which prohibit the release of county-level data to outside entities. In addition, Kansas opted not to present state- and county-specific Asian and Pacific Islander counts and rates for these years. The national rates presented include data for Kansas. Please visit U.S. Cancer Statistics data website for detailed technical documentation.

Description:

The chronic condition prevalence index was created by following the CDC’s guidelines for creating estimates of high prevalence of multiple chronic diseases. Twelve chronic diseases were included: obesity, hypertension, high cholesterol, coronary heart disease, chronic obstructive pulmonary disease, asthma, chronic kidney disease, diabetes, cancer (excluding skin cancer), depression, stroke, and arthritis. A composite score was calculated by ordering each disease estimate and assigning a score of 0 to ZCTAs in the bottom 25th percentile, 1 for ZCTAs in the middle (25th to 75th percentiles) or 2 to those in the top 25th percentile. These scores were summed so each ZCTA has a score range of 0-24. This data was then rated as “lowest” if it was in the bottom 25th percentile, “moderate” if in the middle percentile (25-75 percentile) and “highest” if in the top 25th percentile.

The predominant chronic conditons was calculated from CDC’s PLACES data. The same twelves chronic conditions that went into the chronic condition prevalence index were also used. The predominant chronic condition in a given geographic area is the one that accounts for the highest percentage of individuals that experience that chronic condition.

Description:

The CDC posts daily data updates on COVID-19 vaccination administration and state distribution, assembled from data reported by state public health agencies and offices. Healthcare providers report doses administered to federal, state, territorial, and local agencies within 72 hours of administration, but there could be an additional lag before data reaches the CDC. Each entity may use various reporting methods, including immunization information systems, Vaccine Administration Management System, which supports temporary, mobile, and satellite clinics, in addition to direct submissions. The CDC’s data might differ from state systems and dashboards due to unfound duplicates during the agency’s data consolidation or due to varying reporting practices. Visit https://covid.cdc.gov/covid-data-tracker/#vaccinations to see data validation and collection protocols. Doses delivered and administered for U.S. States, D.C., and Puerto Rico are cumulative counts of COVID-19 vaccine doses reported to the Federal COVID Vaccine Operation delivered since December 14, 2020. Doses delivered to the U.S. Virgin Islands, Palau, Micronesia, Marshall Islands, Guam, American Samoa, and Northern Marianas Islands include those marked as shipped in CDC’s Vaccine Tracking System (VTrckS) since December 13, 2020. Doses delivered and administered in a state or territory also include those delivered and administered in pharmacies and the Federal Pharmacy Partnership for Long-Term Care (LTC) Program in the jurisdiction as reported in VTrckS. Doses administered are attributed to the jurisdiction in which the vaccine was administered. People receiving one or more doses represent the total number of people who have received at least one vaccine dose. People receiving two doses represent the number of people who have received a second dose of the vaccine. The number of people receiving one or more doses and the number of people receiving two doses was determined based on CDC’s information by state, territorial, and local public health agencies and federal entities on dose number, administration date, recipient I.D, and date of submission. A dose number was determined for nearly all reported doses administered; some missing data for dose number resulted in people receiving one or more doses and people receiving two doses not equaling the total doses. Rates per 100,000 represent the number of total doses delivered, the number of total doses administered, the number of people receiving one or more doses, and people receiving two doses per 100,000 residents of all ages. These metrics use the U.S. Census Bureau Annual Estimates of the Resident Population for the United States and Puerto Rico, 2019. 2018 U.S. Census Bureau population estimates and estimates from the CIA World Factbook are used for American Samoa, Federated States of Micronesia, Guam, Northern Mariana Islands, Republic of Palau, Marshall Islands, and U.S. Virgin Islands. Emergency Use Authorization has been granted for using the Pfizer-BioNTech vaccine among persons aged 16 and older and using the Moderna vaccine among persons aged 18 and older. Therefore, vaccine use is limited among those under age 18, representing approximately 22% of the U.S. population. Jurisdictions may use more targeted population counts for the denominators in their rate calculations (i.e., persons over 18 or over 16 years old), resulting in values different from those reported on the CDC COVID Tracker. In some limited circumstances, people might receive vaccinations outside of their state or territory of residency. These rates currently account for vaccinations occurring in the jurisdiction where the vaccination was administered.

Occasionally, states will overreport or underreport vaccine administration and distribution causing “Insufficient Data” to appear on the map for Change in number of people who are fully COVID-19 vaccinated, and Percent change in people who are fully COVID-19 vaccinated indicators.

Description:

The Centers for Disease Control (CDC) Flu Activity & Surveillance System dataset provides estimates of flu activity for states and territories of the U.S. Flu activity indicators are a measure of the proportion of visits to healthcare providers for influenza-like illness (ILI) symptoms. Estimates are collected from public health facilities participating in the Outpatient Influenza-like Illness Surveillance Network (ILINet). These data may disproportionately represent certain populations within a state; for instance, a severe flu outbreak in one city or region may cause the statewide activity level to be High, even if flu activity is low or minimal in other areas throughout the state. State health departments may have more geographically precise information available; contact information for these departments is available in FluView. Geographic spread of influenza is reported directly to CDC by state epidemiologists. This is a measure of how much of each state is affected by flu, and is not a measure of the severity of influenza activity. Weekly data and state and local surveillance information are available at the CDC Influenza Surveillance website. ILI activity and geographic spread measures are provided weekly. To obtain seasonal values, PolicyMap calculated the average of the numerical activity levels for all weeks ending in a given season. Flu season is defined as the period beginning in October and ending in May. Only states with at least 24 weeks of activity per season are included in these calculations.

New York City reports flu data to CDC separately from New York State; as such, New York State flu activity and geographic spread measures do not take New York City into account.

Description:

The Heat and Health Index (HHI) helps identify communities where people are most likely to feel the effects of heat on their health, in order to build towards a healthier and more heat-resilient future for all. The (HHI) is a national tool that incorporates historical temperature, heat-related illness, and community characteristics data at the ZIP code or ZCTA level to identify areas most likely to experience negative health outcomes from heat and help communities prepare for heat in a changing climate. Each ZCTA has a single ranking for the overall HHI and rankings for individual components so that users can make informed decisions to prepare for and prevent the negative health impacts from heat in their communities.

The HHI consists of 25 data indicators of heat and health vulnerability to help communities prepare for warming temperatures in a changing climate. These indicators are grouped into 4 modules.

• The Historical Heat and Health Burden module captures measures of previous experience with heat at the local level (ZCTA or ZIP code).
• The Sensitivity module is comprised of pre-existing health conditions that may increase risk of negative health outcomes when the individual with the condition is exposed to extreme heat.
• The Sociodemographic module encompasses social and demographic characteristics that increase exposure or sensitivity to heat or lessen one”s ability to cope with extreme heat.
• The Natural and Built Environment module focuses on characteristics of the natural and built environment that increase exposure or sensitivity to heat or lessen one”s ability to cope with extreme heat.

Data in the HHI come from the Centers for Disease Control and Prevention (CDC), the National Emergency Medical Services Information System (NEMSIS), the United States Census Bureau, the Multi-Resolution Land Characteristics Consortium (MRLC), and the Environmental Protection Agency (EPA).

Description:

The Centers for Disease Control (CDC) dataset provides the number of births, the number and percent of infants born with birth weight under 2,500 ounces (low birthweight), the number and percent of infants born with birth weight under 1,500 ounces (very low birthweight), the number and percent of infants born vaginally or by cesarean section, the number and percent of births where prenatal care began during the first trimester and the number and percent of births where prenatal care was received in only the third trimester or not at all,. the number and percent of births where the number of prenatal visits met the recommendations from the American College of Obstetricians and Gynecologists (ACOG) (expected number of visits is 8–14 for a term pregnancy), the number and percent of births by mothers with specified maternal health condition (eclampsia, gestational diabetes, pre-pregnancy diabetes, gestational hypertension, pre-pregnancy hypertension). The CDC only reports numbers of births for counties with populations exceeding 100,000.

The CDC also provides numbers and rates for mothers under age 20. Additionally, this dataset includes the number and percent of births to mothers under the age of 20, with break outs for mother under age 18 and mothers 18 and 19 for select years. Data on prenatal care is only available for counties with populations of 100,000 or more.

PolicyMap has suppressed data for geographies where .4% or more of the birth characteristics, prenatal care, or maternal health condition indicators were unknown to the CDC.

Beginning in 2007, data are reported from the 2003 U.S. standard Certificate of Live Birth, with additions of information on birth anomalies and several less variables related to maternal risk factors than the previous 2003 revision.

Description:

The Centers for Disease Control (CDC) dataset provides the number of infant deaths, and the rate of deaths to infants for every 1000 live births by maternal residents of the US. The CDC only reports numbers of births for counties with populations of 100,000 or more and number and rate of infant deaths for counties with populations of 250,000 or more. It suppresses the rate where there are fewer than 20 deaths reported. Adult mortality data are taken from the National Center for Health Statistics’ Compressed Mortality file as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. The Compressed mortality file provides the number and rate of deaths, by age group and cause of death as reported through the tenth revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10). Data on PolicyMap represent deaths from Alzheimer’s disease, cancer, coronary heart disease, chronic lower respiratory disease, COVID-19, stroke, and chronic lower respiratory disease among those aged 45 or older, from 2000 through 2015. Deaths from homicide, suicide, motor vehicle traffic, and accidental injury for all age groups. These causes have topped the CDC’s list of leading causes of death since 2005. Underlying cause-of-death is indicated on the death certificate by the physician. The National Center for Health Statistics determines one cause of death when more than one cause or condition is entered by the physician. PolicyMap shows mortality data from 2000 through 2021. Adults ages 35 and older are used as a base category for deaths from disease because these age groups represent most of the deaths from the four leading causes. Rates are calculated per 100,000 population 35 and over in the source data using population estimates based on 2000 and 2010 U.S. Census counts. The CDC’s National Center for Health Statistics released an estimated model of drug overdose data in its Data Visualization Gallery. Smoothed crude death rate estimates were generated using Hierarchical Bayesian models with spatial and temporal random effects. Bayesian hierarchical modeling “borrows strength” across geographic areas and allows estimates to be generated for counties that have small populations. Updated county-level estimates now include point estimates rather than estimate ranges. The CDC adds a disclaimer to this dataset that in certain states and years, for example New Jersey (2009) and West Virginia (2005, 2009), the rates may be lower than expected due to a large number of unresolved cases or misclassification of ICD-10 codes. More information on the CDC’s methodology is available here. Opioid and narcotic poisoning data comes from the CDC’s Multiple Cause of Death files. Drug overdose deaths were classified using the Tenth Revision (ICD-10) of the International Classification of Disease underlying-cause-of-death codes for drug poisonings (overdose): X40-44 (unintentional), X60-64 (suicide), X85 (homicide), and Y10–Y14 (undetermined intent). The types of opioid involved in drug overdose deaths were classified following the ICD-10 codes: and T40.1 (heroin), T40.2 (natural and semisynthetic opioids), T40.3 (methadone), and T40.4 (synthetic opioids, other than methadone). The category for all opioid overdoses includes all these categories (T40.1, T40.2, T40.3, and T40.4). T40.0 (opium) was not included since fewer than 10 people are reported each year as having died from opium overdose in the nation. Deaths involving multiple types of opioids are recorded in each applicable category, therefore the US totals may include overcounting. Heroin is an illegally-made semi-synthetic opioid derived from morphine. “Natural and semisynthetic opioids” is a category of prescription opioids, which includes natural opioid analgesics (codeine, morphine, etc.) and semi-synthetic opioid analgesics (hydrocodone, hydromorphone, oxycodone, and oxymorphone), but excludes heroin. Methadone is a prescribed synthetic opioid used to treat moderate to severe pain, and also withdrawal symptoms in those addicted to heroin or other narcotics. “Synthetic opioids, other than methadone” is a category of opioids commonly available by prescription and includes drugs such as fentanyl and tramadol, but excludes methadone. The CDC does not differentiate between deaths from pharmaceutical fentanyl and illegally-made fentanyl, and deaths from both forms are included in the data. While medically not considered a narcotic, cocaine is legally classified as such and is included in the CDC’s definition of narcotics along with opioids. The types of narcotics involved in drug overdose deaths were classified following the ICD-10 codes: T40.6 (other and unspecified narcotics), and T40.5 (cocaine). The category for all narcotics overdoses includes T40.1, T40.2, T40.3, T40.4, T40.5 and T40.6. The methods used to classify deaths on death certificates may lead to a significant undercount of opioid-related deaths, which could inaccurately portray the severity of this public health problem. Because of reporting discrepancies and nonspecific language, it is likely that national statistics underestimate by a substantial fraction the amount of opioid analgesic- and heroin-related deaths. Additionally, the degree of underestimation varies based on states’ death certification systems. For more information undercounting opioid-related deaths visit https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4547584/. To provide context for a given area, it is helpful to also look at how many overdose deaths are recorded with no additional drug information. These were classified according to the ICD-10 code of T50.9 (other or unspecified drugs).

