Notes on Selected Indicators
Life expectancy values were taken from a study conducted in collaboration between St. Louis County Public Health and the University of Minnesota Duluth. This study reported values at the neighborhood level and those values were assigned directly to the neighborhoods in this study.
Violent crime values were calculated by geocoding on year of crime data from Duluth Police Department, filtering out nonviolent crimes, and then aggregating to the level of census tracts. Values reported are rates per 1,000 population.
Income inequality is measured by the Gini Coefficient. The Gini Coefficient is an attempt to represent the level of income inequality in a population with a single number ranging from 0 to 1.
In a population where 10% of the population earns 10% of all income, 20% of the population earns 20% of all income, and so on, the Gini Coefficient would be 0, representing “total equality”. In a population where a single individual earns 100% of all income, the Gini Coefficient would be 1, representing maximum inequality. In Duluth census tracts, observed values range from 0.362 to 0.516.
Racial and Ethnic Diversity
Racial and ethnic diversity is measured using the Shannon-Weiner Diversity Index (SWDI). The value represents the level of uncertainty in determining the racial or ethnic background of a randomly selected individual based on the proportion of backgrounds of the population of a neighborhood.
For example, if you had a bag containing 9 red marbles and 1 blue marble, the SWDI of that bag would be very low; you could guess that a randomly selected marble would be red and be correct 90% of the time. If, on the other hand, you had a bag containing 5 red marbles and 5 blues marbles then the SWDI would be much higher; you would expect to guess correctly only 50% of the time.
Note that, because of the math used to calculate the SWDI, it is considered a unitless scale: the values don’t correspond to specific amounts or proportions. The lowest possible value is zero, but the highest possible value depends on the population being studied. For Duluth census tracts, observed values range from 0.27 to .97 with higher values representing greater diversity.
Excluded HCI Indicators
The core indicators “Motor Vehicle Collisions” and “Preventable Hospitalizations” were excluded due to an unavailability of reliable and timely local data. The core indicator “School Proximity to Traffic” was excluded because all values are zero. The optional indicators “Access to Mainstream Financial Services”, “Proximity to Brownfield Sites”, and “Proximity to Superfund Sites” were excluded due to a combination of unavailability of reliable data and irrelevance within the study area. The demographic and contextual measure “Park Quality” was excluded because comparison across cities is outside the scope of this study.
Household Transportation Costs
The geographic resolution of Household Transportation Costs is specified in the administrator’s guide as census block group; at the time of this study, these data were only available at the census tract level, which is one degree coarser.