In 2023, Washington, D.C. had the highest population density in the United States, with 11,130.69 people per square mile. As a whole, there were about 94.83 residents per square mile in the U.S., and Alaska was the state with the lowest population density, with 1.29 residents per square mile. The problem of population density Simply put, population density is the population of a country divided by the area of the country. While this can be an interesting measure of how many people live in a country and how large the country is, it does not account for the degree of urbanization, or the share of people who live in urban centers. For example, Russia is the largest country in the world and has a comparatively low population, so its population density is very low. However, much of the country is uninhabited, so cities in Russia are much more densely populated than the rest of the country. Urbanization in the United States While the United States is not very densely populated compared to other countries, its population density has increased significantly over the past few decades. The degree of urbanization has also increased, and well over half of the population lives in urban centers.

VITAL SIGNS INDICATOR Population (LU1)

FULL MEASURE NAME Population estimates

LAST UPDATED October 2019

DESCRIPTION Population is a measurement of the number of residents that live in a given geographical area, be it a neighborhood, city, county or region.

DATA SOURCES U.S Census Bureau: Decennial Census No link available (1960-1990) http://factfinder.census.gov (2000-2010)

California Department of Finance: Population and Housing Estimates Table E-6: County Population Estimates (1961-1969) Table E-4: Population Estimates for Counties and State (1971-1989) Table E-8: Historical Population and Housing Estimates (2001-2018) Table E-5: Population and Housing Estimates (2011-2019) http://www.dof.ca.gov/Forecasting/Demographics/Estimates/

U.S. Census Bureau: Decennial Census - via Longitudinal Tract Database Spatial Structures in the Social Sciences, Brown University Population Estimates (1970 - 2010) http://www.s4.brown.edu/us2010/index.htm

U.S. Census Bureau: American Community Survey 5-Year Population Estimates (2011-2017) http://factfinder.census.gov

U.S. Census Bureau: Intercensal Estimates Estimates of the Intercensal Population of Counties (1970-1979) Intercensal Estimates of the Resident Population (1980-1989) Population Estimates (1990-1999) Annual Estimates of the Population (2000-2009) Annual Estimates of the Population (2010-2017) No link available (1970-1989) http://www.census.gov/popest/data/metro/totals/1990s/tables/MA-99-03b.txt http://www.census.gov/popest/data/historical/2000s/vintage_2009/metro.html https://www.census.gov/data/datasets/time-series/demo/popest/2010s-total-metro-and-micro-statistical-areas.html

CONTACT INFORMATION vitalsigns.info@bayareametro.gov

METHODOLOGY NOTES (across all datasets for this indicator) All legal boundaries and names for Census geography (metropolitan statistical area, county, city, and tract) are as of January 1, 2010, released beginning November 30, 2010, by the U.S. Census Bureau. A Priority Development Area (PDA) is a locally-designated area with frequent transit service, where a jurisdiction has decided to concentrate most of its housing and jobs growth for development in the foreseeable future. PDA boundaries are current as of August 2019. For more information on PDA designation see http://gis.abag.ca.gov/website/PDAShowcase/.

Population estimates for Bay Area counties and cities are from the California Department of Finance, which are as of January 1st of each year. Population estimates for non-Bay Area regions are from the U.S. Census Bureau. Decennial Census years reflect population as of April 1st of each year whereas population estimates for intercensal estimates are as of July 1st of each year. Population estimates for Bay Area tracts are from the decennial Census (1970 -2010) and the American Community Survey (2008-2012 5-year rolling average; 2010-2014 5-year rolling average; 2013-2017 5-year rolling average). Estimates of population density for tracts use gross acres as the denominator.

Population estimates for Bay Area PDAs are from the decennial Census (1970 - 2010) and the American Community Survey (2006-2010 5 year rolling average; 2010-2014 5-year rolling average; 2013-2017 5-year rolling average). Population estimates for PDAs are derived from Census population counts at the tract level for 1970-1990 and at the block group level for 2000-2017. Population from either tracts or block groups are allocated to a PDA using an area ratio. For example, if a quarter of a Census block group lies with in a PDA, a quarter of its population will be allocated to that PDA. Tract-to-PDA and block group-to-PDA area ratios are calculated using gross acres. Estimates of population density for PDAs use gross acres as the denominator.