For more information on the data visit https://wonder.cdc.gov/mcd-icd10.html.

Description:

The National Center for Health Statistics (NCHS) released the results of their Small-area Life Expectancy Estimates Project in September of 2018. For this project, NCHS calculated abridged life tables at the Census tract level. An “abridged life table” is a series of estimates of life expectancy for people of different ages grouped into multi-year cohorts. To create these abridged life tables, NCHS worked with the National Vital Statistics System to geocode residences recorded on death certificates from 2010 to 2015 (inclusive). The Department of Housing and Urban Development Geocode Center performed the geocoding. Maine and Wisconsin are excluded from this study because they did not have geocoded death certificates for 2010. The life table calculations also required Census tract level population estimates. NCHS worked with the Census Bureau to create a set of custom 6-year population estimates for the years 2010–2015 using data from the 2010 Census and the 2011–2015 ACS for use in this project. The NCHS used information on the location of residence and age of the deceased combined with the population estimates to build age patterns or “schedules of mortality” for 4,639 “model tracts.” Model tracts met two criteria—each had a population over 5,000 and one or more recorded deaths in each age group during the 6-year window. Using schedules of mortality from the model tracts, the NCHS developed statistical models to predict death rates based on demographic, socioeconomic, and geographic variables. They used these models to fill in data for tracts with smaller populations, and for tracts with age groups that had “missing deaths” or no recorded deaths within an age group. For the model tracts, reported values for all age ranges are calculated directly from the death certificates and population estimates. For the tracts with some missing deaths, values predicted by the statistical models are used for the age groups that had no recorded deaths, and observed values are used for all other age ranges. For some tracts with low populations, all values are based on predictions from the statistical models. See the complete documentation of the methodology here: https://www.cdc.gov/nchs/data/series/sr_02/sr02_181.pdf.

Description:

The Centers for Disease Control and Prevention HIV Incidence and Case Surveillance Branch provides the number of estimated active HIV infection cases among people aged 13 and older from state and local health departments. These data are available for states and counties. These data represent the place of residence at earliest HIV diagnosis; duplicate records from different states are reconciled by the source. Some states without confidential name-based HIV infection reporting have elected not to release state and/or county-level data.

CDC provides estimates on new HIV diagnoses for a given year. HIV data for a given year presented on PolicyMap represent the number of confirmed diagnoses of HIV infection or infection classified as stage 3 (AIDS) confirmed by laboratory analysis as of December 31 of that year and reported to the CDC by June 30 of the following year. These data may not be limited to new infections (i.e., incidence); rather, these data represent new diagnoses, as the date of infection may vary by individual. Estimates are statistically-adjusted values based upon actual case counts reported to CDC by state and local health departments. The CDC suppressed values in areas with fewer than 5 reported cases and/or population less than 100 as well as in the county with the lowest population in states where only one other county’s data were suppressed.

Data on the number of new cases of chlamydia, gonorrhea, and syphilis reported each year, and the rate of new STD cases reported for every 100,000 residents, by state and county are available from CDC. Data are based on cases of STDs reported to state and local health departments. Data is reported by both public and private agencies, such as STD clinics, counseling/testing sites, drug treatment clinics, family planning clinics, and private physicians. The CDC collects data from regional jurisdictions, and publishes the data in an annual report, which can be downloaded here: https://www.cdc.gov/std/stats/default.htm.

Syphilis is presented as a combined sum of cases classified in either primary or secondary stages of the disease. Other categories of syphilis – not included in the data – are latent (without symptoms), tertiary (late stage), and congenital (transferred from mother to child). Primary and secondary forms of the disease are the most infectious and therefore important when considering the risk of transfer and spread of disease.

Some variability in the amount of in the amount of reporting may exist across the country. Chlamydia, gonorrhea, and syphilis are considered Nationally Notifiable, which means that regional jurisdictions provide information to the CDC on a voluntary basis. A nationally notifiable disease is not necessarily reportable by law within a given state. Because of incomplete diagnosis and reporting, the number of STD cases reported is less than the actual number of cases occurring. The level of consistency may vary between local jurisdictions, reporting agencies, and reporting years. In some areas, reporting from public sources is thought to be more complete than reporting from private sources.

Incidence rates were calculated by the CDC using total population as the denominator. The population denominators used to compute these rates for the 50 states and the District of Columbia were based on the U.S. Census Bureau population estimates utilizing the OMB compliant race categories. Each rate was calculated by dividing the number of cases for the calendar year by the population for that calendar year and then multiplying the number by 100,000 Ten-year percent change variables for incidence of STDs and five-year percent change for HIV incidence were calculated by PolicyMap.

Starting with 2019 data, STI data presented in the AtlasPlus follows the 2017 Council of State and Territorial Epidemiologists (CSTE) data re-release rules. DSTDP suppresses data at any geographic level based on the following 2 conditions:

• 20% of individuals in a group have the specified disease
• Denominator population is less than 100
To prevent back-calculation of suppressed cells, primary suppression is augmented with complementary (or secondary) suppression in which data for additional groups are suppressed.

Description:

Opioid prescription rates per 100 persons, reported by the CDC, are calculated using population estimates from the Population Estimates Program, U.S. Census Bureau. Opioid prescriptions, including buprenorphine formulations commonly prescribed for treating pain (Belbuca and Butrans), codeine, fentanyl, hydrocodone, hydromorphone, methadone, morphine, oxycodone, oxymorphone, propoxyphene, tapentadol, and tramadol, were identified using National Drug Codes.Cough and cold formulations containing opioids and buprenorphine, an opioid partial agonist used for treatment of opioid use disorder as well as for pain, are not included. In addition, methadone dispensed through methadone treatment programs is not included. Source for all CDC prescription dispensing data comes from IQVIA Xponent (previously know as QuintilesIMS). IQVIA Xponent is based on a sample of approximately 56,500 retail (non-hospital) pharmacies, which dispense nearly 93% of all retail prescriptions in the United States. For this database, a prescription is a new or refilled prescription dispensed at a retail pharmacy in the sample and paid for by commercial insurance, Medicaid, Medicare, cash or its equivalent, and other third-party coverage. This database does not include mail-order prescriptions. Geographic location is based on the location of the prescriber. For the calculation of dispensing rates, numerators are the projected total number of prescriptions dispensed annually at the state, county, or national level. Annual resident population denominators were obtained from the U.S. Census Bureau. After a steady increase in overall opioid prescribing rates from 2006, total opioid prescriptions peaked in 2012 at 255 million and a rate of 81.3 prescriptions per 100 people. The rate does not represent the percent of the population receiving opioid prescriptions. Since an individual may receive multiple prescriptions in a year, many counties have rates that are greater than 100 prescriptions per 100 persons. Counties displayed as having insufficient data may indicate counties with no retail pharmacies, counties where no retail pharmacies were sampled, or counties where the prescription volume was erroneously attributed to an adjacent, more populous county according to the sampling rules used.

This data differs from the data shown in the July 2017 issue of CDC Vital Signs, which featured different facets of opioid prescribing from 2006 to 2015. For more information visit https://www.cdc.gov/vitalsigns/opioids/index.html.

Description:

Social Vulnerability encompasses demographic and socioeconomic factors—such as poverty, limited access to transportation, and crowded housing—that make certain communities more susceptible to hazards and stressors. These stressors may include natural or human-made disasters, like tornadoes or chemical spills, as well as disease outbreaks, such as COVID-19.

The CDC/ATSDR Social Vulnerability Index (SVI), managed by the Geospatial Research, Analysis & Services Program (GRASP), is a place-based index, database, and mapping tool designed to identify and quantify socially vulnerable communities. This index helps public health officials and local planners better prepare for and respond to emergencies, aiming to reduce human suffering, economic loss, and health inequities. Over time, the CDC/ATSDR SVI has evolved, and it strongly discourages comparisons across different versions of the database. Each database calculates percentile scores by ranking census tracts relative to others within the same year, so scores from different years are not directly comparable. Adjustments to the SVI categories have been made periodically to reflect updates in U.S. Census Bureau data and incorporate the latest research.

The current SVI uses 16 U.S. Census variables from the 5-year American Community Survey (ACS) to identify communities that may need support during or after disasters. These variables are grouped into four themes representing major areas of social vulnerability, which are then combined into a single measure. The GRASP program assigned each geography a percentile ranking for each variable and calculated an overall score for each category by summing these percentiles. Each of the four themes is then assigned a percentile ranking, which contributes to the overall Social Vulnerability Index. Based on the overall SVI score, communities are classified into four vulnerability levels: Low, Low to Moderate, Moderate to High, and High, dividing all tracts or counties into quantiles. A percentile ranking represents the proportion of tracts (or counties) with equal or lower vulnerability than a given tract or county. For example, a ranking of 0.85 signifies that 85% of tracts (or counties) in the state or nation are less vulnerable, while 15% are more vulnerable. Find more information on the CDC’s Social Vulnerability Index at https://svi.cdc.gov/.

Description:

The data presents cancer incidence rates and number of new cases of cancer by type per year in state and county level geographies. The data is available by race/ethnicity and cancer type. This data is collected by the CDC from public health surveillance systems by using either their published reports or public use files. Many state departments of health publish state-specific cancer data. This data may be more recent or may provide more detail than the data published nationally. For all data years, rates and counts are suppressed if fewer than 16 cases were reported in a specific category, such as cancer type, race and/or ethnicity, age, and state. This suppression is to ensure confidentiality and reliability of rate estimates.

For 2017-2021 data, Indiana did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas because state legislation and regulations prohibit the release of county-level data to outside entities.

For 2016-2020 data, Indiana and Nevada did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas and Minnesota because state legislation and regulations prohibit the release of county-level data to outside entities. County data from Virginia are suppressed due to incomplete data. Due to a coding issue reported by North Dakota and Wisconsin, state- and county-specific counts and rates by race and ethnicity are not presented for North Dakota. For Wisconsin, state- and county-specific counts and rates are not presented for Hispanic persons. These data for all races and ethnicities are included in national rates.

For 2015-2019 data, Nevada did not meet publication criteria and was excluded from the analysis. County data are not available from Kansas and Minnesota because of state legislation and regulations which prohibit the release of county-level data to outside entities. In addition, Kansas opted not to present state- and county-specific Asian and Pacific Islander counts and rates for these years. The national rates presented include data for Kansas. Please visit U.S. Cancer Statistics data website for detailed technical documentation.

Description:

The Centers for Medicare and Medicaid Services’ Chronic Conditions Data Warehouse contains claims information for persons enrolled in the Medicare fee-for-service (FFS) program. Only information for beneficiaries enrolled in both Part A and Part B is included; information for beneficiaries who have died during the study year is included. Non-FFS Medicare beneficiaries are those with partial Part A and/or Part B coverage and people and who were enrolled in Parts A and B Medicare and Medicare Advantage plan. Medicare Part A (hospital insurance) and Part B (medical insurance) cover individuals ages 65 and over who are receiving Social Security, people who have received disability benefits for at least two years, people who have amyotrophic lateral sclerosis (Lou Gehrig’s disease) and receive disability benefits, and people who have end-stage renal disease (permanent kidney failure) and receive maintenance dialysis or a kidney transplant. Individuals with Medicare Advantage (Part C) and Medicare Prescription Drug Plan (Part D) coverage are not represented in the data.

Chronic health condition data is based on CMS administrative enrollment and claims data for Medicare fee-for-service beneficiaries. A Medicare beneficiary is considered to have a chronic condition if there is a CMS claim indicating that the beneficiary received a service or treatment for that specific condition. Beneficiaries may have more than one of the chronic conditions listed.

In March 2022, the CMS Chronic Conditions Warehouse released an updated algorithm for their Chronic Condition indicators. Due to this algorithm update, some Chronic Conditions indicators may experience significant changes to prevalence rates (i.e., percentages). This update applies to 2021 data and forward. For more information, please visit CMS Mapping Medicare Disparities Tool Technical Documentation. Please note that Alcohol Use Disorder and Drug Use Disorder conditions are available for years 2007-2018, and 2021 due to CMS’s data availability.

