Lit Review Variables Exploration

Author

Nissim Lebovits

Published

June 16, 2023

Overview

Variables by Study

The table below lists the studies and datasets that I’ve identified as most relevant to this research. The variables column lists all of the input variables that each study or model uses, while the approach column lists the analytical technique, where relevant. In this section, the variables are given as listed in the original study. Below, I have used string cleaning to aggregate the variables by theme to identify the most common topics accounted for in these studies.

title author_s date variables approach
Socioeconomic vulnerability and climate risk in coastal Virginia Sadegh Eghdami, Andrew M. Scheld, Garrick Louis 2023 Household median income (inflation-adjusted); Housing median value; Renters' share of households; Poverty rate of households; Share of social assistance receivers; Share of Black households; Share of below diploma; Share of bachelor’s' degree & higher; Unemployment rate; Ratio of military to civilian employment; Share of no-internet access households; Share of over 65 population Multidimensional regressions
How do community-level climate change vulnerability assessments treat future vulnerability and integrate diverse datasets? A review of the literature Emma J. Windfeld, James D. Ford, Lea Berrang-Ford and Graham McDowell 2019 Median age; Per capita income; Median value of owner-occupied housing units; Gross rent; Physicians per capita; Percent voting Obama in 2008; Birth rate per capita; Net internal migration per capita; Percent land in farms; Percent population Black or African American alone; Percent population American Indian and Alaska Native Alone; Percent population Asian alone; Percent population Hispanic or Latino; Percent population under 5 years old; Percent total population under 65 years old; Percent unemployed in civilian force age 16 and over; Average household size; Percent of households earning more than $75,000/year; percent population below poverty line; Percent housing units occupied by renters; Number of farm operators per capita; Percent housing units that are mobile homes; Percent population age 25 and over with no HS degree; Housing units per square mile; Private housing building permits; Manufacturing establishments; Sales, receipts, and value of shipments for all firms; Number of firms; Farm income; Percent population 16 and over in the labor force; Percent females 16 and over in civilian labor force; Percent population employed in extractive industries; Percent population employed in transport, utility, information; Percent population employed in services (education, arts, other); Percent population in nursing homes; Hospital beds per capita; Percent population change 2000-2010; Percent urban population; Percent population that is female; Percent households with female householder, no husband; Percent households with Social Security income PCA + k-means (compared)
Clusters of community exposure to coastal flooding hazards based on storm and sea level rise scenarios—implications for adaptation networks in the San Francisco Bay region Michelle A. Hummel, Nathan J. Wood, Amy Schweikert, Mark T. Stacey, Jeanne Jones, Patrick L. Barnard, Li Erikson 2017 Number and percent of residents; Number and percent of employees; Number of critical facilities; Length of critical infrastructure; Ethnicity; Non-White race; Age; Tenancy; Group quarters; Number and percent of employees in each of five classes: government and critical facilities, manufacturing, services, natural resources, trade PAM clustering (uses mediods, not centroids)