Annual population estimates for metropolitan areas outside the Bay Area are from the Census and are benchmarked to each decennial Census. The annual estimates in the 1990s were not updated to match the 2000 benchmark.

The following is a list of cities and towns by geographical area: Big Three: San Jose, San Francisco, Oakland Bayside: Alameda, Albany, Atherton, Belmont, Belvedere, Berkeley, Brisbane, Burlingame, Campbell, Colma, Corte Madera, Cupertino, Daly City, East Palo Alto, El Cerrito, Emeryville, Fairfax, Foster City, Fremont, Hayward, Hercules, Hillsborough, Larkspur, Los Altos, Los Altos Hills, Los Gatos, Menlo Park, Mill Valley, Millbrae, Milpitas, Monte Sereno, Mountain View, Newark, Pacifica, Palo Alto, Piedmont, Pinole, Portola Valley, Redwood City, Richmond, Ross, San Anselmo, San Bruno, San Carlos, San Leandro, San Mateo, San Pablo, San Rafael, Santa Clara, Saratoga, Sausalito, South San Francisco, Sunnyvale, Tiburon, Union City, Vallejo, Woodside Inland, Delta and Coastal: American Canyon, Antioch, Benicia, Brentwood, Calistoga, Clayton, Cloverdale, Concord, Cotati, Danville, Dixon, Dublin, Fairfield, Gilroy, Half Moon Bay, Healdsburg, Lafayette, Livermore, Martinez, Moraga, Morgan Hill, Napa, Novato, Oakley, Orinda, Petaluma, Pittsburg, Pleasant Hill, Pleasanton, Rio Vista, Rohnert Park, San Ramon, Santa Rosa, Sebastopol, Sonoma, St. Helena, Suisun City, Vacaville, Walnut Creek, Windsor, Yountville Unincorporated: all unincorporated towns

The Census Bureau released revised delineations for urban areas on December 29, 2022. The new criteria (contained in this Federal Register Notice) is based primarily on housing unit density measured at the census block level. The minimum qualifying threshold for inclusion as an urban area is an area that contains at least 2,000 housing units or has a population of at least 5,000 persons. It also eliminates the classification of areas as “urban clusters/urbanized areas”. This represents a change from 2010, where urban areas were defined as areas consisting of 50,000 people or more and urban clusters consisted of at least 2,500 people but less than 50,000 people with at least 1,500 people living outside of group quarters. Due to the new population thresholds for urban areas, 36 urban clusters in California are no longer considered urban areas, leaving California with 193 urban areas after the new criteria was implemented.

The State of California experienced an increase of 1,885,884 in the total urban population, or 5.3%. However, the total urban area population as a percentage of the California total population went down from 95% to 94.2%. For more information about the mapped data, download the Excel spreadsheet here.

Please note that some of the 2020 urban areas have different names or additional place names as a result of the inclusion of housing unit counts as secondary naming criteria.

Please note there are four urban areas that cross state boundaries in Arizona and Nevada. For 2010, only the parts within California are displayed on the map; however, the population and housing estimates represent the entirety of the urban areas. For 2020, the population and housing unit estimates pertains to the areas within California only.

Data for this web application was derived from the 2010 and 2020 Censuses (2010 and 2020 Census Blocks, 2020 Urban Areas, and Counties) and the 2016-2020 American Community Survey (2010 -Urban Areas) and can be found at data.census.gov.

For more information about the urban area delineations, visit the Census Bureau's Urban and Rural webpage and FAQ.

To view more data from the State of California Department of Finance, visit the Demographic Research Unit Data Hub.

VITAL SIGNS INDICATOR

Population (LU1)

FULL MEASURE NAME

Population estimates

LAST UPDATED

February 2023

DESCRIPTION

Population is a measurement of the number of residents that live in a given geographical area, be it a neighborhood, city, county or region.

