Visualising Racial Inequity
Interactive data visualisation exploring racism, segregation, poverty & domestic violence in Chicago
- Data Visualisation, Data Analysis
Project Overview
Interactive Visual Narrative Exposing Racial Injustice
Decades of research have documented the disproportionately high rate of domestic violence among Black women and its relationship to poverty, neighbourhood disadvantage, and racism. But these findings have largely been relegated to academic journals. I created this interactive data visualisation narrative to depict the high rate of domestic violence among Black women in Chicago.
Explore the interactive data visualisation and read my complete dissertation for more detail on my final project for the Digital Media Master of Arts programme at NUI Galway.
Project Type
Master's Degree Final Project
My Roles
Data Analyst
Data Visualisation Artist
Content Strategist
Researcher
Achievements
First-Class Honours
Research
Refocusing Project on Domestic Violence & Racism
My final project did not go according to plan. I intended to create a data visualisation about the relationship between domestic violence rates and COVID-19 lockdown measures. But as I began conducting the literature review, the most consistent theme I found in existing scholarship was the strong relationship between domestic violence, race, and poverty.
After George Floyd was murdered by a police officer and mass protests for racial justice spread across the world, I felt a critical need to scuttle my original plans and change the topic of my final project. Black women’s experiences are often minimised in both discussions of domestic violence and discussions of racism. Black women’s unique vulnerability to domestic violence is well known in academic research and Black communities, but this disparity is rarely discussed elsewhere. I wanted to create a data visualisation that exposes this narrative to elevate awareness and create momentum for action.
The Project
Visualising Race, Poverty & Domestic Violence
Existing literature indicated that poverty and neighbourhood disadvantage show strong relationships to both individual and neighbourhoodlevel rates of domestic violence. Compared to white women, Black women have higher levels of poverty, are more likely to live in disadvantaged neighbourhoods, and experience domestic violence at higher rates. The large racial disparity in domestic violence rates reduces or disappears after controlling for individual and neighbourhood-level poverty. Present-day poverty, segregation, and racial inequity show strong relationships to historical segregation and redlining policies that locked Blacks out of homeownership.
Based on these findings, I developed research questions and a methodology to guide my data analysis.
Research Questions
- What is the relationship between the historical redlining map of Chicago and the geographic distribution of present-day domestic violence crime?
- What is the relationship between the geographic distribution of domestic violence crime and levels of Black segregation and poverty in Chicago?
- What data gaps and flaws emerge in the analysis, and how do they affect the findings and narrative of the story?
Methodology
- I conducted a quantitative time-series analysis of five years of geocoded domestic violence crime data in Chicago and racial and economic data for Chicago census tracts.
- I plotted this data on a present-day Chicago map showing individual census tracts.
- I also compared the data to a historical segregation map of Chicago from 1940.
Data Analysis
Research Findings
Results showed significant support for positive relationships between race, segregation, poverty, and domestic violence. Less support was found for a positive relationship between historically redlined areas of the city and current levels of domestic violence, partially as a result of limitations in the analysis. Missing data and high margins of error in nonwhite areas of Chicago emerged in the analysis.
Historical Segregation Compared to Present-Day Domestic Violence Rate
The comparison between the current location of concentrated domestic violence crime and historically redlined areas of Chicago provided mixed results. On the South Side of the city, the hypothesis was partially supported, with higher rates of domestic violence appearing in some areas with historical D-grade ratings that prevented residents from getting mortgages. But the high concentration of domestic violence on the West Side was not predicted by D-graded areas on the redlining map.
The relationship between historical redlining and current neighbourhood segregation and disadvantage is complex and multifaceted. This complexity prevents this research from drawing any definitive conclusions about the relationship between redlining and modern levels of domestic violence.
Degree of Black Segregation Compared to Domestic Violence Rate
The percentage of Black residents in a census tract was highly correlated with the number of crimes per 100 residents in the same geographic area. Visualisations show a clear pattern of domestic violence crime locations overlapping segregated Black areas on the South and West Sides of the city.
The strong correlation between the Black population and the rate of domestic violence is a finding that must be interpreted with care. Multiple studies have found that after controlling for individual poverty and neighbourhood disadvantage, the wide discrepancy between domestic violence rates among Blacks and whites decreases or disappears. Black women are also more likely to report domestic violence incidents than white women, which could over-represent domestic violence rates in predominantly Black census tracts.
Poverty Levels Compared to Domestic Violence Rate
Poverty also showed a clear correlation with elevated domestic violence rates across multiple different measures. Determining poverty is a complex matter involving many different factors; this study focuses on the percentage of unemployment, the percentage of residents living below the poverty line, the percentage of residents using food stamps, and the median household income in each census tract. Broadly, all measures of poverty positively correlated with higher rates of domestic violence. It is not a coincidence that the Blackest neighbourhoods in Chicago are also the poorest; this pattern is both a symptom of racism and segregation and a factor perpetuating racial inequality.
High Margin of Error for Child Poverty Rate in Nonwhite Areas
Analysis reveals that though the margin of error for the child poverty rate is high across the entire city, nonwhite census tracts have consistently higher margins of error than white census tracts. Differing population sizes and the percentage of children living in each census tract account for a great deal of this difference, as white census tracts tend to have higher populations. However, this pattern still emerges when majority-white tracts are compared to majority-Black tracts with similar populations. Missing data about poverty and extremely high margins of error in Black census tracts provide an indication of whose lives are valued by the Census Bureau and whose data is seen as worth collecting.
Final Data Visualisation
Data Visualisation Narrative
View the final Tableau data visualisation story summarising my research. For an optimal experience, view the data visualisation on a desktop or laptop in full-screen mode.
Reflection
What I Learned
This was my first large-scale data analysis and visualisation project, and the skills I learned throughout the process will benefit me greatly when analysing and visualising data in UX and content-strategy projects.
Achievements
- Mastered Tableau data visualisation software
- Received First-Class Honours in Digital Media MA from the National University of Ireland Galway
Insights
- My ability to harness the narrative potential of visualisations will be a valuable asset in future analysis of user research and usability testing.
- The limitations of off-the-shelf software created some usability issues I was unable to resolve. In future projects, I will consider developing my coding skills to create more bespoke visualisations that are not constrained by Tableau’s functionality.