CMS demographics, spending, and service utilization data available on PolicyMap comes from CMS’s Fee-for-Service Geographic Variation Public Use File (FFS GV PUF). This public use file is based primarily on information from CMS’s Chronic Conditions Data Warehouse.

In May 2024, CMS released 2022 data for the FFS GV PUF and updated historic data for years 2014-2021. This is due to CMS using different classification systems to define physician services between the 2014-2022 and 2007-2013 data years. For more information on methodology changes, please visit the FFS GV PUF Technical Documentation May 2024 Update.

All dollar amounts in this data set are standardized by CMS to adjust for factors that result in different payment rates for the same service, including local variations in wages and payments Medicare makes to hospitals to advance program goals (including training doctors). The standardized values represent what Medicare would have paid in the absence of those adjustments. Because the state of Maryland is exempt from reporting special payments to Medicare, costs in Maryland were standardized using different factors than the nationwide model.

Description:

Centers for Medicare and Medicaid Services’ (CMS) opioid prescription claims include information about Medicare and Medicaid prescriptions. CMS opioid claims and prescribing rates are available for all opioids as well as extended-release and long-acting (ER/LA) opioid formulations. Prescription numbers and rates for ER/LA opioids are also counted within the “all opioids” indicators. ER/LA opioids are designed to deliver more stable dosing for chronic pain patients and reduce dosage frequency but may pose a higher risk of overdose when misused. Certain methadone types are considered ER/LA. For a list of opioids included see https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Downloads/OpioidDrugList.zip.. Medicare Opioid Prescriptions CMS Medicare opioid prescription claims come from Medicare Part D claims prescribed by health care providers, as collected from Part D Prescriber public use files. Medicare opioid prescribing rates were calculated as the rate of opioid prescription claims per 100 total Medicare Part D prescription claims, including both prescriptions and refills. Medicare opioid prescription data includes claims for beneficiaries enrolled in Medicare Advantage Prescription Drug Plans and stand-alone Prescription Drug plans, but does not include prescriptions for patients on Medicaid, those with commercial insurance, or self-pay patients. Approximately 70% of Medicare beneficiaries have Medicare prescription drug coverage either from a Part D plan or a Medicare Advantage Plan offering Medicare prescription drug coverage. In 2017, Medicare Part D spending was $155 billion, while U.S. retail prescription drug spending was around $333 billion. Due to data redactions for geographies with 10 or fewer claims, county and Zip code Medicare opioid claim totals may not add up to state totals or may be lower than the true program totals. Medicaid Opioid Prescriptions CMS Medicaid opioid prescription claims come from claims where a portion was paid through Medicaid, as reported through the Medicaid State Drug Utilization Data. Medicaid opioid prescribing rates were calculated as the rate of opioid prescription claims per 100 total prescription claims for which Medicaid paid a portion. CMS Medicaid opioid prescription data includes claims prescribed through Fee-For-Service (FFS) programs and Manage Care Organizations (MCO). In 2017, Medicaid spending on prescription drugs was $68 billion, while U.S. retail prescription drug spending was around $333 billion.

As of the 2025 data release, Medicaid opioid claims are available at the zip, county, and state level. However, data may be sparse for smaller geographies due to CMS data processing methods. County-and zip-level measures presented in this tool were aggregated from the Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF). In order to be included in this tool, state-level TAF data had to meet quality checks for each data year. Due to known data quality issues in prior years, TAF data is presented only for data years 2016 and onward. In order to protect the privacy of beneficiaries, suppression is applied to the data. At each aggregated geographic level, any derived opioid claim counts between 1 and 10 are suppressed (replaced with missing a.k.a. Insufficient Data values). Secondary suppression is applied in cases where a single subgroup category is primary suppressed.For example, if the Fee-for-Service opioid claim count for a given state is primary suppressed but the Managed Care opioid claim count is not primary suppressed, then the Managed Care opioid claim count for that state must be secondary suppressed to prevent disclosure of the Fee-for-Service opioid claim count suppressed value. Secondary suppression prevents backing into a primary suppressed value by using the values from the other opioid claim counts such as total opioid claim count. These claims include all types of opioids.

Description:

The Hospital Compare dataset is part of a data repository maintained by the Centers for Medicare & Medicaid Services (CMS), focusing on the quality of care at over 4,500 Medicare-certified hospitals (including acute care hospitals, critical access hospitals (CAHs), children’s hospitals, and hospital outpatient departments) across the country. The dataset was created in collaboration with organizations representing consumers, doctors, hospitals, employers, accrediting organizations, and other federal agencies, as part of an overall effort to improve patient safety and care. The Hospital Compare dataset on PolicyMap includes data on:

• General information (Overall Rating, Mortality, Safety of Care, Readmission, Patient Experience, Effectiveness of Care, Timeliness of Care, Effective use of Medical Imaging): The overall hospital rating is given in stars from 1 to 5, while all other measures are designated as either being below, same, or above the national average. Data for these measures is compiled through the Inpatient/Outpatient Quality Reports and other programs mandatory for Medicare-certified hospitals. All comparisons on national average were standardized to ensure a common scale and direction for each measure. This implies that hospitals that perform above average on mortality or readmission comparison have a higher standardized z-score on these measures, based on lower mortality and readmission rates than the national average. The overall star rating in this section is intended primarily for acute care hospitals, due to which CMS may have omitted the measure for specialty hospitals. More details on the methodology for calculating overall ratings can be found here.
• Survey of patients’ experiences – (Ratings on Care transition, Cleanliness, Communication, Pain Management, Staff Responsiveness, Quietness, Discharge Information, and overall hospital ratings): All of these measures are assigned a star rating from 1 to 5. Data for these measures is compiled using the Hospital Consumer Assessment of Healthcare Providers and Systems survey [HCAHPS], which is administered to a random sample of adult inpatients after discharge. In order to receive HCAHPS Star Ratings, hospitals must have at least 100 completed HCAHPS surveys over a given four-quarter period. While all of the star ratings are based on direct patient responses, the summary star rating is calculated as a weighted measure using all categories of patient responses, including overall patient rating. More details on the methodology for calculating HCAHPS star ratings can be found here.
• Healthcare Associated Infections (HAI): HAI measures show how often patients in a particular hospital contract certain infections during the course of their medical treatment, when compared to like hospitals. These infections can often be prevented when healthcare facilities follow guidelines for safe care. Hospitals currently submit information on central line-associated bloodstream infections (CLABSIs), catheter-associated urinary tract infections (CAUTIs), colon and abdominal hysterectomy surgical site infections (SSIs), MRSA Bacteremia, and C.difficile laboratory-identified events. More details on the methodology for calculating HCAHPS star ratings can be found here.

This dataset is available on PolicyMap as point data based on hospital location, and can be viewed upon clicking each respective point. The CMS Hospital Compare data was joined by PolicyMap to hospital locations using data from HRSA. HRSA hospital location data can be found here.

Description:

The Medicare chronic condition prevalence index rating was created by following the CDC’s guidelines for creating estimates of high prevalence of multiple chronic diseases. Nineteen CMS Medicare chronic diseases were included: Alzheimer’s disease, arthritis, asthma, atrial fibrillation, autism spectrum disorder, cancer (excluding skin), COPD, depression, diabetes, heart disease, heart failure, hepatitis (chronic viral B & C), high cholesterol, hypertension (high blood pressure), HIV/AIDS, chronic kidney disease, osteoporosis, schizophrenia, and stroke. These county-level chronic disease prevalence estimates come from the CMS medicare chronic conditions data set. A composite score was calculated by ordering each disease estimate and assigning a score of 0 to counties in the bottom 25th percentile, 1 for counties in the middle (25th to 75th percentiles) or 2 to those in the top 25th percentile. These scores were summed so each county has a score range of 0-24. This data was then rated as “lowest” if it was in the bottom 25th percentile, “moderate” if in the middle percentile (25-75 percentile) and “highest” if in the top 25th percentile.

Description:

The Centers for Medicare & Medicaid Services’ national downloadable file contains individual-level data on medical providers who accept Medicare assignment. Data in the national downable file is collected from providers’ Medicare enrollment records and includes information on provider credentials, sub-specialties, practice locations, and whether Medicare assignment is accepted as payment in full. PolicyMap has aggregated this data to create facility-level points that group providers practicing the same sub-specialty at the same location. If a provider practices at multiple locations, they are counted at each applicable location.

In addition, PolicyMap has incorporated data from CMS’ PY 2022 Clinician Public Reporting: Overall MIPS Performance dataset to include facility-level information on the MIPS scores of providers at a location. CMS calculates MIPS scores at both the individual and group levels; PolicyMap uses only group-level scores to better reflect the collective performance of providers at a facility. When multiple groups practice at the same location, their scores are averaged.

Each point on the map is labeled with either the name of the facility where the providers are located or, when there is only a single provider, the provider’s name. When providers of different sub-specialties practice at the same facility, multiple points may appear at the same location.

Description:

The Civil Rights Data Collection (CRDC) contains wide-ranging data on equity among public schools. The data is based on school district responses to a biennial survey conducted by the Office of Civil Rights in the Department of Education. The indicators in this dataset can be grouped into the following categories:

• School Information
• School Enrollment (includes students by race and ethnicity, students with Limited English Proficiency, and students with disabilities)
• School Staff (FTE equivalent)
• Gifted and Talented Programs
• Retention
• In School Suspension (ISS)
• Out of School Suspension (OSS)
• Expulsions
• Harrassment and Bullying
• Criminal Offenses
• Referrals to Law Enforcement
• Arrests
• Restraint
• Seclusion
• Corporal Punishment.

Percent and rate indicators have all been derived by PolicyMap. School locations, names, district names, level, type, and charter school status are all from the NCES CCD.

Description:

The Centers for Disease Control (CDC) Flu Activity & Surveillance System dataset provides estimates of flu activity for states and territories of the U.S. Flu activity indicators are a measure of the proportion of visits to healthcare providers for influenza-like illness (ILI) symptoms. Estimates are collected from public health facilities participating in the Outpatient Influenza-like Illness Surveillance Network (ILINet). These data may disproportionately represent certain populations within a state; for instance, a severe flu outbreak in one city or region may cause the statewide activity level to be High, even if flu activity is low or minimal in other areas throughout the state. State health departments may have more geographically precise information available; contact information for these departments is available in FluView. Geographic spread of influenza is reported directly to CDC by state epidemiologists. This is a measure of how much of each state is affected by flu, and is not a measure of the severity of influenza activity. Weekly data and state and local surveillance information are available at the CDC Influenza Surveillance website. ILI activity and geographic spread measures are provided weekly. To obtain seasonal values, PolicyMap calculated the average of the numerical activity levels for all weeks ending in a given season. Flu season is defined as the period beginning in October and ending in May. Only states with at least 24 weeks of activity per season are included in these calculations.

New York City reports flu data to CDC separately from New York State; as such, New York State flu activity and geographic spread measures do not take New York City into account.

Description:

The Escrow Requirements under the Truth in Lending Act rule (known as the Escrows Rule) requires that certain creditors create escrow accounts for a minimum of five years for higher-priced mortgage loans (HPMLs), except HPMLs made by certain small creditors that operate predominantly in rural or underserved counties. Rural counties are defined by using the USDA Economic Research Service’s urban influence codes, and underserved counties are defined by reference to data collected under the Home Mortgage Disclosure Act (HMDA).

Description:

Data obtained from the Convenient Care Association on March 31, 2017. Includes only members of the Convenient Care Association.