DATA SOURCE

California Department of Finance: Population and Housing Estimates - http://www.dof.ca.gov/Forecasting/Demographics/Estimates/

Table E-6: County Population Estimates (1960-1970)

Table E-4: Population Estimates for Counties and State (1970-2021)

Table E-8: Historical Population and Housing Estimates (1990-2010)

Table E-5: Population and Housing Estimates (2010-2021)

Bay Area Jurisdiction Centroids (2020) - https://data.bayareametro.gov/Boundaries/Bay-Area-Jurisdiction-Centroids-2020-/56ar-t6bs

Computed using 2020 US Census TIGER boundaries

U.S. Census Bureau: Decennial Census Population Estimates - http://www.s4.brown.edu/us2010/index.htm- via Longitudinal Tract Database Spatial Structures in the Social Sciences, Brown University

1970-2020

U.S. Census Bureau: American Community Survey (5-year rolling average; tract) - https://data.census.gov/

2011-2021

Form B01003

Priority Development Areas (Plan Bay Area 2050) - https://opendata.mtc.ca.gov/datasets/MTC::priority-development-areas-plan-bay-area-2050/about

CONTACT INFORMATION

vitalsigns.info@bayareametro.gov

METHODOLOGY NOTES (across all datasets for this indicator)

All historical data reported for Census geographies (metropolitan areas, county, city and tract) use current legal boundaries and names. A Priority Development Area (PDA) is a locally-designated area with frequent transit service, where a jurisdiction has decided to concentrate most of its housing and jobs growth for development in the foreseeable future. PDA boundaries are current as of December 2022.

Population estimates for Bay Area counties and cities are from the California Department of Finance, which are as of January 1st of each year. Population estimates for non-Bay Area regions are from the U.S. Census Bureau. Decennial Census years reflect population as of April 1st of each year whereas population estimates for intercensal estimates are as of July 1st of each year. Population estimates for Bay Area tracts are from the decennial Census (1970-2020) and the American Community Survey (2011-2021 5-year rolling average). Estimates of population density for tracts use gross acres as the denominator.

Population estimates for Bay Area tracts and PDAs are from the decennial Census (1970-2020) and the American Community Survey (2011-2021 5-year rolling average). Population estimates for PDAs are allocated from tract-level Census population counts using an area ratio. For example, if a quarter of a Census tract lies with in a PDA, a quarter of its population will be allocated to that PDA. Estimates of population density for PDAs use gross acres as the denominator. Note that the population densities between PDAs reported in previous iterations of Vital Signs are mostly not comparable due to minor differences and an updated set of PDAs (previous iterations reported Plan Bay Area 2040 PDAs, whereas current iterations report Plan Bay Area 2050 PDAs).

The following is a list of cities and towns by geographical area:

Big Three: San Jose, San Francisco, Oakland

Bayside: Alameda, Albany, Atherton, Belmont, Belvedere, Berkeley, Brisbane, Burlingame, Campbell, Colma, Corte Madera, Cupertino, Daly City, East Palo Alto, El Cerrito, Emeryville, Fairfax, Foster City, Fremont, Hayward, Hercules, Hillsborough, Larkspur, Los Altos, Los Altos Hills, Los Gatos, Menlo Park, Mill Valley, Millbrae, Milpitas, Monte Sereno, Mountain View, Newark, Pacifica, Palo Alto, Piedmont, Pinole, Portola Valley, Redwood City, Richmond, Ross, San Anselmo, San Bruno, San Carlos, San Leandro, San Mateo, San Pablo, San Rafael, Santa Clara, Saratoga, Sausalito, South San Francisco, Sunnyvale, Tiburon, Union City, Vallejo, Woodside

Inland, Delta and

Splitgraph serves as an HTTP API that lets you run SQL queries directly on this data to power Web applications. For example:

See the Splitgraph documentation for more information.

Contained within the Atlas of Canada Poster Map Series, is a poster showing population density across Canada. There is a relief base to the map on top of which is shown all populated areas of Canada where the population density is great than 0.4 persons per square kilometer. This area is then divided into five colour classes of population density based on Statistics Canada's census divisions.

This map shows four of these densely populated areas are in California. The San Francisco-Oakland and San Jose Urban Areas rank second and third, respectively. That the New York Metropolitan area ranks fifth on this list shows that this density ranking is greatly affected by the nature of the land area designated as urban. Census Urban Areas comprise an urban core and associated suburbs. California's urban and suburban areas are more uniform in density when compared to New York's urban core and suburban periphery which have vastly different densities. Delano ranks fourth because it has a very small land area and its population is augmented by two large California State Prisons housing 10,000 inmates.