A place-based socioeconomic status index: Measuring social vulnerability to flood hazards in the context of environmental justice Liton Chakraborty, Horatiu Rus, Daniel Henstra, Jason Thistlethwaite, Daniel Scott 2019 Female; Female labor force participation; Age; Senior; Children under 5 years of age; Children under 15 years of age; Psychological disability; Physical disability; Unattached one-person; Unattached elderly; Lone parents; More than three children in a census family; Household size; Official language knowledge; English/French; First-generation status; Foreign born Canadian citizens; Aboriginal Peoples; Indian/Inuit/Metis; Year of immigration; White; Black; South Asian; Chinese; Filipino; Latin American; No certificate/diploma; Post-secondary certificate; Shelter-cost-to-income ratio; Government transfer; Low income; Dwelling value; Income; Management; Business, finance, and administration; Health; Education, law, social community and government service; Sales and service; Unemployed; Not in the labor force; House with major repair; Crowded home; Period of home construction; Dwelling is in apartment with 5+ stories built before 1980; Renters; No private vehicle/Public transit; Population density; Mobility; Dwelling size PCA
SVI CDC 2020 Below 150% poverty; Unemployed; Housing cost burden; No high school diploma; No health insurance; Aged 54 and older; Aged 17 and younger; Civilian with disability; Single-parent households; English language proficiency; Hispanic or Latino; Black or African American (not Hispanic or Latino); Asian (not Hispanic or Latino); American Indian or Alaska Native (not Hispanic or Latino); Native Hawaiian or Pacific Islander (not Hispanic or Latino); Two or more races (not Hispanic or Latino); Other races (not Hispanic or Latino); Multi-unit structures; Mobile homes; Crowding; No vehicle; Group quarters Simple ranked aggregation
SoVI Cutter (see "Social Vulnerability to Environmental Hazards") 2014 Percent Asian; Percent Black; Percent Hispanic; Percent Native American; Percent Population under 5 years or 65 and over; Percent Children Living in 2-parent families; Median Age; Percent Households Receiving Social Security Benefits; Percent Poverty; Percent Households Earning over $200,000 annually; Per Capita Income; Percent Speaking English as a Second Language with Limited English Proficiency; Percent Female; Percent Female Headed Households; Nursing Home Residents Per Capita; Percent of population without health insurance (County Level ONLY); Percent with Less than 12th Grade Education; Percent Civilian Unemployment; People per Housing Unit; Percent Renters; Median Housing Value; Median Gross Rent; Percent Mobile Homes; Percent Employment in Extractive Industries; Percent Employment in Service Industry; Percent Female Participation in Labor Force; Percent of Housing Units with No Car; Percent Unoccupied Housing Units; Hospitals per Capita (County Level ONLY); PCA
Fault zone regulation, seismic hazard, and social vulnerability in Los Angeles, California: Hazard or urban amenity? Toke et al 2014 NA Index
Quantifying coastal flood vulnerability for climate adaptation policy using principal component analysis Wu 2021 Developed Land Use; Municipal & Private Open Space; Tidal Wetlands; Inland Wetlands; Forestry Area; Artificial Filled Land; Well-drained Soil; Aquifer Protection Area; Impervious Land Cover; Population Density; Population with No HS Diploma; Population with a Disability; Households with No Vehicle; Per Capita Income; Population Below Poverty; Unemployed Population; Median House Value; Aggregated House Value; Socio-Economic Vulnerability Index; FEMA 100-year Flooding Area; CIRCA Projected 100-year Flooding Area with 20-inch SLR in 2050; Hurricane (Category 4) Surge Inundation Area; Developed Land in Flooding Area; Roadway in Flooding Area; Buildings in Flooding Area; Critical Facilities in Flooding Area; Population Exposed to Coastal Flooding; Mean Elevation; Erosion Susceptible Land; Land Loss with 20-inch SLR in 2050; PCA
Planning for Climate Adaptation: Evaluating the Changing Patterns of Social Vulnerability and Adaptation Challenges in Three Coastal Cities Bin Kashem et al 2016 Black or African American population; Hispanic population; Asian population; Native American population; Population under 5 years old; Population 65 years or older; Group quarters population; Foreign-born population; Household size; Female population; Female-headed households; Female labor force participation; Public transportation dependence; Education attainment; Unemployment rate; Manufacturing employment; Employment in service occupations; Poverty rate; Social security recipients; Average household income; Renter-occupied housing units; Number of mobile homes; Average gross rent; Average home value; Population in civilian labor force; Housing density PCA