Description:

The Community Reinvestment Act (CRA), which was enacted by Congress in 1977, is intended to encourage depository institutions to help meet the credit needs of the communities in which they operate, including low- and moderate-income neighborhoods, consistent with safe and sound banking operations. CRA requires that each insured depository institution’s record in helping meet the credit needs of its entire community be evaluated periodically. These examinations are conducted by federal agencies: the Board of Governors of the Federal Reserve System (FRB), the Federal Deposit Insurance Corporation (FDIC), the Office of the Comptroller of the Currency (OCC), and the Office of Thrift Supervision (OTS). That record is taken into account in considering an institution’s application for deposit facilities, including mergers and acquisitions. In order to gauge CRA performance, the evaluation looks for bank activity in low- and moderate-income neighborhoods, nonmetropolitan distressed and underserved areas, and federally designated disaster areas. These areas are identified by calculating tract income level. This is the Median Family Income (MFI) of each tract divided by Area Median Family Income (AMFI). For tracts located outside of an MSA/MD, the MFI used in the denominator is the statewide non-MSA/MD MFI. This figure is calculated using incomes from all areas of a state that are not assigned to MSA/MDs. As of 2022, FFIEC calculates tract income level using the Census’ 2016-2020 American Community Survey estimates. For additional information on data and calculations see: http://www.ffiec.gov/geocode/help3.aspx The tract income level is defined as follows: If the Median Family Income % is < 50% then the Income Level is Low. If the Median Family Income % is >= 50% and < 80% then the Income Level is Moderate. If the Median Family Income % is >= 80% and < 120% then the Income Level is Middle. If the Median Family Income % is >=120% then the Income Level is Upper. If the Median Family Income % is 0% then the Income Level is Not Known. Tracts are CRA eligible if they are low- or moderate-income, or if they are nonmetropolitan middle income tracts designated by FFIEC as distressed or underserved. Distressed middle income tracts are those with: (1) Unemployment rate at least 1.5 times the national average or (2) Poverty rate of 20% or greater or (3) Population loss of 10% or more between the 2010 and 2020 census, or a net migration loss of 5% or more between 2010 and 2020. Underserved middle-income tract are those designated by the Economic Research Service of the United States Department of Agriculture with an “urban influence code” of 7, 10, 11 or 12. Lists of these tracts are released annually and available on the CRA website at: http://www.ffiec.gov/cra/examinations.htm. PolicyMap also provides the distressed and underserved designation for tracts that were designated as such in the previous year but not the current year because of the allowance of a one-year “lag period.” According to the source, “this lag period will be in effect for the 12 months immediately following the date when a census tract that was designated distressed or underserved is removed from the designation list. Revitalization or stabilization activities undertaken during the lag period will receive consideration as community development activities if they would have been considered to have a primary purpose of community development if the census tract in which they were located were still designated as distressed or underserved.” For more information, see page 24 of this Federal Register document: https://www.gpo.gov/fdsys/pkg/FR-2016-07-25/pdf/2016-16693.pdf

To identify tracts that are “designated disaster areas” consult the Federal Emergency Management Agency (FEMA) website: http://wwww.fema.gov. Disaster designations are also mapped on PolicyMap and can be found in the Federal Guidelines menu under FEMA Disaster Declarations.

Description:

Data Axle provides locations and characteristics of a wide range of businesses, both large and small. PolicyMap licenses data on pharmacies, and presents the locations, addresses, store status, and information on a parent chain, such as CVS or Walgreeens if the pharmacy is part of a chain. Pharmacies are labeled as “Closed” if a location existed in DataAxle’s pharmacy data from August, 2024, and no longer exist in either the May 2025 or August 2025 data updates. This label is an assumption by PolicyMap to help users identify where pharmacy locations used to exist, and may not be inclusive of all closed pharmacies.

Description:

Data Axle provides locations and characteristics of a wide range of businesses, both large and small. PolicyMap licenses data on retail-based healthcare locations. These may be based in large pharmacy chains such as a CVS or Walgreens, or be independent. We present the locations, addresses, and information on a parent chain such as CVS or Walgreeens if the retail clinic is part of a chain.

Description:

The Agency for Healthcare Research and Quality’s (AHRQ) Compendium of U.S. Health Systems is a suite of files that provide information on hospitals, group practices, outpatient sites, nursing homes, and home health care organizations across the United States. Facilities in these files are linked through a “system” file that identifies which facilities belong to the same health system. The group practice file has not been published on PolicyMap because the data were not provided at the address level.

AHRQ defines a health system as follows: “A health system includes at least one hospital and at least one group of physicians that provides comprehensive care (including primary and specialty care) who are connected with each other and with the hospital through common ownership or joint management.” Additionally, a health system must include at least one non-federal acute care hospital, at least 50 physicians, and at least 10 primary care physicians.

Organizations in the source are aggregated to their highest level of ownership, meaning subsidiary systems of larger national systems are not listed separately. As a result, AHRQ notes that not all health systems operating in certain states or local markets may appear in the source. For more information on these files and AHRQ’s methodology for identifying health systems, please visit:https://www.ahrq.gov/chsp/data-resources/compendium.html.

Description:

The Department of Homeland Security’s Yearbook of Immigration Statistics is an annual publication on documented foreign nationals in the United States. PolicyMap contains state and CBSA-level data on the number of people granted Legal Permanent Resident (LPR) status by region of birth and by selected countries. If the volume of immigrants receiving green cards in any year was more than 15,000 people, the country was included.

Description:

The Energy Information Administration (EIA) U.S. Energy Mapping System provides the locations and capacity of operable electric generating plants which includes all plants that are operating, on standby, or short- or long-term out of service with a combined nameplate capacity of 1 MW or more. Generator-level data is collected as part of the EIA-860 Annual Electric Generator Report. Geographic coordinates are assigned to the plant locations in the source data. Thematic indicators of electricity generation capacity were determined based on a spatial join performed by PolicyMap of geocoded plant locations and standard Census geographic boundaries. Generator-level megawatt output capacity was aggregated for county and state boundaries.

References in the data to “renewable” energy sources refers to energy from biomass, hydropower, geothermal, wind, and solar. This is consistent with how the EIA classifies renewable energy sources as outlined on their renewable sources webpage.

Description:

The EPA regularly updates its Brownfields Sites Report List. The points in PolicyMap are as of April 2024. The coordinates used in PolicyMap are provided by the EPA. PolicyMap removes points that do not appear in the site’s listed state. The points shown on PolicyMap include brownfield sites that have received assessment, cleanup, and/or redevelopment funding from the EPA. Brownfields designated by states or local entities, sites that may qualify for but have not received EPA assessment funding, and underground storage tanks are not included on the map. Each point represents a transfer of funds related to a known brownfield site. Multiple points for the same brownfield location indicate multiple actions over a period of time; the entity receiving funds may differ.

PolicyMap also stamps brownfield site locations to Census Tract boundaries to produce a data layer showing tracts with a presence of brownfields.

Description:

The Environmental Protection Agency’s (EPA) Smart Location Database provides data on the relationship between land use and transportation efficiency. The Smart Location Database (SLD) summarizes several demographic, employment, and land use variables for Census block groups. PolicyMap made an excerpt from this larger body of work available on its platform. Frequency of transit service provides a general metric of the quality of public transit options in an area. EPA calculated transit frequency through an analysis of General Transit Feed Specification (GTFS) data between 4:00 and 7:00 PM on a weekday. Then, for each block group, EPA identified transit routes with service that stops within 0.4 km (0.25 miles). Finally, EPA summed aggregate service frequency by block group. Values for this metric are expressed as service frequency per hour of service. GTFS is a transit data reporting standard that allows public transit agencies to publish data in a standard format. EPA also calculated frequency of transit service per square mile by dividing frequency of transit service per hour by total land acreage then converting to units per square mile. Where the total land acreage was zero, total block group acreage was used as the denominator.

To create the indicators on job or workforce accessibility by auto travel, the EPA joined an origin-destination matrix to employment and demographic data from the 2010 Census. Although the transit accessibility indicators were analyzed the same way as the auto accessibility, it was analyzed for evening peak travel period only, as this is normally the period of relatively intense levels of transit service.

Description:

Locations of Superfund sites are from EPA’s Superfund Enterprise Management System. The Superfund program is an evaluation program for active and inactive hazardous waste sites. Sites on the National Priorities List (NPL) are included on PolicyMap, as well as sites proposed for the NPL or formerly on the NPL.

Human exposure and groundwater migration information are environmental indicators based on metrics set by the EPA. These indicators are used to measure progress made through site cleanup activities.

Description:

The Federal Bureau of Investigation’s Uniform Crime Reporting (UCR) Program compiles standardized incident reports from local law enforcement agencies in order to produce reliable, uniform, and national crime data. The UCR Program collects data on known offenses and persons arrested by law enforcement agencies, however, it does not record the findings of a court, coroner, jury, or the decision of a prosecutor. Previously, PolicyMap received county and city level data directly from the FBI that was used to publish incidents and rates of crime for counties, places, and states. Due to this source no longer being available, PolicyMap has transitioned to using crime estimates from the UCR Program’s Crime in the United States (CIUS) reports to publish state and MSA level data.

The CIUS report provides data on the volume and rate of violent and property crime offenses for the nation and by state using Summary Reporting System data and summarized data from the National Incident-Based Reporting System (NIBRS). As the UCR Program is voluntary, offenses may be estimated by the UCR if an agency does not provide twelve months of complete data. The estimation process considers factors such as the following: population size covered by the agency; type of jurisdiction, e.g., police department versus sheriff’s office; and geographic location. The CIUS report provides crime statistics for all fifty states, Metropolitan Statistical Areas (MSAs), and Metropolitan Divisions (MDs), which are subdivisions of MSAs that have a core population of at least 2.5 million people. More information on the CIUS estimation process can be found in the CIUS report’s methodology.

On January 1, 2021, the FBI’s UCR Program transitioned from the Summary Reporting System to the National Incident-Based Reporting System (NIBRS) for crime data collection. Due to this transition, participation in 2021 remained below a statistically acceptable level to be estimated at a national level. Therefore, the UCR program chose to publish a limited release of the CIUS report which did not include state and MSA estimates of crime data. PolicyMap is monitoring the UCR program page to see if a supplemental version of this report is published.

Limited data for 2022 were available for Florida, Illinois, Maryland, and Pennsylvania. Limited data for 2023 were available for Florida. Due to data availability, The City of New York, NY is not included in the 2022 New York-Newark-Jersey City, NY-NJ-PA Metro Area.

Description:

The Federal Bureau of Investigation’s Uniform Crime Reporting (UCR) Program compiles standardized incident reports from local law enforcement agencies in order to produce reliable, uniform, and national crime data. The UCR Program is voluntary, and includes data for only counties and cities with population over 10,000. As a result, coverage is not universal. The UCR Program collects data on known offenses and persons arrested by law enforcement agencies. The UCR Program does not record the findings of a court, coroner, jury, or the decision of a prosecutor.

The Hate Crime Statistics Program within the FBI’s (UCR) Program collects data regarding criminal offenses that were motivated, in whole or in part, by the offender’s bias against a race, ethnicity, ancestry, gender, gender identity, religion, disability or sexual orientation and were committed against persons, property, or society. Hate crime data is captured by including the element of bias in offenses already being reported to the UCR Program. State hate crime counts reflect the sum of all reported offenses from agencies within the state that submitted data to the FBI. The state population count used in the rate calculations is the total population of the state as reported in the Census’s Population Estimates Program. The State of Hawaii does not participate in the Hate Crime Statistics Program. Due to variation in reporting and hate crime definitions changing over time, FBI hate crime statistics should not be compared across states, and should not be compared from one year to another. An agency can report up to four bias motivation types per offense. Multiple-bias offenses are not common, but when they occur, they are double-counted in the value of the total number of hate crimes.

Description:

The Federal Deposit Insurance Corporation releases data on failures and assistance transactions of financial institutions in the United States and its territories. This data draws on information from two FDIC databases: Failures and Assistance Transactions United States and Other Areas (Table BF01, available here: https://banks.data.fdic.gov/explore/failures) which is updated on an ongoing basis, and the FDIC Institution Directory (https://www5.fdic.gov/idasp/warp_download_all.asp), which is updated weekly. The data includes banks that have failed since October 1, 2000. The fields in the data for the assets and deposits of the acquiring bank are from the most recent quarterly report by the FDIC at the time of the most recent data update (this may be different than the acquiring bank information at the time of the initial bank closing).

The FDIC’s Summary of Deposits (SOD) is an annual survey of branch offices for all FDIC-insured institutions, including U.S. branches of foreign banks. Data is updated by the FDIC annually.

Description:

Every two years, the FDIC sponsors the National Survey of Unbanked and Underbanked Households to collect data on the number of U.S. households that are unbanked and underbanked; their demographic characteristics such as household type, race/ethnicity, age, and level of education; and their reasons for being unbanked or underbanked. This survey is conducted by the U.S. Census Bureau as a special supplement to the Current Population Survey (CPS). The FDIC undertakes this effort to address a gap in the availability of comprehensive data on the number of unbanked and underbanked households in the United States.These estimates are based on data aggregated from the 2019, 2021, and 2023 surveys, pulled from the multi-year microdata published by the FDIC at https://www.fdic.gov/household-survey/data-downloads-and-resources.