With 3.5 persons per square kilometre, Canada is one of the countries with the lowest population densities in the world. Census metropolitan areas (CMAs) with the highest population densities—Toronto (866), Montréal (854), Vancouver (735), Kitchener (546), Hamilton (505), and Victoria (475)—were located close to United States border.

The Africa Population Distribution Database provides decadal population density data for African administrative units for the period 1960-1990. The databsae was prepared for the United Nations Environment Programme / Global Resource Information Database (UNEP/GRID) project as part of an ongoing effort to improve global, spatially referenced demographic data holdings. The database is useful for a variety of applications including strategic-level agricultural research and applications in the analysis of the human dimensions of global change.

This documentation describes the third version of a database of administrative units and associated population density data for Africa. The first version was compiled for UNEP's Global Desertification Atlas (UNEP, 1997; Deichmann and Eklundh, 1991), while the second version represented an update and expansion of this first product (Deichmann, 1994; WRI, 1995). The current work is also related to National Center for Geographic Information and Analysis (NCGIA) activities to produce a global database of subnational population estimates (Tobler et al., 1995), and an improved database for the Asian continent (Deichmann, 1996). The new version for Africa provides considerably more detail: more than 4700 administrative units, compared to about 800 in the first and 2200 in the second version. In addition, for each of these units a population estimate was compiled for 1960, 70, 80 and 90 which provides an indication of past population dynamics in Africa. Forthcoming are population count data files as download options.

African population density data were compiled from a large number of heterogeneous sources, including official government censuses and estimates/projections derived from yearbooks, gazetteers, area handbooks, and other country studies. The political boundaries template (PONET) of the Digital Chart of the World (DCW) was used delineate national boundaries and coastlines for African countries.

For more information on African population density and administrative boundary data sets, see metadata files at [http://na.unep.net/datasets/datalist.php3] which provide information on file identification, format, spatial data organization, distribution, and metadata reference.

References:

Deichmann, U. 1994. A medium resolution population database for Africa, Database documentation and digital database, National Center for Geographic Information and Analysis, University of California, Santa Barbara.

Deichmann, U. and L. Eklundh. 1991. Global digital datasets for land degradation studies: A GIS approach, GRID Case Study Series No. 4, Global Resource Information Database, United Nations Environment Programme, Nairobi.

UNEP. 1997. World Atlas of Desertification, 2nd Ed., United Nations Environment Programme, Edward Arnold Publishers, London.

WRI. 1995. Africa data sampler, Digital database and documentation, World Resources Institute, Washington, D.C.

Contained within the 3rd Edition (1957) of the Atlas of Canada is a map that shows the density of the Canadian population for 1951. The first map display Western provinces, while the second map concentrates on southern Ontario and the Maritimes. Only the most populous areas are covered. Population density is illustrated by denoting the number of inhabitants per square mile. It shows a significant difference in the population distribution across Canada, mainly in urban and metropolitan areas. The cities with greater inhabitants are clusters within Capital cities, and a even larger concentration south, near the U.S. border, in particular along ocean or inland coastlines.

Canada CA: Population Density: People per Square Km data was reported at 4.350 Person/sq km in 2021. This records an increase from the previous number of 4.239 Person/sq km for 2020. Canada CA: Population Density: People per Square Km data is updated yearly, averaging 3.127 Person/sq km from Dec 1961 (Median) to 2021, with 61 observations. The data reached an all-time high of 4.350 Person/sq km in 2021 and a record low of 2.038 Person/sq km in 1961. Canada CA: Population Density: People per Square Km data remains active status in CEIC and is reported by World Bank. The data is categorized under Global Database’s Canada – Table CA.World Bank.WDI: Population and Urbanization Statistics. Population density is midyear population divided by land area in square kilometers. Population is based on the de facto definition of population, which counts all residents regardless of legal status or citizenship--except for refugees not permanently settled in the country of asylum, who are generally considered part of the population of their country of origin. Land area is a country's total area, excluding area under inland water bodies, national claims to continental shelf, and exclusive economic zones. In most cases the definition of inland water bodies includes major rivers and lakes.;Food and Agriculture Organization and World Bank population estimates.;Weighted average;