Risks of sea level rise to disadvantaged communities in the United States Martinich et al 2011 SoVI PCA
A Validation of Metrics for Community Resilience to Natural Hazards and Disasters Using the Recovery from Hurricane Katrina as a Case Study Burton 2014 Social capacity % population that is not elderly; % population with vehicle access; % population with telephone access; % population that doesn’t speak English as a second language; % population without a disability; % population that is not institutionalized or infirmed; % population that is not a minority; % population with at least a high school diploma; % population living in high-intensity urban areas; Social assistance programs per 1,000 population; Adult education and training programs per 1,000 population; Child care programs per 1,000 population; Community services (recreational facilities, parks, historic sites, libraries, museums) per 1,000 population; Internet, television, radio, and telecommunications broadcasters per 1,000 population; Psychosocial support facilities per 1,000 population; Health services per 1,000 population; Equity Ratio % college degree to % no high school diploma; Ratio % minority to % nonminority population; % homeownership; % working age population that is employed; % female labor force participation; Per capita household income; Mean sales volume of businesses; Economic diversity % population not employed in primary industries; Ratio of large to small businesses; Retail centers per 1,000 population; Commercial establishments per 1,000 population; Resource equity Lending institutions per 1,000 population; Doctors and medical professionals per 1,000 population; Ratio % white to % nonwhite homeowners; % commercial establishments outside of high hazard zones (flood, surge); Density of commercial infrastructure; % population covered by a recent hazard mitigation plan; % population participating in Community Rating System (CRS) for flood; % households covered by National Flood Insurance Program Policies; Preparedness % population with Citizen Corps program participation; % workforce employed in emergency services (firefighting, law enforcement, protection); Number of paid disaster declarations; Development % land cover change to urban areas from 1990 to 2000; Housing type % housing that is not a mobile home; % housing not built before 1970; after 1994; Response and recovery % housing that is vacant rental units; Hotels and motels per square mile; Fire, police, emergency relief services, and temporary shelters per 1,000 population; % fire, police, emergency relief services, and temporary shelters outside of hazard zones; Schools (primary and secondary education) per square mile; Principal arterial miles; Number of rail miles; Density of single-family detached homes; % building infrastructure not in flood and storm surge inundation zones; % building infrastructure not in high hazard erosion zones; Social capital Religious organizations per 1,000 population; Social advocacy organizations per 1,000 population; Arts, entertainment, and recreation centers per 1,000 population; Civic organizations per 1,000 population; Creative class % workforce employed in professional occupations; Professional, scientific, and technical services per 1,000 population; Research and development firms per 1,000 population; Business and professional organizations per 1,000 population; National Historic Registry sites per square mile; Sense of place % population born in a state and still residing in that state; % population that is not an international migrant; Risk and exposure % land area that does not contain erodible soils; % land area not in an inundation zone (100/500-year flood and storm surge combined); % land area not in high landslide incidence zones; Number of river miles; % land area that is nondeveloped forest; % land area with no wetland decline; % land area with no land-cover/land-use change; % land area under protected status; % land area that is arable cultivated land; % land area that consists of windbreaks and environmental plantings; % land area that is a wetland, swamp, marsh, mangrove, sand dune, or natural barrier; % land area that is developed open space; Frequency of loss-causing weather events (hail, wind, tornado, hurricane); NA

