Economic Mobility in the United States

This set of ten visualizations explores economic mobility in the United States. It seeks to communicate how where you’re from or what you look like might affect your chances of ending up in jail or earning more/less income than your parents.

The research data presented below comes from two data sets distributed by Opportunity Insights, a non-partisan, not-for-profit organization located at Harvard University.

One data set reports predicted outcomes for incarceration and income percentile by U.S Census Tract. Predicted outcomes are a result of study participants born between 1978 and 1983. All participants were born at the 25th income percentile of national income. The study collected information on race and incarceration rates in 2010 for study participants. It collect information on race and income percentile in 2014-2015. Through this set of data, it’s possible to draw insights on if rates of economic mobility vary based upon race and geographic location. The data dictionary for this data set can be found here..

The second data set from Opportunity Insights is a collection of data from the U.S Census Bureau by U.S Census Tract. The collection is primarily a function of the Decennial Census and the American Community Survey. This data set allows researchers to draw insights regarding the neighborhood characteristics of U.S Census Tracts. The data dictionary for this data set can be found here.

Finally, there are four other data sets used below that come from the U.S Census Bureau and the usmap R package. These data sets are meant to provide meaningful names for U.S Counties, States, and U.S Census Regions.

-This is the data source of U.S Census Region Names

-This data source ties U.S FIPS Codes with U.S Census Region Division Codes

-This data source holds U.S State Abbreviations and Names by FIPS Codes

-U.S County Names come from the usmaps R package

Incarceration

The first variable this exploration will cover is incarceration. The plot below shows the distribution in incarceration of study participants by U.S Census Region and Race for all U.S Census tracts. Please note that the range is limited at 2 standard deviations from the mean incarceration for each U.S Census Region.

The next visualization is interactive. It shows the average incarceration proportion by state for all races in the study.

Income

Next, this exploration will cover the income percentile associated with study participants. The following visualization shows how the income percentile of study participants varies by U.S Census Region and Race.

The next visualization is interactive. It shows the average income percentile by county for all races in the study.

Next, we view the average income percentile by state and race for study participants in each U.S Census region. The median for each census region is denoted in red for comparative purposes.

Finally, we view both variables (income and incarceration) together on a single scatter plot. Information is aggregated by U.S County and Race for comparative purposes. Three counties with high incarceration are noted directly in the plot.

Comparing Incarceration and Income with Community Characteristics

The following set of visualizations uses data from the Decennial Census and American Community Survey to track how neighborhood characteristics within U.S Census tracts might influence income and incarceration for the study participants.

The first visualization is a correlogram of the key variables that help define both data sets. Income shows interesting correlation patterns with four characteristics below. Incarceration does not share as strong of a relationship with the characteristics relative to income, so income is explored exclusively going forward. Definitions for each of the variables is provided below:

Income - Mean household income rank for children whose parents were at the 25th percentile of the national income distribution. Incomes for children were measured as mean earnings in 2014-2015 when they were between the ages 31-37. Household income is defined as the sum of own and spouse’s income.

Jail - Fraction of children born in 1978-1983 birth cohorts with parents at the 25th percentile of the national income distribution who were incarcerated on April 1st, 2010. Incarceration is defined as residing in a federal detention center, federal prison, state prison, local jail, residential correctional facility, military jail, or juvenile correctional facility.

White - Racial Shares for Whites in the decennial Census. All races (except Latino) exclude Hispanics and Latinos.

Latino - Racial Shares for Latinos in the decennial Census. All races (except Latino) exclude Hispanics and Latinos.

Black - Racial Shares for Blacks in the decennial Census. All races (except Latino) exclude Hispanics and Latinos.

Immigrant - Number of foreign born residents in the 2010 Census divided by the sum of native and foreign born residents. Obtained from the ACS 2006-2010.

College - Number of people aged 25 or older who have a bachelor’s degree, master’s degree, professional school degree, or doctorate degree, divided by the total number of people aged 25 or older in a tract. We use the 2006-2010 ACS to obtain the estimate for 2010.

Single Parent - The number of households with females heads (and no husband present) or male heads (and no wife present) with own children under 18 years old present divided by the total number of households with own children present (1990 and 2000 estimates are from the decennial Census, and the 2010 estimate is from the 2006-2010 estimate).

Poverty - Share of individuals in the tract below the federal poverty line, measured in the decennial Census of the relevant year for the 1990 and 2000 estimates, and measured in the 2006-2010 ACS for the 2010 estimate.

Commute - Share of workers 16 years and over who do not work at home whose commute is shorter than 15 minutes. Measured in the 2006-2010 ACS

Given the high correlation between the Single Parent variable and income, we’ll take a closer look at how it varies. The next visualization shows the average prooportion of single parent households by state and U.S Census region. The median single parent household proportion is denoted in red for comparative purposes.

The next visualization directly compares the income percentile with the average proportion of single parent households by race and state. Outliers for both variables are labeled for each race. In this case, an outlier is more than two standard deviations away from the mean of each variable for each race.

The correlogram also showed a wide variability among racial share density in U.S Census Tracts and the income in those U.S Census Tracts. The visualization below explores the relationship in further detail. The income columns denote the income percentile for each race. The density rows denote the racial share density for each race. The two are placed on the same scatter plot to show differences in how Black, White, and Latino income vary given the density of Black, White, and Latino populations in U.S Census Tracts.