Like with all estimates derived from survey data, the values published in these data layers are associated with some uncertainty. In order to help users understand the reliability of the estimates, PolicyMap published margins of error at the 90% confidence level associated with each estimate. This means that there is a 90% likelihood that if every household in the given geography was interviewed, the count or percent would be within the margin of error above or below the estimate derived from the survey sample. For example, 4.5% of households in Colorado were unbanked, with a margin of error of 1.6%. This may be expressed as 4.5% ± 1.6%, which means that there is a 90% likelihood that the true percentage of unbanked households in Colorado was between 2.9% and 6.1% (expressing the margin of error in this way is known as a “confidence interval”).

PolicyMap has also published a data flag to help users interpret these margins of error. The data flags are based on the relative standard error, or the ratio of the standard error to the count or precent. The RSE is calculated by dividing the standard error of the estimate by the estimate itself, then multiplying the result by 100. Estimates considered “reliable” have standard errors that are 15% or less of the estimate. Estimates tagged “use with caution” have standard errors between 15% and 30% of the estimate. Esimates with standard errors greater than 30% of the estimate have been suppressed.

Description:

The housing price index (HPI) is a broad measure of the movement of single-family house prices based on transactions involving conforming, conventional mortgages purchased or securitized by Fannie Mae or Freddie Mac. The HPI is a weighted, repeat-sales index, meaning that it measures average price changes in repeat sales or refinancings on the same properties.

The annual indices for smaller geographies should be considered developmental. As with the standard FHFA HPIs, revisions to these indexes may reflect the impact of new data or technical adjustments. Indexes are calibrated using appraisal values and sales prices for mortgages bought or guaranteed by Fannie Mae and Freddie Mac. An index is not reported in cases where sample sizes on mortgage transactions are too small for a given geographic area. These indices were developed as part of FHFA Working Paper 16-01 (Bogin, Alexander N., Doerner, William M. and Larson, William D. (2019). “Local House Price Dynamics: New Indices and Stylized Facts”. Real Estate Economics, volume 47, issue 2, pages 365-398). The initial FHFA WP 16-01 is accessible at http://www.fhfa.gov/papers/wp1601.aspx.

Description:

Credit security is the ability of households to access mainstream credit and keep current on debt payments. The Credit Insecurity Index, created by the Federal Reserve Bank of New York, is intended to capture all the components that inform a community’s credit health in a composite measure. The score is meant to be interpreted in terms of place, such as a state or a county. The Credit Insecurity Index is reflective of the credit conditions experienced collectively by the individuals and households that make up the geography being examined. While it is important to understand credit insecurity at an individual or household level, assessing and comparing community-level credit insecurity can help researchers, practitioners, and community stakeholders understand the broader implications credit insecurity has for a community’s ability to access affordable credit, be resilient in the face of shocks, pursue economic opportunities, and take on community-wide investments.

The calculation for the Credit Insecurity Index involves adding the share of people who are not credit included to a value that represents the degree to which the population that is credit included is nonetheless credit constrained. In other words, the Credit Insecurity Index Score = (Not Credit Included) + (Credit Included x .25*(Credit Constraints)) where Credit Constraints = No Revolving Credit + Credit Over-Utilization + Deep Subprime Credit Score + Struggling or Consistently Delinquent Payment History.

Description:

Feeding America’s Map the Meal Gap analysis is published annually to assess food insecurity and food costs across the United States, with estimates modeled at national, state, county, and congressional district levels. The data can help illustrate how food insecurity varies across different communities and population groups.

Dataset includes:

• Food insecurity rates and estimated counts (overall and for children)
• Breakdowns by race/ethnicity (Black, Hispanic, White non-Hispanic)
• Breakdowns by age (older adults and seniors, state-level only)
• Localized meal costs and food budget shortfall estimates
• State-level SNAP income thresholds and access gaps

Feeding America uses data from the Current Population Survey (CPS), American Community Survey (ACS), Bureau of Labor Statistics (BLS), and NielsenIQ. The relationship between food insecurity and factors such as poverty, unemployment, homeownership, and disability is modeled at the state level, then applied to local demographic and economic conditions.

Notes:

• SNAP income eligibility thresholds vary by state and affect access to nutrition programs.
• Methodology changes in 2023 limit comparability of food budget shortfall, local meal cost, and meal gap estimates across years.
• Modeled estimates represent approximations and are not direct counts.

More detail on Feeding America’s methodology can be found in the technical brief:
https://www.feedingamerica.org/research/map-the-meal-gap/how-we-got-the-map-data

Source: Ribar, D.C., Harris, V., Dewey, A., Dawes, S., and Engelhard, E. (2025). Map the Meal Gap: An Analysis of Local Food Insecurity and Food Costs in the United States in 2023. Feeding America National Organization.

Description:

The Community Reinvestment Act (CRA), which was enacted by Congress in 1977, is intended to encourage depository institutions to help meet the credit needs of the communities in which they operate, including low- and moderate-income neighborhoods, consistent with safe and sound banking operations. CRA requires that each insured depository institution’s record in helping meet the credit needs of its entire community be evaluated periodically. That record is taken into account in considering an institution’s application for deposit facilities, including mergers and acquisitions. CRA examinations are conducted by the federal agencies that are responsible for supervising depository institutions: the Board of Governors of the Federal Reserve System (FRB), the Federal Deposit Insurance Corporation (FDIC), the Office of the Comptroller of the Currency (OCC), and the Office of Thrift Supervision (OTS). PolicyMap extracted the database of lending activity from the Peer Small Business Data. PolicyMap aggregated the number of loans by amount of loan and by borrower revenue. PolicyMap also aggregated the number, average amount and percent of loans by top small business lenders and by top small farm lenders in order to construct categories that would be useful to policymakers and descriptive of neighborhoods and markets.

The 2004-2011 data is at the 2000 Census boundaries. The 2012-2019 data is at the 2010 boundaries. The 2022 data and beyond is at the 2020 boundaries. For percent changes, PolicyMap created a bridge table across 2000, 2010, and 2020 geographies in order to calculate previous years of data at proper Census boundaries. These previous years of data calculations are used for comparison to the current year of data.

Description:

The Religious Congregations and Membership Study, carried out by the Association of Statisticians of American Religious Bodies (ASARB), was conducted with significantly different methodologies in 2000 and 2010. 2000 data were collected by the ASARB and include statistics for 149 religious groups, including number of churches and adherents. Dale E. Jones, Sherri Doty, Clifford Grammich, James E. Horsch, Richard Houseal, Mac Lynn, John P. Marcum, Kenneth M. Sanchagrin and Richard H. Taylor supervised the collection. These data originally appeared in Religious Congregations & Membership in the United States, 2000: An Enumeration by Region, State and County Based on Data Reported by 149 Religious Bodies, published by the Glenmary Research Center. The 2000 data excludes most of the historically African-American denominations and some other major groups. In an effort to correct for this, in 2002 the ASARB released an adjusted rate of adherence to all denominations per 1,000 people. The adjusted rate is included on PolicyMap; because of this correction some counties will have rates in excess of 1000. For more on the corrections see Roger Finke and Christopher P. Scheitle’s article Accounting for the Uncounted at http://www.thearda.com/mapsReports/Accounting%20for%20the%20Uncounted.pdf. 2010 data were also collected by ASARB and include statistics for 236 religious groups, including number of churches and adherents. In contrast to the 2000 study, researchers obtained mailing lists for the eight largest historically African-American denominations. In addition to including membership information gathered from this list, online church locators were identified and used to identify additional congregation locations. For each congregation located in this way, a membership of 100 was assigned. However, it is important to note that, while the figures for African-American denominations are more accurate than those imputed for the 2000 U.S. Religion Census, the figures are still significantly lower than those reported by the denominations in the Yearbook of American and Canadian Churches, 2010. In total, the 236 groups reported 344,894 congregations with 150,686,156 adherents, comprising 48.8 percent of the total U.S. population in 2010.

The data reported on Jews and Muslims are estimates rather than counts. For more information on how these estimates were calculated, including changes in the estimation methodologies from the 2000 to 2010 surveys, see: http://www.thearda.com/mapsReports/rcms_notes.asp.

Description:

These data come from research put out by Harvard University and the University of California at Berkeley as part of The Equality of Opportunity Project. Through this project, the researchers set out to examine geographical differences in economic mobility rates throughout the country and to look at the impact of tax expenditures on intergenerational mobility. As part of this study, the researchers released data on the probability that a child growing up with parents with an annual household income in a certain income quintile will have an annual household in a certain quintile as an adult. On PolicyMap, we used this data to display the percent chance that children from low- and middle-income families will achieve certain income ranges as adults. These data are mapped to Commuting Zones (CZs), which PolicyMap created using geographic crosswalks provided by the source. Based on Census data, CZs are geographical aggregations of counties based on commuting patterns that are similar to metro areas but also cover rural areas. Children are assigned to the CZ based on their location at age 16 (no matter where they live today), and the location is thus interpreted as where the child grew up.

For a full report on the researchers’ findings, see http://obs.rc.fas.harvard.edu/chetty/tax_expenditure_soi_whitepaper.pdf. For more information about the Equality of Opportunity Project or the Commuting Zone geography, visit http://www.equality-of-opportunity.org.

Description:

PolicyMap downloads the geocoded Head Start locations using the Head Start locator at the website listed above. Head Start locations are classified as Early Head Start, Head Start, Migrant or Seasonal Head Start, or American Indian and Alaskan Native Head Start. Individual centers receive funding from a grantee authority and are located in defined federal regions. Head Start locations can be filtered on whether the operating status of the center is open, closed, or was unreported according to the source.

Description:

The Healthcare Cost and Utilization Project (HCUP) is a Federal-State-Industry partnership, sponsored by the Agency for Healthcare Research and Quality (AHRQ), that brings together the data collection efforts of State data organizations, hospital associates, private data organizations, and the Federal government to create a national resource of encounter-level healthcare data. HCUP opioid-related hospital use includes both hospital inpatient stays, recorded as hospital discharges, and emergency department (ED) visits. Each discharge or visit is recorded as a separate event, regardless of how many times an individual patient may visit a hospital in a year. Inpatient stays are when patients are admitted to and treated in community hospitals. Community hospitals are defined as short-term, non-Federal, general, and other hospitals, excluding hospital units of other institutions (e.g., prisons). Included among community hospitals are obstetrics and gynecology, otolaryngology, orthopedic, cancer, pediatric, public, and academic medical hospitals. Excluded are community hospitals that are also long-term care facilities such as rehabilitation, psychiatric, and alcoholism and chemical dependency hospitals. ED visits are defined as treat-and-release hospital encounters (i.e., they do not result in a hospital admission to the same hospital). The number of inpatient stays and ED visits are adjusted to account for missing data. Inpatient stays and ED visits for opioid-related hospital use include both opioid-related disorders and opioid poisoning and other adverse effects. Specifically, hospital discharges and ED visits related to opioid use were identified by any diagnosis in the following ranges of ICD-10-CM and ICD-9-CM codes: F11 series (except F11.21); T40 series (0X1, 0X4, 0X5, 1X1, 1X4, 2X1, 2X4, 2X5, 3X1, 3X4, 3X5, 4X1, 4X4, 4X5, 601, 604, 605, 691, 694, 695); 304.00-304.02; 304.70-304.72; 305.50-305.52; 965.00-965.02; 965.09; 970.1; E850.0-E850.2; E935.0-E935.2; E940.1. Inpatient stays and ED associated with multiple opioid diagnosis codes are only recorded once. For more information on these codes and how they changed from ICD-9-CM to ICD-10-CM codes, please visit https://www.hcup-us.ahrq.gov/datainnovations/ICD-10CaseStudyonOpioid-RelatedIPStays042417.pdf Age refers to the age of the patient at admission. Discharges or visits missing age are excluded from results reported by age. Income is based on the median household income of the patient’s ZIP Code of residence. Quartiles are defined so that the total U.S. population is evenly distributed across four groups. The value ranges for the national income quartiles vary by year. Income quartile is missing if the patient is homeless or foreign. Patient location is based on the National Center for Health Statistics scheme to study the relationship between urbanization and health. For this dataset, there are five categories: large central metropolitan (counties in metropolitan statistical areas (MSAs) of 1 million or more population that contain the entire population of the largest principal city of the MSA, have their entire population contained in the largest principal city of the MSA, or contain at least 250,000 inhabitants of any principal city of the MSA); large fringe metropolitan (suburbs) (counties in MSAs of 1 million or more population that did not qualify as large central metropolitan counties); medium metropolitan (counties in MSAs of populations of 250,000 to 999,999); small metropolitan (counties in MSAs of population less than 250,000); and rural (comprised of counties in micropolitan statistical areas and nonmetropolitan counties that did not qualify as micropolitan). Information about insurance type is a hospital’s response to the question, “Who is expected to pay the hospital for a given service?”, which may be different than the actual means of payment. Expected primary payers include: Medicare, Medicaid, private insurance, and the uninsured. Discharges and ED visits with other, missing, or invalid expected primary payer are not reported in Fast Stats reporting by payer. These excluded records typically represent approximately 3 to 6 percent of all discharges or visits. Discharges with the expected primary payer of self-pay, charity, and no charge are classified as uninsured. For HCUP Partner organizations that identify State and local programs serving low-income, uninsured populations (e.g., Indian Health Services, county indigent, migrant health programs, Ryan White Act, Hill-Burton Free Care), discharges for these payers also are classified as uninsured. About one-third of the HCUP Partner organizations include this level of detail in their coding of expected payer. The rate of inpatient stays and rate of ED visits are calculated per 100,000 people (U.S. residents). Population-based rates are presented for trends of opioid-related inpatient stays and ED visits reported overall and by age, community-level income, and patient location. HCUP uses population and demographic data from Claritas, a vendor that compiles U.S. Census Bureau data. Rates are not calculated for expected payer (insurance type) because there is no current source of national population insurance estimates that align with HCUP’s definition of expected primary payer.