Counts of Variables by Frequency Across All Studies

Because each study lists and/or measures each variable in a slightly different way, I’ve used string cleaning to try to identify the most common themes across all of the studies. All topics occurring more than once are shown by frequency in the bar plot below. After that, a table lists all the identified variables in descending order of frequency.

Note that, because this string matching is imperfect, this chart shouldn’t be interpreted as an exact count of how often each theme is mentioned in the literature, but rather as a way of getting a sense of, proportionally, how often items occur. Each theme probably has a margin of error of +/- an occurrence or two.

Finally, “employment by industry” is an overcount due to it showing up once per industry (e.g., “employment in extractive industries”, employment in service industry”, etc.) in Cutter’s SoVI (2014) as well as other papers based on it, such as Hummel et al (2017).

Show string cleaning code
data <- data |>  
          mutate(variables = tolower(variables))

to_remove <- c("per 1,000 population", "population", "social capacity", "that is", "average", "median", "percent", "%", "alone",
               "canadian citizens", "(inflation adjusted)", "sense of place", "share of", "or latino", "or african american", "line",
               "aggregated", "\\)", "\\(", "not hispanic", "rate", "total", "socio economic vulnerability index", "county level only", 
               "mobility", "education attainment", "sovi")

pattern <- paste(to_remove, collapse = "|")

data$variables <- str_replace_all(data$variables, "  ", " ") |>
                      str_replace_all("dwelling value", "housing value") |>
                      str_replace_all("-", " ") |>
                      str_replace_all("home value", "housing value") |>
                      str_replace_all("house value", "housing value") |>
                      str_remove_all(pattern) |> # note that this line is string remove, not replace
                      str_replace_all("  ", " ") |>
                      str_replace_all("dwelling value", "housing value") |>
                      str_replace_all("-", " ") |>
                      str_replace_all("home value", "housing value") |>
                      str_replace_all("house value", "housing value") |>
                      str_replace_all("working age employed", "unemployment") |>
                      str_replace_all("with vehicle access", "vehicle access") |>
                      str_replace_all("without a disability", "disability") |>
                      str_replace_all("with a disability", "disability") |>
                      str_replace_all("unemployed in civilian force age 16 and over", "unemployment") |>
                      str_replace_all("unemployed", "unemployment") |>
                      str_replace_all("renters' households", "renters") |>
                      str_replace_all("poverty of households", "poverty") |>
                      str_replace_all("below poverty", "poverty") |>
                      str_replace_all("senior", "over 65") |>
                      str_replace_all("65 years or older", "over 65") |>
                      str_replace_all("of without health insurance", "health insurance") |>
                      str_replace_all("of housing units with no car", "vehicle access") |>
                      str_replace_all("number and of residents", "total population") |>
                      str_replace_all("non white race", "minority") |>
                      str_replace_all("not a minority", "minority") |>
                      str_replace_all("households with no vehicle", "vehicle access") |>
                      str_replace_all("no vehicle", "vehicle access") |>
                      str_replace_all("no private vehicle/public transit", "vehicle access") |>
                      str_replace_all("low income", "poverty") |>
                      str_replace_all("latin american", "hispanic") |>
                      str_replace_all("indian/inuit/metis", "native american") |>
                      str_replace_all("housing units per square mile", "housing density") |>
                      str_replace_all("housing units that are mobile homes", "mobile homes") |>
                      str_replace_all("housing type housing not a mobile home", "mobile homes") |>
                      str_replace_all("females 16 and over in civilian labor force", "female participation in labor force") |>
                      str_replace_all("density of single family detached homes", "housing density") |>
                      str_replace_all("civilian unemployment", "unemployment") |>
                      str_replace_all("change 2000 2010", "population change") |>