The TIGER/Line Files are shapefiles and related database files (.dbf) that are an extract of selected geographic and cartographic information from the U.S. Census Bureau's Master Address File / Topologically Integrated Geographic Encoding and Referencing (MAF/TIGER) Database (MTDB). The MTDB represents a seamless national file with no overlaps or gaps between parts, however, each TIGER/Line File is designed to stand alone as an independent data set, or they can be combined to cover the entire nation. After each decennial census, the Census Bureau delineates urban areas that represent densely developed territory, encompassing residential, commercial, and other nonresidential urban land uses. In general, this territory consists of areas of high population density and urban land use resulting in a representation of the "urban footprint." There are two types of urban areas: urbanized areas (UAs) that contain 50,000 or more people and urban clusters (UCs) that contain at least 2,500 people, but fewer than 50,000 people (except in the U.S. Virgin Islands and Guam which each contain urban clusters with populations greater than 50,000). Each urban area is identified by a 5-character numeric census code that may contain leading zeroes.

The 2015 cartographic boundary shapefiles are simplified representations of selected geographic areas from the U.S. Census Bureau's Master Address File / Topologically Integrated Geographic Encoding and Referencing (MAF/TIGER) Database (MTDB). These boundary files are specifically designed for small-scale thematic mapping. When possible, generalization is performed with the intent to maintain the hierarchical relationships among geographies and to maintain the alignment of geographies within a file set for a given year. Geographic areas may not align with the same areas from another year. Some geographies are available as nation-based files while others are available only as state-based files. The records in this file allow users to map the parts of Urban Areas that overlap a particular county. After each decennial census, the Census Bureau delineates urban areas that represent densely developed territory, encompassing residential, commercial, and other nonresidential urban land uses. In general, this territory consists of areas of high population density and urban land use resulting in a representation of the "urban footprint." There are two types of urban areas: urbanized areas (UAs) that contain 50,000 or more people and urban clusters (UCs) that contain at least 2,500 people, but fewer than 50,000 people (except in the U.S. Virgin Islands and Guam which each contain urban clusters with populations greater than 50,000). Each urban area is identified by a 5-character numeric census code that may contain leading zeroes. The primary legal divisions of most states are termed counties. In Louisiana, these divisions are known as parishes. In Alaska, which has no counties, the equivalent entities are the organized boroughs, city and boroughs, municipalities, and for the unorganized area, census areas. The latter are delineated cooperatively for statistical purposes by the State of Alaska and the Census Bureau. In four states (Maryland, Missouri, Nevada, and Virginia), there are one or more incorporated places that are independent of any county organization and thus constitute primary divisions of their states. These incorporated places are known as independent cities and are treated as equivalent entities for purposes of data presentation. The District of Columbia and Guam have no primary divisions, and each area is considered an equivalent entity for purposes of data presentation. The Census Bureau treats the following entities as equivalents of counties for purposes of data presentation: Municipios in Puerto Rico, Districts and Islands in American Samoa, Municipalities in the Commonwealth of the Northern Mariana Islands, and Islands in the U.S. Virgin Islands. The entire area of the United States, Puerto Rico, and the Island Areas is covered by counties or equivalent entities. The boundaries for counties and equivalent entities are as of January 1, 2010.

The SCAG_ATDB_Healthy shapefile contains Census tract level food access, retail density, park access, tree canopy coverage, and Healthy Places Index (HPI) score data of the SCAG region. Food access data for 2015 (data source: USDA FARA 2017) includes the percentage of the urban population residing less than 1/2 mile from a supermarket/large grocery store, or the percentage of the rural population living less than 1 mile from a supermarket/large grocery store. Retail density data (data source: EPA Smart Location Database 2010) includes the gross retail, entertainment, and education employment density (jobs/acre) on unprotected land. Park access data (data source: HCI/CalLands Database 2010) includes the percentage of population living within a half-mile of a park, open space, or beach. Tree canopy coverage data (data source: HCI/National Land Cover Database 2011) includes population-weighted percentage of census tract area with tree canopy coverage. The HPI score (version: December 2017) is composed of diverse non-medical economic, social, political and environmental factors that influence physical and cognitive function, behavior and disease. These factors are often called health determinants or social determinants of health and form the root causes of health advantage. Indicator data used for HPI comes from publicly available sources and is produced at a census tract level. The HPI score was derived from 8 domain scores, 25 Individual indicators + race/ethnicity percent (8057 CTs). HPI materials will be made freely available online for use by communities and public and private agencies. More info at: http://phasocal.org/ca-hpi/

20 year Projected Urban Growth scenarios. Base year is 2000. Projected year in this dataset is 2020.