For more information on the data and methodology visit https://www.hcup-us.ahrq.gov/faststats/OpioidUseServlet.

Description:

PolicyMap downloads Hospitals, Federally Qualified Health Center (FQHC), Home Health Agencies, Nursing facilities, Psychiatric & Community Mental Health Services, and other health facilities points from the HRSA Geospatial Database. These geocoded locations from the HRSA Geospatial Data Warehouse are from a “Provider of Service” extract from the Online Survey and Certification Reporting System database maintained by Centers for Medicare and Medicaid Services. They are included in the HRSA Warehouse because they are the most readily-obtainable data on various classes of health care facility such as hospitals, hospices, rural health clinics, etc.

The Nursing Facility locations provided by HRSA are those facilities participating in Medicare and Medicaid for individuals requiring nursing care and assistance with daily life activities. The Hospitals are those facilities participating in Medicare and Medicaid Services for individuals requiring temporary or long-term medical treatment. The Critical Access Hospitals are those institutions participating in Medicare and Medicaid and meeting the following requirements: being located in rural areas and being located more than 35 miles from any other Hospital or Critical Access Hospital, having no more than 25 inpatient beds and maintaining an average length of stay of 96 hours per patient for acute inpatient care, and providing 24 hour emergency care services.

“Federally Qualified Health Centers (FQHCs)” (often referred to as “Community Health Centers”) receive funding under the Health Center Cluster federal grant program to provide care for underserved populations. The types of providers eligible include Community Health Centers, Migrant Health Centers, Health Care for the Homeless Programs, Public Housing Primary Care Programs, and care providers for some tribal organizations.

On PolicyMap there is also a dataset called “Community Health Centers and Look-Alikes”, which PolicyMap downloads from the Health Resources and Services Administration (HRSA) website. This includes those receiving grants and community health centers that are eligible but not currently receiving grant funding. Although they are not receiving grants, these providers – or “look-alikes” – are eligible for some benefits including enhanced reimbursement from Medicare and Medicaid. Mapping both FQHCs and “look-alikes” might provide a fuller picture of the health-care safety net in a community.

PolicyMap joins individual health center performance data to the Community Health Centers and Look-Alikes point dataset. HRSA tracks this performance data via the Uniform Data System (UDS), which is a reporting requirement for grantees of the following HRSA primary care programs, as defined in the Public Health Service Act: Community Health Center, Migrant Health Center, Health Care for the Homeless, Public Housing Primary Care. Because the UDS data is self-reported, some performance data did not match up with the Community Health Centers and Look-Alikes dataset, resulting in a 95.8% match. For information about health center performance data, see: https://bphc.hrsa.gov/datareporting/index.html.

Medically Underserved Areas (MUAs) are census tracts designated by the Health Resources and Services Administration as having too few primary care providers, high infant mortality, high poverty, and/or high elderly population. See: http://muafind.hrsa.gov/. Medically Underserved Populations (MUP) are areas where a specific population group is underserved, including groups with economic, cultural, or linguistic barriers to primary medical care. If a population group does not meet the criteria for an MUP, but exceptional conditions exist which are a barrier to health services, they can be designated with a recommendation from the state’s Governor.

Due to what HRSA terms a “source data error”, some areas have multiple designations. In these instances, if any designation is MUA, MUA is shown on the map. If MUP and Governor are designated for a single area, MUP is shown. If multiple IMU scores are provided, the lowest score is shown on the map. Although MUA and MUP data is shown on PolicyMap at the tract level, it is provided by HRSA at the tract, county, and minor civil division (MCD). County and MCD level data is shown at the tract level. In cases where a tract was only partially covered by an applicable MCD, it was labeled as not being an MUA or MUP.

The data layers on PolicyMap related to health care professions, health facilities, and hospital utilization are all from HRSA’s Area Health Resource File (AHRF). This data is compiled by HRSA from multiple original data sources, which means that different indicators are available for different years. PolicyMap calculated all the rates in this dataset using the Census’s population estimates for the appropriate year. For more information about the AHRF, see: https://data.hrsa.gov/data/about.

Health Professional Shortage Areas (HPSAs) are defined by the Health Resources and Services Administration (HRSA) as areas that need more health providers in primary care, dental health, or mental health. All HPSAs are defined on the basis of three basic criteria: the ratio of population to health providers, percent of population below the federal poverty level, and travel time to the nearest source of care outside the HPSA area. Dental HPSAs also consider an area’s water flouridation status. Mental HPSAs also consider substance and alcohol abuse prevalence, and percentage of the population over the age of 65 or under the age of 18. Primary Care HPSAs also consider infant mortality rate and low birth weight rate. HPSAs may be designated as “geographic”, “population”, or “point” shortage areas. A geographic HPSA is an area where all residents may experience a shortage of providers. A population HPSA is an area where a specific group of people may experience a shortage of providers. A point HPSA is a specific facility, clinic, or health center that may experience a shortage of providers. PolicyMap excludes point HPSAs as they may not be reflective of the larger population or geographic area. If a census tract is designated as a HPSA more than once, the status and scores for the most recently updated HPSA are shown. HRSA updates HPSA status frequently, so there may be a lag between the data shown on PolicyMap and the latest designation from HRSA. Go to https://data.hrsa.gov/tools/shortage-area to check whether a site of interest lies within a HPSA.

Maternity Care Health Professional Target Areas (MCTAs) are areas within an existing Primary Care Health Professional Shortage Areas (HPSA) that are experiencing a shortage of maternity health care professionals. Maternity Care Target Areas can receive a score between 0-25. What goes into the MCTA score: Population-to-Full-Time-Equivalent Maternity Care Health Professional Ratio [5 points max], Percentage of Population With Income at or Below 200 Percent of the Federal Poverty Level (FPL) [5 points max], Travel Distance/Time to Nearest Source of Accessible Care Outside of the MCTA [5 points max], Fertility Rate [2 points max], Social Vulnerability [2 points max], Maternal Health Indicators, Pre-Pregnancy Obesity [1 point max], Pre-Pregnancy Diabetes [1 point max], Pre-Pregnancy Hypertension [1 point max], Cigarette Smoking [1 point max], Prenatal Care Initiation in the 1st Trimester [1 point max], Behavioral Health Factor [1 point max]

Description:

As part of its Affirmatively Furthering Fair Housing (AFFH) initiative, HUD released a series of opportunity indices. To assist PolicyMap users engaged in AFFH efforts, PolicyMap has loaded the following opportunity indices to PolicyMap: low poverty index, labor market engagement index, school proficiency index, low transportation cost index, transit trips index, jobs proximity index and environmental health index. For more information about the component data and methodology that HUD used in creating these indices, please consult their AFFH data documentation here.

Description:

The U.S. Department of Housing and Urban Development (HUD) submits annual reports to Congress about homelessness in the United States. The reports include point-in-time counts of homeless persons on a single night in January based on local community counts. These counts are submitted annually to HUD by local Continuums of Care (CoC) as part of the competitive funding process. HUD updates CoC boundaries annually and re-calculates historical point-in-time counts with the new boundary. Due to this, PolicyMap has chosen to present data for all years at the most current boundary.

Because of pandemic-related disruptions to counts of unsheltered homeless people in January 2021, caution should be taken when using data on unsheltered homeless or comparing these values to other years. For more information see the AHAR Report here: https://www.huduser.gov/portal/sites/default/files/pdf/2021-AHAR-Part-1.pdf.

Description:

The Community Development Block Grant (CDBG) program, which was enacted by Congress in 1974, is intended to fund local government activities benefiting low and moderate income people such as ensuring access to affordable housing, creating jobs through retention and expansion of local businesses, and providing basic services to the most vulnerable residents. CDBG funds are allocated on a formula basis and grantees need to submit a Consolidated Plan showing that the activities funded will benefit in majority low and moderate income people. This dataset includes the primary locations of CDBG activities that have occurred since 1996. Below are the CDBG activity categorizations used by HUD: Asset Acquisition – activity related to acquisition, including disposition, clearance and demolition, and clean-up of contaminated Sites/brownfields. Economic Development – activity related to economic development, including commercial or industrial rehab, commercial or industrial land acquisition, commercial or industrial construction, commercial or industrial infrastructure development, direct assistance to businesses, and micro-enterprise assistance. Housing – activity related to housing, including multifamily rehab, housing services, code enforcement, operation and repair of foreclosed property and public housing modernization. Public Improvements – activity related to public improvements, including senior centers, youth centers, parks, street improvements, water/sewer improvements, child care centers, fire stations, health centers, non-residential historic preservation, etc. Public Services – activity related to public services, including senior services, legal services, youth services, employment training, health services, homebuyer counseling, food banks, etc. Other – activity related to urban renewal completion, non-profit organization capacity building, and assistance to institutions of higher education.

HUD provided valid geocoding for 99% of the activity locations, and points with identical activity information have been removed from this dataset. To learn more about the CDBG program, please see the following HUD website: https://hub.arcgis.com/datasets/HUD::community-development-block-grant-activity/about.

Description:

The Community Development Block Grant (CDBG) program, which was enacted by Congress in 1974, is intended to fund local government activities benefiting low and moderate income people such as ensuring access to affordable housing, creating jobs through retention and expansion of local businesses, and providing basic services to the most vulnerable residents. CDBG funds are allocated on a formula basis and grantees need to submit a Consolidated Plan showing that the activities funded will benefit in majority low and moderate income people. In order to be eligible for CDBG funds on an area basis, at least 51% of the activity’s beneficiaries must be of low and moderate income. The U.S. Department of Housing and Urban Development (HUD) defines as low and moderate income all individuals living in a household with income below 80% of the area median family income. Through FY 2013, HUD published a summary of the number of people who are low and moderate income at the split block group level. During this time, HUD based its figures on the decennial Census. In FY 2014, HUD started basing its figures on the Census’ American Community Survey (2006-2010). Because ACS has a smaller sample and is less precise than the decennial Census, for FY 2014, HUD published a summary of the number of people who are low and moderate income at the block group, but not the split block group, level. For FY 2023, HUD based its figures on the 2016-2020 ACS. In addition, HUD publishes every year a list of “exception grantees,” which are areas with smaller overall concentrations of low and moderate income people. In these areas, the block groups in the highest quartile in terms of concentration of persons of low and moderate income are deemed eligible. The list of exception grantees can be found at: https://www.hudexchange.info/manage-a-program/acs-low-mod-summary-data-exception-grantees/. Finally, HUD publishes every year a list of block groups in “uncapped areas.” These are metropolitan areas that are allowed to use different income limits – as specified by HUD – in their Consolidated Plans. To find out if you are an “uncapped” area and what your special income limits are please visit the HUD website at: https://www.hudexchange.info/manage-a-program/acs-low-mod-summary-data-uncapped-grantees/.

PolicyMap displays the number and percentage of persons with medium income (household income below 120% AMI), low and moderate income (household income below 80% AMI), and low income (household income below 50% AMI) at the block group, census tract, place, and county geographies. This data is displayed with the most recent ACS estimates used in LMISD data, not the most recent ACS release.