                      str_replace_all("black households", "black") |>
                      str_replace_all("american indian and alaska native", "native american") |>
                      str_replace_all("american indian or alaska native", "native american") |>
                      str_replace_all("aboriginal peoples", "native american") |>
                      str_replace_all("children under 5 years of age", "under 5 years old") |>
                      str_replace_all("female participation in labor force", "female labor force participation") |>
                      str_replace_all("household income", "income") |>
                      str_replace_all("age 25 and over with no hs degree", "no diploma") |>
                      str_replace_all("below diploma", "no diploma") |>
                      str_replace_all("children under 15 years of age", "under 15 years old") |>
                      str_replace_all('civilian with disability', "disability") |>
                      str_replace_all("crowded home", "crowding") |>
                      str_replace_all("english language proficiency", "nonnative english") |>
                      str_replace_all("english/french", "nonnative english") |>
                      str_replace_all("first generation status", "foreign born") |>
                      str_replace_all("no high school diploma", "no diploma") |>
                      str_replace_all("not an international migrant", "foreign born") |>
                      str_replace_all("not elderly", "over 65") |>
                      str_replace_all("not in the labor force", "in civilian labor force") |>
                      str_replace_all("physical disability", "disability") |>
                      str_replace_all("per capita household income", "income") |>
                      str_replace_all("psychological disability", "disability") |>
                      str_replace_all("south asian", "asian") |>
                      str_replace_all("that doesn't speak english as a second language", "nonnative english") |>
                      str_replace_all("speaking english as a second language with limited english proficiency", "nonnative english") |>
                      str_replace_all("value of owner occupied housing units", "housing value") |>
                      str_replace_all("with at least a high school diploma", "diploma") |>
                      str_replace_all("with less than 12th grade education", "no diploma") |>
                      str_replace_all("working age  employed", "employment") |>
                      str_replace_all("year of immigration", "foreign born") |>
                      str_replace_all("per capita income", "income") |>
                      str_replace_all("that doesn't speak english as a second language", "nonnative english") |>
                      str_replace_all("shelter cost to income ratio", "housing cost burden") |>
                      str_replace_all("ratio minority to nonminority", "minority") |>
                      str_replace_all("ratio white to nonwhite homeowners", "minority") |>
                      str_replace_all("official language knowledge", "nonnative english") |>
                      str_replace_all("no internet access households", "no internet access") |>
                      str_replace_all("no certificate/diploma", "no diploma") |>
                      str_replace_all("housing not built before 1970", "housing age") |>
                      str_replace_all("households receiving social security benefits", "social security") |>
                      str_replace_all("households with social security income", "social security") |>
                      str_replace_all("16 and over in the labor force", "in civilian labor force") |>
                      str_replace_all("social security recipients", "social security") |>
                      str_replace_all("people per housing unit", "crowding") |>
                      str_replace_all("period of home construction", "housing age") |>
                      str_replace_all("other races", "minority") |>
                      str_replace_all("two or more races", "minority") |>
                      str_replace_all("that doesn’t speak english as a second language ", "nonnative english") |>
                      str_replace_all("number of mobile homes", "mobile homes") |>
                      str_replace_all("no health insurance", "health insurance") |>
                      str_replace_all("no internet access", "internet access") |>
                      str_replace_all("ethnicity", "hispanic") |>
                      str_replace_all("with no hs diploma", "no diploma") |>
                      str_replace_all("white", "minority")

