Generally defined, vulnerability is the potential for loss due to hazards ranging from loss of property, natural resources, or human life. Social vulnerability represents those social, economic, demographic, and housing characteristics that influence a community’s ability to prepare for, respond to, cope with, recover from, and adapt to environmental hazards. The hazards-of-place model (Cutter 1996) combines social vulnerability and biophysical vulnerability (physical characteristics of hazards and environment) to determine the overall hazard vulnerability of places.

• Socioeconomic Status (Income, Political Power, Prestige): Socioeconomic status affects the ability of a community to absorb losses and cope with hazard impacts. Wealth enables communities to better prepare for disasters through mitigation and absorb and recover from losses more quickly using insurance, social safety nets, and entitlement programs. Low status communities have little ability to absorb losses due to poverty and disadvantaged populations. 
• Gender: Women often have a more difficult time during recovery than men because of sector-specific employment (e.g., personal services), lower wages, and family care responsibilities.
• Race and Ethnicity: These factors impose language and cultural barriers and affect access to post-disaster funding and occupation of high-hazard areas.
• Age: Extremes of age affect movement out of harm’s way and require outside supervision and care. Parents lose time and money caring for children when day care facilities are affected; the elderly may have mobility constraints or medical and cognitive concerns increasing the burden of care before, during, and after the emergency.
• Employment Loss: The potential loss of employment following a disaster increases the existing number of unemployed workers in a community. Such losses compound the impact of the hazard and leads to a slower recovery from the disaster.  At an individual level, employment loss equates to a lower ability to pay for necessary goods and services, effectively lowering the ability to prepare and recovery from disasters.
• Residential Property: Home value is an indicator of financial capacity.  The value and quality of residential construction affect potential losses and recovery. Expensive homes are costly to replace, mobile homes are easily destroyed by water and winds.  The viability of neighborhoods based on the number of unoccupied housing units also contributes to slower long term recovery. 
• Renters: People rent because they are transients, do not have the financial resources for home ownership, or do not want the responsibility of home ownership. They often lack access to information about financial aid during recovery and are not covered by current federal disaster recovery programs. In extreme cases, renters lack sufficient shelter options when lodging becomes uninhabitable or too costly to afford.
• Occupation: Some occupations, especially those characterized as primary extractive industries, may be severely affected by a hazard event. Primary sector jobs are impacted first during major disasters.  For example, self-employed fishermen suffer when their means of production is lost (boats), and they may not have the requisite capital to resume work in a timely fashion; therefore, they may seek alternative employment. The same is true of migrant workers engaged in agriculture.  Low-skilled service jobs (housekeeping, child care, and gardening) may suffer similarly as disposable income fades and the need for services declines.
• Family Structure: Families with large numbers of dependents and/or single-parent households often have limited resources to outsource care for dependents and thus must juggle work responsibilities with care for family members. All these factors affect coping with and recovering from hazards.
• Education: Education is linked to socioeconomic status in that higher educational attainment affects lifetime earnings, and limited education constrains the ability to understand warning information and access recovery information.
• Medical Services and Access: Health care providers, including physicians and hospitals, are important post-event sources of relief. The lack of proximate medical services lengthens the time needed to obtain short-term relief and achieve longer-term recovery from disasters. Nursing homes represent an increase in socially vulnerable people as the resident populations are less able to independently cope with disasters. The availability of health insurance is another factor influencing social vulnerability. 
• Social Dependence: People who are totally dependent on social services (social security, food assistance) for survival are already economically and socially marginalized and require additional support in the post-disaster period.
• Special-needs Population: Special-needs populations (infirm, institutionalized, transient, homeless) are difficult to identify, let alone measure and monitor. Yet it is this segment of society that invariably is left out of recovery efforts, largely because of this invisibility in communities.

The majority of the sources used by the HVRI are obtained from the five-year American Community Survey estimates compiled by the U.S. Census Bureau. Data are also obtained from the Geographic Names and Information System (GNIS), and model-based Small Area Health Insurance Estimates (SAHIE) published by the U.S Census Bureau. Data variables are used to represent the population characteristics that affect social vulnerability (See list above). For instance, the number of people older than 65 and the number of people under 5 years old were used to represent the socially vulnerable population due to age.

County-level socioeconomic and demographic data were used to construct an index of social vulnerability to environmental hazards, called the Social Vulnerability Index (SoVI®) for the United States. SoVI® is an empirically-based comparative assessment of social vulnerability of places. The current configuration uses data from 2010-2014 and 29 distinct variables for county study areas. The original SoVI® methodology and results was published in the following article: Cutter, S.L., B.J. Boruff, and W.L. Shirley. 2003. “Social Vulnerability to Environmental Hazards.” Social Science Quarterly 84(2): 242–261.

The SoVI® is constructed using a statistical procedure called a principal components analysis.  The output (factors that are generated) are then labeled and their influence on social vulnerability determined (increases or decreases).  The factor scores and their directional adjustments (increases or decreases vulnerability) are then put into an additive model to generate the total score.  The scores are then mapped using standard deviations from the mean, normally using either 3 or 5 classes.  Get the SoVI® recipe.  [PDF]

The SoVI® was created as a comparative index at a county-level for the entire United States. Therefore, the SoVI® scores need to be displayed in relation to each other. Generally, the SoVI® is classified using standard deviations. Social vulnerability scores that are greater than 1.5 standard deviations above the mean are considered the most socially vulnerable places and are usually mapped in red, while scores below 1.5 standard deviations less than the mean are the least vulnerable places usually mapped in blue.

The total SoVI® score is represented as a numeric value, but it has no inherent mathematical properties.  Because the score is a relative score and not an absolute score, it cannot be used to compare two places directly (e.g., a county with a SoVI® score of 10 does not have double the vulnerability of a county with a SoVI® score of 5).  SoVI® scores are used to show the relative placement of a county relative to others on the continuum of scores with variable ranges.  As such, SoVI® scores should be classed (e.g., by standard deviation) for mapping and analysis purposes or can be examined using percentile ranks. 

The variable names represent short-hand notations for longer descriptions of the variables.  The codebook is listed below.

QASIAN — Percent Asian
QBLACK — Percent Black
QSPANISH — Percent Hispanic
QINDIAN — Percent Native American
QAGEDEP — Percent Population under 5 years or 65 and over
QFAM — Percent Children Living in 2-parent families
MEDAGE — Median Age
QSSBEN — Percent Households Receiving Social Security Benefits
QPOVTY — Percent Poverty
QRICH — Percent Households Earning over $200,000 annually
PERCAP — Per Capita Income
QESL — Percent Speaking English as a Second Language with Limited English Proficiency
QFEMALE — Percent Female
QFHH — Percent Female Headed Households
QNRRES — Nursing Home Residents Per Capita
QNOHLTH — Percent of population without health insurance (County Level ONLY)
QED12LES — Percent with Less than 12th Grade Education
QCVLUN — Percent Civilian Unemployment
PPUNIT — People per Housing Unit
QRENTER — Percent Renters
MDHSEVAL — Median Housing Value
MDGRENT — Median Gross Rent
QMOHO — Percent Mobile Homes
QEXTRCT — Percent Employment in Extractive Industries
QSERV — Percent Employment in Service Industry
QFEMLBR — Percent Female Participation in Labor Force
QNOAUTO — Percent of Housing Units with No Car
QUNOCCHU — Percent Unoccupied Housing Units
HOSPTPC — Hospitals per Capita (County Level ONLY)