Description:

The HUD Initiative for Renewal Communities, Empowerment Zones, and Enterprise Communities (RC/EZ/EC) provided federal grants, tax incentives, and partnerships with government, for-profit, and non-profit entities to promote job creation and economic development in economically distressed communities. After the original authorization by the Omnibus Reconciliation Act of 1993, which designated 9 empowerment zones and 95 enterprise communities, there were three rounds of community applications, culminating in extension of the benefits provided through the Community Renewal Tax Relief Act of 2000. Ultimately, 19 Enterprise, 67 Renewal, and 44 Empowerment communities were designated or extended between 1994 and 2011.

Description:

Fair Market Rent (FMR) is established by the Department of Housing and Urban Development (HUD) for each fiscal year. FMR is used primarily to determine payment standard amounts for Federal housing assistance programs. FMR is a gross rent estimate and includes the shelter rent, plus the cost of all tenant-paid utilities, except telephones, cable or satellite television service, and internet service. The levels at which FMR is set is expressed as a percentile point within the rent distribution of standard-quality rental housing units. The most recent FMR reflects the estimated 40th and 50th percentile rent levels, meaning that 40% or 50% of rental units can be rented at or below this threshold. To calculate FMRs, HUD uses the U.S. Ceneus Bureau’s most recent 5-year ACS data as “base rents” for calculations. For more information on the methodology used by HUD, please see: https://www.huduser.gov/portal/datasets/fmr.html

Small Area FMRs (SAFMRs) are FMRs calcualted at the Zip Code level. HUD calculates SAFMR using standard quality gross rent data provided by the Census Bureau for ZIP Code Tabulation Areas (ZCTAs) when ZCTA data is statistically reliable. When ZCTA data is not reliable, HUD uses the Rent Ratio method; a calculation method where HUD divides the median gross rent across all bedrooms for the ZCTA by the similar median gross rent for the metropolitan area of the ZCTA. If the all bedroom level data is not reliable, HUD may use neighboring ZCTA data to calculate this data. For more information on this methodology, please see: https://www.huduser.gov/portal/datasets/fmr/smallarea/index.html#year2026 Prior to FY2025, HUD only published SAFMR data for metro area zip codes. Therefore, zip codes in non-metro areas prior to 2025 have been published with the “Insufficient Data” label. From 2025 on, data is available for both metro and non-metro area zip codes.

Description:

The US Department of Housing and Urban Development (HUD) releases annual funding allocations for the programs administered by the Office of Community Planning and Development (CPD). These programs include: Community Development Block Grants (CDBG); HOME Investment Partnerships (HOME); Housing Opportunities for Persons with AIDS (HOPWA); and Emergency Shelter Grants (ESG).

These datasets are mapped according to CPD’s custom boundaries for Formula Allocations. Data are mapped to the most recent boundaries available. See: https://hudgis-hud.opendata.arcgis.com/

Description:

The Housing Choice Voucher Marketing Opportunity Index is an index for every Census tract and block group to identify the area’s potential opportunity for Housing Choice Voucher holders seeking housing. It is a measure of neighborhoods’ high quality housing and neighborhood conditions. A higher number indicates a higher potential opportunity for HCV holders seeking housing. It can be used by Public Housing Authorities to help voucher holders find neighborhoods that have low poverty rates, available rental units at or below Fair Market Rent limits, a high level of employment and educational opportunities, and a low density of households who receive housing assistance. The calculation through which the index is calculated is available in a PDF document at the link provided. The calculation through which the index is calculated is available in a PDF document downloadable from the page at the link provided above.

Description:

The Department of Housing and Urban Development (HUD) established Area Median Incomes (AMI) for households of various sizes, which are used to determine eligibility for HUD’s assisted housing programs, including Public Housing, Section 8 Housing Assistance Payments program, Section 202 housing for the elderly, and Section 811 housing for persons with disabilities.

Many non-federal and non-housing programs also use HUD’s income guidelines, often specifying a percentage of the median income that a household’s income must fall below in order to qualify. PolicyMap includes AMI at a variety of percentages for a variety of household sizes. The 30%, 50% (Very Low Income), and 80% (Low Income) of median income by family size as well as the overall area median income are provided by HUD. PolicyMap calculated 60% of Area Median Income by multiplying the 50% threshold by 1.2 and calculated 120% of AMI by multiplying the 50% threshold by 2.4, per instructions in the LIHTC legislation, on HUD’s website, and in communications between PolicyMap and the HUD User electronic help desk resource. The income thresholds as they are calculated in PolicyMap may not be appropriate for your needs if your programs or requirements specify a different method for determining income thresholds. In particular, the Housing and Economic Recovery Act of 2008 (HERA) specifies different Income Limits for qualification levels and rental rates under section 42 of the Internal Revenue Code and projects financed with tax-exempt housing bonds under section 142 of the Code. Projects in service in 2007 or 2008 should rely on the Multifamily Tax Subsidy Income Limits (MTSP). See: http://www.huduser.org/portal/datasets/mtsp.html.

Description:

The Location Affordability Index is a calculation designed by the Department of Housing and Urban Development (HUD) in order to estimate the costs of housing and transportation across the country. Data is provided on the percentage of household income spent on housing, transportation, and the combination of the two. Separate values are also offered for owner and renter households. Annual cost spent on transportation and annual total public transit trips are also included. HUD’s methodology was significantly updated in Version 3 of the Location Affordability Index Model (LAIM), which was released in March of 2019. The biggest difference between Version 2 and 3 of the LAIM was the reduction in geographies available – Version 3 only modelled results for Census Tracts. HUD uses different calculations for eight different types of households, with different sizes, incomes, and number of commuters. They are defined as follows:

A very thorough explanation of the calculations made by HUD are available here: https://www.hudexchange.info/programs/location-affordability-index/.

Description:

PolicyMap downloaded the properties listed in HUD’s LIHTC Database in June 2025. HUD provided valid geocoding for 96% of projects. The LIHTC program was created by the Tax Reform Act of 1986, and gives state and local LIHTC allocating agencies authority to issue tax credits for acquisition, rehabilitation or new construction of low income rental housing.

Description:

The Department of Housing and Urban Development (HUD) conducts a periodic survey of all households living in HUD-subsidized housing. HUD compiles this information into a series of reports called A Picture of Subsidized Households where household data are aggregated by program at various geographies including tract, county, place, and state. The programs in this report include public housing, Housing Choice Vouchers, Mod Rehab, Project Based Section 8, RentSup/RAP, Section 236/BMIR, and Section 202 and 811 Supportive Housing programs. Point-level data on HUD’s multifamily and public housing sites are available on PolicyMap. PolicyMap downloaded and geocoded data on multifamily and public housing sites from three different HUD resources. Wherever possible, PolicyMap linked the data using the property ID. The three datasets included are the Multifamily Assistance and Section 8 Contracts, A Picture of Subsidized Households, and the REAC assessment scores report. All points are geocoded by HUD; where coordinates were not available, the point was not included on the map. 99.99% of public housing properties and 92% of multifamily properties contain coordinates and appear on the map. A Picture of Subsidized Households data is provided by HUD at the contract level, not property level. Multifamily properties may contain multiple contracts. For PolicyMap, contracts are aggregated together to create single values for each property. All contracts available in the Picture data are included, regardless of the contracts listed in the Multifamily Assistance database. Only properties listed in the HUD Multifamily database are included.

State, county, place, and tract level data from Picture of Subsidized Households are aggregated for all HUD subsidy programs, and are also available for the Housing Choice Voucher recipients and for public housing residents. Tract level data for all subsidized households is only available from 2014 on, due to missing data from the source. Data at the tract level for HCV and public housing data is available for all years. Percent calculations against the general population were made by PolicyMap by using data from the American Community Survey.

Description:

Qualified Census Tracts: PolicyMap downloads data on Low-Income Housing Tax Credit (LIHTC) Qualified Census Tracts (QCT) from tables at HUD’s website. A Qualified Census Tract is any census tract (or equivalent geographic area defined by the Bureau of the Census) in which at least 50 percent of households have an income less than 60 percent of the Area Median Gross Income (AMGI). There is a limit on the number of Qualified Census Tracts in any Metropolitan Statistical Area (MSA) or Primary Metropolitan Statistical Area (PMSA) that may be designated to receive an increase in eligible basis: all of the designated census tracts within a given MSA/PMSA may not together contain more than 20 percent of the total population of the MSA/PMSA. For purposes of HUD designations of Qualified Census Tracts, all non-metropolitan areas in a state are treated as if they constituted a single metropolitan area. Difficult Development Areas: PolicyMap downloads data on Difficult Development Areas (DDA) from tables at HUD’s website. A Difficult Development Area is any area designated by the Secretary of HUD as an area that has high construction, land, and utility costs relative to the Area Median Gross Income (AMGI). All designated Difficult Development Areas in Metropolitan Statistical Areas (MSA) or Primary Metropolitan Statistical Areas (PMSA) may not contain more than 20 percent of the aggregate population of all MSAs/PMSAs, and all designated areas not in metropolitan areas may not contain more than 20 percent of the aggregate population of all non-metropolitan counties.

Description:

This dataset, provided by the US Department of Housing and Urban Development (HUD), identifies census tracts that meet or exceed HUD’s established thresholds for racially and ethnically-concentrated areas of poverty (R/ECAPs). These tracts have a non-white population that is greater than or equal to 50% and meet either of the following poverty criteria: the poverty rate of a tract is 1) higher than 40% or 2) more than three times the average poverty rate of tracts in the metropolitan area. The racial/ethnic threshold is lowered to 20% for tracts that are located outside of metropolitan/micropolitan areas. HUD used component data from the decennial census and the American Community Survey to determine which geographies met these criteria in 2010 and 2020, and component data from Brown Longitudinal Tract Database, based on decennial census data from 2000 and 1990, to determine R/ECAP geographies for 1990 and 2000. Data from 1990-2010 are at the 2010 census tract boundaries. Data from 2020 is at the 2020 census tract boundaries.

Description:

Renewal Communities(RC), Empowerment Zones(EZ) and Enterprise Communities(EC) are part of a federally funded community renewal initiative to revitalize distressed urban and rural areas. Businesses located within these three designations are eligible for specific benefits. The Renewal Community tax incentives are worth approximately $5.6 billion to eligible businesses of all sizes in Renewal Communities. These incentives encourage businesses to open, expand, and to hire local residents. The incentives include employment credits, a 0% tax on capital gains, accelerated depreciation through Commercial Revitalization Deductions, and other incentives. See http://www.hud.gov/utilities/intercept.cfm?/offices/cpd/economicdevelopment/library/taxincentivesrc.pdf for complete details. The Empowerment Zone tax incentives are worth approximately $5.3 billion to small and large businesses in Empowerment Zones. These incentives encourage businesses to open and expand and to hire local residents. Empowerment Zone incentives include employment credits, low-interest loans through EZ facility bonds, reduced taxation on capital gains, and other incentives. See http://www.hud.gov/utilities/intercept.cfm?/offices/cpd/economicdevelopment/library/taxincentivesez.pdf for complete details. HUD does not provide a detailed description of Enterprise Communities.

Incentive zones that were authorized before 2000 were specified in terms of 1990 Census Tracts. In PolicyMap it is only possible to display shading for 2000 Census Tracts. If 75% or more of the area of a 2000 Census Tract was deemed an Empowerment Zone, Renewal Community, or Enterprise Community in 1990 (according to the overlap of the 1990 boundary file), then that Census Tract is designated to be of that Zone or Community in PolicyMap.

Description:

The Multifamily Assistance and Section 8 Contracts Database was created to provide HUD partners/clients with a way of measuring the potential impact of expiring project-based subsidy contracts in their communities. PolicyMap linked the HUD Multifamily Assistance Properties to the HUD Multifamily Assistance and Section 8 Contracts to show the details for up to four contracts per property. The most recent four contracts are shown; if a property has more than four contracts, the number of additional older contracts is displayed. PolicyMap downloaded and geocoded data on HUD’s multifamily sites from three different resources at HUD and, wherever possible, PolicyMap linked the data using the property ID. The three datasets included are the Multifamily Assistance and Section 8 Contracts, A Picture of Subsidized Households and the REAC assessment scores report. PolicyMap was able to locate 92% of multifamily properties on a map.