variables count
employment by industry 14
housing value 7
income 7
minority 7
unemployment 7
black 6
hispanic 6
native american 6
no diploma 6
poverty 6
asian 5
disability 5
female labor force participation 5
foreign born 5
mobile homes 5
vehicle access 5
age 4
female 4
nonnative english 4
over 65 4
crowding 3
gross rent 3
group quarters 3
household size 3
housing density 3
in civilian labor force 3
renters 3
social security 3
under 5 years old 3
density 2
female headed households 2
health insurance 2
housing age 2
housing cost burden 2
adult education and training programs 1
after 1994 1
aged 17 and younger 1
aged 54 and older 1
aquifer protection area 1
artificial filled land 1
arts, entertainment, and recreation centers 1
bachelor’s' degree & higher 1
below 150 poverty 1
birth per capita 1
born in a state and still residing in that state 1
building infrastructure not in flood and storm surge inundation zones 1
building infrastructure not in high hazard erosion zones 1
buildings in flooding area 1
business and professional organizations 1
business, finance, and administration 1
child care programs 1
children living in 2 parent families 1
chinese 1
circa projected 100 year flooding area with 20 inch slr in 2050 1
civic organizations 1
commercial establishments 1
commercial establishments outside of high hazard zones flood, surge 1
community services recreational facilities, parks, historic sites, libraries, museums 1
covered by a recent hazard mitigation plan 1
critical facilities in flooding area 1
density of commercial infrastructure 1
developed land in flooding area 1
developed land use 1
development land cover change to urban areas from 1990 to 2000 1
diploma 1
doctors and medical professionals 1
dwelling is in apartment with 5+ stories built before 1980 1
dwelling size 1
education, law, social community and government service 1
equity ratio college degree to no diploma 1
erosion susceptible land 1
exposed to coastal flooding 1
farm income 1
fema 100 year flooding area 1
filipino 1
fire, police, emergency relief services, and temporary shelters 1
fire, police, emergency relief services, and temporary shelters outside of hazard zones 1
forestry area 1
frequency of loss causing weather events hail, wind, tornado, hurricane 1
government transfer 1
health 1
health services 1
homeownership 1
hospital beds per capita 1
hospitals per capita 1
hotels and motels per square mile 1
house with major repair 1
households covered by national flood insurance program policies 1
households earning over $200,000 annually 1
households with female householder, no husband 1
housing units occupied by renters 1
hurricane category 4 surge inundation area 1
impervious land cover 1
in nursing homes 1
inland wetlands 1
internet access 1
internet, television, radio, and telecommunications broadcasters 1
land area a wetland, swamp, marsh, mangrove, sand dune, or natural barrier 1
land area arable cultivated land 1
land area developed open space 1
land area nondeveloped forest 1
land area not in an inundation zone 100/500 year flood and storm surge combined 1
land area not in high landslide incidence zones 1
land area that consists of windbreaks and environmental plantings 1
land area under protected status 1
land area with no land cover/land use change 1
land area with no wetland dec 1
land in farms 1
land loss with 20 inch slr in 2050 1
length of critical infrastructure 1
living in high intensity urban areas 1
lone parents 1
management 1
manufacturing establishments 1
mean elevation 1
mean sales volume of businesses 1
more than three children in a census family 1
multi unit structures 1
municipal & private open space 1
national historic registry sites per square mile 1
native hawaiian or pacific islander 1
net internal migration per capita 1
not institutionalized or infirmed 1
number of critical facilities 1
number of farm operators per capita 1
number of firms 1
number of paid disaster declarations 1
number of rail miles 1
number of river miles 1
nursing home residents per capita 1
of households earning more than $75,000/year 1
participating in community rating system crs for flood 1
physicians per capita 1
population change 1
post secondary certificate 1
preparedness with citizen corps program participation 1
principal arterial miles 1
private housing building permits 1
professional, scientific, and technical services 1
psychosocial support facilities 1
public transportation dependence 1
ratio of large to small businesses 1
renter occupied housing units 1
research and development firms 1
resource equity lending institutions 1
response and recovery housing vacant rental units 1
retail centers 1
risk and exposure land area that does not contain erodible soils 1
roadway in flooding area 1
sales and service 1
sales, receipts, and value of shipments for all firms 1
schools primary and secondary education per square mile 1
single parent households 1
social advocacy organizations 1
social assistance programs 1
social assistance receivers 1
social capital religious organizations 1
tenancy 1
that doesn’t speak english as a second language 1
tidal wetlands 1
total population 1
unattached elderly 1
unattached one person 1
under 15 years old 1
under 5 years or 65 and over 1
under 65 years old 1
unoccupied housing units 1
urban 1
voting obama in 2008 1
well drained soil 1
with telephone access 1

Variables Occurring Once But Worth Tracking

Some variables show up only infrequently, either due to limited use or to difficulty in string matching, but are still worth considering. I’ve identified them manually here:

  • Aquifer protection area

  • Artificial filled land

  • Land cover changes in past decade

  • Erosion susceptible land

  • Farm income

  • FEMA 100 year floodplain

  • Forestry area

  • Homeownership rate (or ratio of homeowners to renters)

  • NFIP participation

  • Impervious surface cover

  • Wetlands (total land cover by type of wetland)

  • Internet access Land cover (% by type)

  • Development intensity

  • Households earning more than $75,000/year

  • Households earning more than $200,000/year

  • Population change in the past decade

  • Unoccupied housing units

  • Well-drained soil

  • Has telephone access

  • Works in fishing

  • Works in agriculture

Note that many of these may be considered climate variables, not socioeconomic. Land use and land cover (LULC) came up frequently across the studies that I encountered and is certainly worth incorporating into the analysis.

Finally, many of these variables, especially around commerce or building infrastructure, would be worth considering were it not for the difficulty of collecting and reconciling them at such a large spatial scale (e.g., buildings in flooding area, length of critical infrastructure, covered by a recent hazard mitigation plan).