By 2020, most forecasters agree, California will be home to between 43 and 46 million residents-up from 35 million today. Beyond 2020 the size of California's population is less certain. Depending on the composition of the population, and future fertility and migration rates, California's 2050 population could be as little as 50 million or as much as 70 million. One hundred years from now, if present trends continue, California could conceivably have as many as 90 million residents.

Where these future residents will live and work is unclear. For most of the 20th Century, two-thirds of Californians have lived south of the Tehachapi Mountains and west of the San Jacinto Mountains-in that part of the state commonly referred to as Southern California. Yet most of coastal Southern California is already highly urbanized, and there is relatively little vacant land available for new development. More recently, slow-growth policies in Northern California and declining developable land supplies in Southern California are squeezing ever more of the state's population growth into the San Joaquin Valley.

How future Californians will occupy the landscape is also unclear. Over the last fifty years, the state's population has grown increasingly urban. Today, nearly 95 percent of Californians live in metropolitan areas, mostly at densities less than ten persons per acre. Recent growth patterns have strongly favored locations near freeways, most of which where built in the 1950s and 1960s. With few new freeways on the planning horizon, how will California's future growth organize itself in space? By national standards, California's large urban areas are already reasonably dense, and economic theory suggests that densities should increase further as California's urban regions continue to grow. In practice, densities have been rising in some urban counties, but falling in others.

These are important issues as California plans its long-term future. Will California have enough land of the appropriate types and in the right locations to accommodate its projected population growth? Will future population growth consume ever-greater amounts of irreplaceable resource lands and habitat? Will jobs continue decentralizing, pushing out the boundaries of metropolitan areas? Will development densities be sufficient to support mass transit, or will future Californians be stuck in perpetual gridlock? Will urban and resort and recreational growth in the Sierra Nevada and Trinity Mountain regions lead to the over-fragmentation of precious natural habitat? How much water will be needed by California's future industries, farms, and residents, and where will that water be stored? Where should future highway, transit, and high-speed rail facilities and rights-of-way be located? Most of all, how much will all this growth cost, both economically, and in terms of changes in California's quality of life?

Clearly, the more precise our current understanding of how and where California is likely to grow, the sooner and more inexpensively appropriate lands can be acquired for purposes of conservation, recreation, and future facility siting. Similarly, the more clearly future urbanization patterns can be anticipated, the greater our collective ability to undertake sound city, metropolitan, rural, and bioregional planning.

Consider two scenarios for the year 2100. In the first, California's population would grow to 80 million persons and would occupy the landscape at an average density of eight persons per acre, the current statewide urban average. Under this scenario, and assuming that 10% percent of California's future population growth would occur through infill-that is, on existing urban land-California's expanding urban population would consume an additional 5.06 million acres of currently undeveloped land. As an alternative, assume the share of infill development were increased to 30%, and that new population were accommodated at a density of about 12 persons per acre-which is the current average density of the City of Los Angeles. Under this second scenario, California's urban population would consume an additional 2.6 million acres of currently undeveloped land. While both scenarios accommodate the same amount of population growth and generate large increments of additional urban development-indeed, some might say even the second scenario allows far too much growth and development-the second scenario is far kinder to California's unique natural landscape.

This report presents the results of a series of baseline population and urban growth projections for California's 38 urban counties through the year 2100. Presented in map and table form, these projections are based on extrapolations of current population trends and recent urban development trends. The next section, titled Approach, outlines the methodology and data used to develop the various projections. The following section, Baseline Scenario, reviews the projections themselves. A final section, entitled Baseline Impacts, quantitatively assesses the impacts of the baseline projections on wetland, hillside, farmland and habitat loss.

Canada, with 3.33 people per square kilometre, has one of the lowest population densities in the world. In 2001, most of Canada's population of 30,007,094 lived within 200 kilometres of the United States (along Canada's south). In fact, the inhabitants of our three biggest cities -- Toronto, Montréal and Vancouver -- can drive to the border in less than two hours. Thousands of kilometres to the north, our polar region -- the Yukon, the Northwest Territories and Nunavut -- is relatively empty, embracing 41% of our land mass but only 0.3% of our population. An inset map shows in greater detail the Windsor-Québec Corridor where a high concentration of Canadians live.