This data is displayed with HUD’s A Picture of Subsidized Households, which shows data about the subsidized households at these properties. Picture data is provided by HUD at the contract level, not property level. For PolicyMap, contracts are aggregated together to create single values for each property. All contracts available in the Picture data are included, regardless of the contracts listed in the Multifamily Assistance database. Only properties listed in the HUD Multifamily database are included.

Description:

HUD’s Real Estate Assessment Center (REAC) conducts physical property inspections of roughly 20,000 properties owned, insured or subsidized by HUD per year. The purpose of these inspections is to make sure that assisted families are provided with decent, safe and sanitary housing in good repair. Inspected properties receive an overall score from 0 to 100 based on five criteria: site, building exterior, building systems, common areas and units. Multifamily inspections note whether a site has health and safety violations or smoke detector issues. The REAC score received by the property sets its “standard” and dictates how often HUD will return to evaluate the property:

• A score of 90 points or higher is designated a Standard 1 and required to undergo a physical inspection once every three years.
• A score between 80 and 89 is designated a Standard 2 and required to undergo a physical inspection once every two years.
• A score of less than 80 points is designated a Standard 3 and required to undergo an annual physical inspection.

PolicyMap downloads geocoded data on HUD’s multifamily and public housing sites from three different resources at HUD and wherever possible PolicyMap linked the data using the property ID. The three HUD housing datasets included are the Multifamily Assistance and Section 8 Contracts, A Picture of Subsidized Households and the REAC assessment scores report. All points are geocoded by HUD; where coordinates were not available, the point was not included on the map. In 2025, there was no update for the public housing inspection scores. The latest data available from HUD (2023) was used.

Description:

The Museum Universe Data file contains the locations and basic information of known museums and related organizations. Museums include aquariums, arboretums, botanical gardens, art museums, children’s museums, general museums, historic houses and sites, history museums, nature centers, natural history and anthropology museums, planetariums, science and technology centers, specialized museums, and zoological parks. IMLS compiled this information from federal tax records, private foundations, and third-party commercial vendors. Points were geocoded by IMLS.

Description:

The Public Libraries Survey (PLS) is conducted annually by the IMLS. The data file includes all public libraries and outlets identified by state library agencies in the 50 States, the District of Columbia, and the outlying areas of Guam, the Northern Mariana Islands, Puerto Rico and the U.S. Virgin Islands.

Library outlet locations include central libraries, branches, bookmobiles, and books-by-mail locations. Points were geocoded by IMLS based on addresses provided by the survey respondent (library administrator), and in some cases were matched to the center point of the postal zip code or zip code division. IMLS was able to locate 99.9% of library outlets on a map.

The reporting period varies among localities for states; however, each public library provided data for a 12-month period. When calculating weekly hours, PolicyMap applied IMLS’ methodology–suppressing values where average hours per week is less than 11 or greater than 130. Refer to the Data File Documentation and User’s Guide for more details.

Description:

The International Business Innovation Association is a trade group serving over 2,100 business incubators and related organizations worldwide. Business incubators are programs that provide support services and resources for entrepreneurial companies during their “start-up” phase. InBIA provided PolicyMap with a list of business incubators in the United States.

Description:

The Internal Revenue Service’s Statistics of Income (IRS SOI) division produces annual estimates of migration flows using domestic and foreign tax returns. Returns are matched to returns from the previous year using the primary tax filer’s social security number. If the address associated with the return is located in a different county than the previous year, the return is identified as a migrant. In-migrants are those who filed tax returns in different counties, states, or abroad in the previous year but filed locally in the current year; out-migrants filed taxes locally in the previous year but filed in different counties, states, or abroad for the current year. Individual taxpayers cannot be identified, either directly or indirectly, from these tabulations. The data released by the IRS for these calculations have undergone suppression procedures to ensure no inappropriate disclosure of information. There are two limitations to the completeness of that data that should be considered when using IRS migration data. First, up to the 2009-2010 time period, the data only capture returns processed by late September of the year following the tax year, which covers roughly 95-98 percent of all returns. Second, those who are not required to file tax returns are not counted by the IRS, so lower income people and senior citizens are likely underrepresented. Beginning with 2010-2011, the IRS began using full-year tax return instead of partial-year. Overall, this increases the number of returns by 5 percent, and high-income returns by approximately 25 percent. Caution should be exercised in making comparisons across this time period. More information is available here: https://www.irs.gov/pub/irs-soi/soi-a-inmig-id1509.pdf.

Data from the IRS is given for the year of the tax filing; people generally file their returns the year following the tax year (tax year 2013 returns are filed in 2014). PolicyMap displays the tax year, which is the year the migration occurred, not the filing year.

Description:

Researchers at the Johns Hopkins Bloomberg School of Public Health and the University of Maryland created a statistical model of how likely an adult (age 18+) is to die of COVID-19 as compared to the national average, based on demographic information and underlying health conditions. The researchers developed the model based on two sources of information: (1) multivariate-adjusted estimate of risk associated with gender, social deprivation index and 12 pre-existing conditions from the recently published UK-based OpenSAFELY study, and (2) death rates associated with different age and racial/ethnic groups in the US published by the Centers for Disease Control and Prevention (CDC) after performing external covariate adjustment accounting for the correlation of these factors with other risk factors in the model. The researchers applied their statistical model to publicly available demographic and health data to create (1) an Index of Excess Risk (IER) which describes how likely the average adult in a city (Census place) or state is to die from COVID-19 compared to the national average, (2) estimated proportion and size of the general adult population in a city or state that exceed different risk thresholds, and (3) estimated proportion of the projected deaths among general adult population in a city or state that exceed different risk thresholds. Validation analyses were conducted based on the projected risks and data on recent deaths in the US to show that the model is well calibrated for the US population. They also applied the model to CMS data for the Medicare population age 65 and older to arrive at county- and state-level estimates.

The researchers also published a calculator that assesses individual risk of death from COVID-19 here: http://covid19risktools.com/.

Description:

The Common Core of Data (CCD) is a program of the U.S. Department of Education’s National Center for Education Statistics that collects selected data about all public schools, public school districts and state education agencies in the United States every year. Data are supplied by state education agency officials through the Common Core of Data (CCD), Local Education Agency (School District) Universe Survey, accessed from https://nces.ed.gov/ccd/files.asp. For schools that do not report free and reduced-price lunch program (FRPL) eligibility data data, direct certification counts are used exclusively. Direct certification applies to children from households participating in the Supplemental Nutrition Assistance Program (SNAP), Temporary Assistance for Needy Families (TANF), the Food Distribution Program on Indian Reservations (FDPIR), or (in some states) Medicaid, as well as children who are migrant, experiencing homelessness, in foster care, or enrolled in Head Start. These students are categorically eligible to receive free meals at school. Note that NCES CCD data may indicate participation rates above 100% in some schools due to reporting variations or administrative discrepancies. Eligibility data is suppressed when fewer than 6 students are present to protect confidentiality.

Data Note: Illinois and Virginia opted not to collect this data under the pandemic meal provisions that gave free meals universally, thus leading to gaps in the usual reporting for these states in the 2020-21 period. Additionally, in the August 2025 data update, PolicyMap recalculated and reloaded historic percent data for the 2019-2020 school year.

Description:

The Integrated Postsecondary Education Data System (IPEDS) provides data on colleges, universities, and technical and vocational postsecondary institutions in the United States.

The point-level community college locations data on PolicyMap consist of all the institutions in the IPEDS’ universe classified as 2-but-less-than-4-year or less than 2-year. The 4-year colleges and universities data consist of all institutions in the IPEDS’ universe classified as four or more years. Most of the data in these datasets is for the 2016-2017 academic year (with enrollment data applying to Fall of 2016). Graduation rates, however, apply to August of 2016. Data regarding institutions’ financial performance reflect Government Accounting Standards Board (GASB) reporting. PolicyMap downloaded these data from the IPEDS data center in August 2018. The latitude and longitude of the points were provided by the source.

Description:

The Private School Universe Survey (PSS) is a program of the U.S. Department of Education’s National Center for Education Statistics that collects selected data about all private schools every two years. Data are supplied through a survey that is compiled by administrative personnel in private schools. The NCES definition of a private school is “A private school is not supported primarily by public funds, provides classroom instruction for one or more of grades K-12 or comparable ungraded levels, and has one or more teachers. Organizations or institutions that provide support for home schooling without offering classroom instruction for students are not included.”

Description:

The National Credit Union Administration (NCUA) provides data for all U.S. credit unions as well as the addresses of credit union branches. These data are from the 5300 Call Report, submitted quarterly by credit unions to the NCUA, and downloaded in April 2025 from https://www.ncua.gov/analysis/Pages/call-report-data/quarterly-data.aspx. All information in this data, aside from office type and contact information, applies to the credit union and not the individual branch. Financial counseling/education, online banking, and percent loans delinquent were calculated by PolicyMap. PolicyMap geocoded all branch location points, and was able to locate 95% of the given addresses on a map.

Description:

Technical Preservation Services, a division of the National Park Service, provided PolicyMap with a list of all approved Federal Historic Preservation Tax Incentives Program Part 3 applications from fiscal year 2001 through fiscal year 2022. Part 3 application approvals represent completed and certified projects eligible for the 20% federal tax credit for the rehabilitation of a historic building. The tax credit applies to qualifying costs associated with capital improvements when undertaking a substantial rehabilitation consistent with the historic character of a certified historic building. Project Costs in this data represent the estimated rehabilitation costs, or qualified rehabilitation expenditures (QREs) submitted to the National Park Service on the Part 3 application. The indicated date is for certification of the project by the National Park Service, and is not necessarily when the construction project was completed. The data represent tax credit project approvals since October 1, 2001, through September 30th of the most recently available fiscal year. Street addresses are those submitted to the National Park Service on the Part 3 applications; PolicyMap was able to locate 96% of the addresses on a map. Data are updated annually.

The Federal Historic Preservation Tax Incentives Program is administered by the National Park Service, State Historic Preservation Offices, and the Internal Revenue Service. For more information on Federal Historic Tax Credits visit the National Park Service: http://www.nps.gov/tps/index.htm. For additional information visit the National Trust Community Investment Corporation: https://ntcic.com/invest/htc/

Description:

A team at the New York Times has created a national database of COVID-19 cases and deaths based on local reporting and has made the database available to the public. This includes all “confirmed” cases where a person has tested positive for COVID-19. Since many who were exposed to COVID-19 have not been tested, reported numbers may underrepresent the prevalence of the virus. This data is reported at the state and county levels. The New York Times has modified the county level data in several important ways. The five counties that comprise New York City have been grouped into a single entity called “New York City.” Kansas City, Missouri is reported separately from the counties that it intersects. Some cruise ships or aircraft carriers with infected passengers or crew have been counted according to the county where the vessel has docked or where the affected people were treated. See https://github.com/nytimes/covid-19-data/blob/master/README.md for specific information on geographic exceptions.

PolicyMap calculated the percent change in cases and deaths and rates per 100,000 people. Rates were calculated using the 2014-2018 population estimates published by the U.S. Census Bureau. The populations of Platte, Clay, Jackson, and Cass counties in Missouri excluding the portion of the population that lives within Kansas City were estimated using block group data. This dataset is updated daily by the New York Times, and downloaded and published frequently by PolicyMap.

Description:

PolicyMap calculated local median income as a share of area median income using American Community Survey (ACS) estimates of median household income and median family income. For all tracts and block groups located within Census-defined metropolitan areas, this calculation is local median income as a share of metro-area median income. For tracts and block groups outside of Census-defined metro areas, this is local median income as a share of the non-metro state median income. PolicyMap has chosen specific breaks in the data, following on commonly used and understood guidelines: ≤30% of area median income >30% and ≤50% of area median income >50% and ≤80% of area median income >80% and ≤120% of area median income >120% of area median income However, PolicyMap subscribers can edit the breaks in the ranges if some other set of breaks is preferred.

Please be aware that the thresholds and data sources used in this calculation can vary, and federal agencies may require specific calculations for some program applications. The Community Development Block Grant (CDBG) program defines low and moderate income tracts based on what percent of the population is low or moderate income, rather than by comparing median local values to the surrounding metro area (See HUD Community Development Block Grant Eligibility Criteria above). The Community Reinvestment Act (CRA) specifies what years of income data to include in the calculation – 2000 data for local median income and 2004 data for area median income. (See Community Reinvestment Act Eligibility Criteria above). Both CDBG and CRA low and moderate income calculations can be found on PolicyMap under the Federal Guidelines tab.