Developing the Dimension and Index Scores

The methodology for calculating the indicator scores and JustSouth Index rankings is based on the United Nations’ Human Development Index “goalpost approach” to measuring well-being.  All 50 states and Washington D.C. were given a score on each of the nine indicators in relation to the highest and lowest observable indicator values. The state with highest indicator value was given a score of 1 and the rest of the states receive a standardized score between 0 and 1 according to their respective indicator values.

The three indicator scores under each of the three dimensions were used to calculate the poverty, racial diversity, and immigration exclusion dimension index scores. This was done by calculating the arithmetic mean of each grouping of indicators. We used the arithmetic mean because each indicator is considered independent from the other indicators. The arithmetic mean weights each of the indicators equally, which means that a high or low score on one indicator can drastically affect a state’s dimension score.

With each dimension being a proxy for social justice, the dimensions are not independent of each other. As a result, in order to create an overall index, the JustSouth Index, the geometric mean of the dimension scores were used. The geometric mean “normalizes” the scores being combined, so that no score dominates the weighting. This ensures that very high or low scores on one dimension do not drastically affect a state’s overall score. As with the individual indicator scores each dimension index score falls between the “goal posts” of 0 and 1, with the highest scores closest to 1 and the lowest scores closest to 0. 

Indicator Methodology

Poverty

Average Annual Income of Households in Lowest Income Quartile

Considers the total annual income* of each household in a state to isolate the households in the lowest income quartile. The mean income of households in the lowest income quartile in a state represents the state’s indicator value. Source: U.S. Census Bureau. 2014 American Community Survey Public Use Microdata Sample; Household file

Share of Persons in Lowest Income Quartile without Health Insurance

Compares the total number of persons between
the ages of 15 and 64 who are in the lowest income quartile* in a state to the number of the persons in that income quartile who reported not having any form of public or private health insurance coverage at time of survey. Source: U.S. Census Bureau. 2014 American Community Survey Public Use Microdata Sample; Individual file

Percent of Households in Lowest Income Quartile with a High Housing Cost Burden

Compares the total number of households in a state that are in the lowest income* quartile and rent their home to the number of those households that report spending more than 30 percent of their income in gross rent costs in the past 12 months. Source: U.S. Census Bureau. 2014 American Community Survey Public Use Microdata Sample; Household file

Racial Disparity

Percent of Segregated Schools

Considers the percentage of public schools in a state that have a student population that is 90 percent or more one race and that student population is more than five percentage points different than the total public school population in the county. Source: National Center for Education Statistics, Elementary/Secondary Information System, 2013

White-Minority Wage Gap

Compares the hourly wages of working age (ages 18-64) white persons in a state to the hourly wages of working age non-white persons while controlling for age, level of education, and occupation using a weighted regression analysis to determine the isolated impact of minority status on earnings. Source: 2014 Current Population Survey*

White-Minority Unemployment Rate Gap

Compares the percentage of white persons aged 16 years and over that reported being unemployed to the percentage of non-white persons aged 16 years and over who reported being unemployed. (Unemployment is defined as respondent did not have employment during the last week, were available for work, and had made specific efforts to find employment sometime during the preceding 4-week period.) Source: 2014 Current Population Survey Microdata**

Immigrant Exclusion

Disconnected Immigrant Youth Rate

Compares the total number of foreign-born youth (persons aged 18-25 who were not born in the U.S.) in a state with the number of foreign-born youth who reported that they were not enrolled in school in the last three months and had not worked in the last week. Source: U.S. Census Bureau. 2014 American Community Survey Public Use Microdata Sample; Individual file

Percent of Immigrant Population with Difficulty Speaking English

Considers the number of foreign-born individuals in each state who entered the US in 2011 or earlier who reported speaking English “not well” or “not at all” compared to the total population of foreign- born individuals in each state. Source: U.S. Census Bureau. 2014 American Community Survey Public Use Microdata Sample; Individual file

Gap in Health Insurance Rate
 Between Immigrant and Native-Born Populations

Considers the difference in the percentage of total foreign-born residents of a state (all who reported being born in another country) who reported not having public or private health insurance in 2014 compared to the percentage of all native-born residents of the state who report not having public or private health insurance. Source: 2014 American Community Survey, detailed table created using American FactFinder.

Additional Notes

*All indicators that are based on individual or household income include earned wages, commissions, bonuses, or tips; self-employment income; interest, dividends, net rental income, royalty income, or income from estates and trusts; Social Security or Railroad Retirement; Supplemental Social Security Income; public assistance or welfare payment; retirement, survivor or disability pensions; Veterans’ payments, unemployment compensation, and child support payments.

All income and earnings data are reported in 2014 inflation adjusted dollars. 

**Current Population Survey data was analyzed using the coded extracts provided by the Economic Policy Institute’s Economic Analysis and Research Network. All income and earnings data are reported in 2014 income adjusted dollars.

Detailed statistical output and tables generated by Jeanie Donovan using the U.S. Census Bureau’s Public Use Microdata Sample (PUMS) file, Community Population Survey, and National Center for Education Statistics Elementary/ Secondary Information (ESLi) System.

For more detailed statistical methodological information please contact: mailto:jjdonova@loyno.edu