Design Rationale

Process of Designing the Dashboard

We started this dashboard by identifying reliable data sources that could provide comprehensive measures of housing affordability across New York City neighborhoods. We gathered data from the U.S. Census Bureau's American Community Survey 5-Year Estimates, which provided median household income by zip code and gross rent as a percentage of household income. This dataset was crucial because it's one of the few sources that calculates rent burden directly and provides it at a granular geographic level. The Census data covers all five NYC boroughs and is regularly updated, making it ideal for understanding current affordability challenges.

After we established our data source, we did a deep dive into the available fields and determined what data stories we wanted to tell. We recognized that rent burden needed to be shown in multiple ways to capture both its geographic distribution and its relationship to income inequality. We developed four visualizations that work together to answer different questions: where is rent burden highest, how does it relate to income, and how do boroughs compare? We wanted this dashboard to convey that rent burden isn't just a citywide average but a geographically concentrated crisis that disproportionately affects low-income neighborhoods.

Our dashboard incorporates four distinct visualizations to provide a comprehensive view of rent burden across NYC neighborhoods and boroughs:

  1. Choropleth Map (Neighborhood Level): The first visualization we chose was a choropleth map showing rent burden by zip code. We wanted to provide a granular geographic view that shows exactly where housing affordability is worst, since rent burden varies dramatically even within the same borough. Rent burden is quantitative continuous data (percentages ranging from roughly 10% to 47%), so we chose a sequential color scale from light pink to dark red. This color scheme was deliberate because warm colors like red naturally convey intensity and urgency, which matches the stressful nature of spending too much on housing. Darker red immediately signals "crisis" to viewers, while lighter pink indicates more manageable burden. We used zip code boundaries as our geographic unit because they align with how most New Yorkers think about neighborhoods and provided the right balance between detail and readability.

  2. Scatterplot (Income vs. Rent Burden): In addition to showing where rent burden is highest, we wanted to reveal why it's happening by examining the relationship between income and housing costs. Therefore, we chose a scatterplot with median household income on the x-axis and rent burden on the y-axis. Each point represents a neighborhood, and we color-coded points by borough using our consistent color scheme (Bronx blue, Brooklyn orange, Manhattan red, Queens teal, Staten Island green). This visualization shows the clear inverse relationship between income and rent burden: as income decreases, the percentage spent on housing increases. The scatterplot format is ideal for showing correlation between two continuous variables and allows viewers to see both the overall trend and identify outliers or clusters.

  3. Bar Chart (Borough Averages): We recognized that while neighborhood-level data is detailed, some viewers need a simpler borough-level comparison to understand the big picture. Therefore, we chose a bar chart showing average rent burden by borough. Bar charts are perfect for comparing discrete categories (the five boroughs) on a single quantitative measure. The height of each bar uses position encoding, which humans perceive very accurately, making it easy to see that the Bronx has the highest average burden at over 33%. We maintained the same borough color scheme as the other visualizations to create visual continuity across the dashboard.

  4. Choropleth Map (Borough Level): Finally, we included a simplified choropleth map that aggregates rent burden at the borough level. This map uses the same light-to-dark sequential color scale but with larger geographic units, providing a high-level geographic overview that complements the detailed neighborhood map. This visualization serves viewers who want to understand general patterns without getting lost in neighborhood-level detail.

The consistent use of the light-to-dark color scale across the maps and the borough color scheme across all four visualizations creates a unified visual language. This repetition helps viewers build mental associations and reduces cognitive load when moving between visualizations. The dashboard layout places the two maps on the left to anchor viewers geographically, while the analytical charts on the right dive deeper into relationships and comparisons.