Source

This is the dataset used in this book: https://github.com/ageron/handson-ml/tree/master/datasets/housing to illustrate a sample end-to-end ML project workflow (pipeline). This is a great book - I highly recommend!

The data is based on California Census in 1990.

About the Data (from the book):

"This dataset is a modified version of the California Housing dataset available from Luís Torgo's page (University of Porto). Luís Torgo obtained it from the StatLib repository (which is closed now). The dataset may also be downloaded from StatLib mirrors.

The following is the description from the book author:

This dataset appeared in a 1997 paper titled Sparse Spatial Autoregressions by Pace, R. Kelley and Ronald Barry, published in the Statistics and Probability Letters journal. They built it using the 1990 California census data. It contains one row per census block group. A block group is the smallest geographical unit for which the U.S. Census Bureau publishes sample data (a block group typically has a population of 600 to 3,000 people).

The dataset in this directory is almost identical to the original, with two differences: 207 values were randomly removed from the total_bedrooms column, so we can discuss what to do with missing data. An additional categorical attribute called ocean_proximity was added, indicating (very roughly) whether each block group is near the ocean, near the Bay area, inland or on an island. This allows discussing what to do with categorical data. Note that the block groups are called "districts" in the Jupyter notebooks, simply because in some contexts the name "block group" was confusing."

About the Data (From Luís Torgo page):

http://www.dcc.fc.up.pt/%7Eltorgo/Regression/cal_housing.html

This is a dataset obtained from the StatLib repository. Here is the included description:

"We collected information on the variables using all the block groups in California from the 1990 Cens us. In this sample a block group on average includes 1425.5 individuals living in a geographically co mpact area. Naturally, the geographical area included varies inversely with the population density. W e computed distances among the centroids of each block group as measured in latitude and longitude. W e excluded all the block groups reporting zero entries for the independent and dependent variables. T he final data contained 20,640 observations on 9 variables. The dependent variable is ln(median house value)."

End-to-End ML Project Steps (Chapter 2 of the book)

1. Look at the big picture
2. Get the data
3. Discover and visualize the data to gain insights
4. Prepare the data for Machine Learning algorithms
5. Select a model and train it
6. Fine-tune your model
7. Present your solution
8. Launch, monitor, and maintain your system

The 10-Step Machine Learning Project Workflow (My Version)

1. Define business object
2. Make sense of the data from a high level
   • data types (number, text, object, etc.)
   • continuous/discrete
   • basic stats (min, max, std, median, etc.) using boxplot
   • frequency via histogram
   • scales and distributions of different features
3. Create the traning and test sets using proper sampling methods, e.g., random vs. stratified
4. Correlation analysis (pair-wise and attribute combinations)
5. Data cleaning (missing data, outliers, data errors)
6. Data transformation via pipelines (categorical text to number using one hot encoding, feature scaling via normalization/standardization, feature combinations)
7. Train and cross validate different models and select the most promising one (Linear Regression, Decision Tree, and Random Forest were tried in this tutorial)
8. Fine tune the model using trying different combinations of hyperparameters
9. Evaluate the model with best estimators in the test set
10. Launch, monitor, and refresh the model and system

Estimated number of persons by quarter of a year and by year, Canada, provinces and territories.

This table presents the 2021 and 2016 population and dwelling counts, land area, population density and population ranking for census metropolitan areas or census agglomerations. It also shows the percentage change in the population and dwelling counts between 2016 and 2021.

A raster representing Greater Sage-grouse (hereafter sage-grouse) space-use and lek abundance. A higher pixel value corresponds to a greater amount of likelihood that the area is utilized by sage-grouse. Values are the result of combining a kernel density estimation on lek abundances with a raster representing distance to lek. The kernel density was calculated using maximum lek abundances observed between the most recent population nadir for the Great Basin region (2013) and the most recent lek counts available (2021). Polygons representing high-space use areas of Greater Sage-grouse (hereafter sage-grouse) space-use and lek abundance. Areas represent the 85 percent isopleth of the abundance and space-use index (ASUI) as well as a 5-kilometer buffer around remote leks that did not fall within the 85 percent isopleth, so that remote leks were not under-represented.