Design Rationale

Process of Designing the Dashboard

We started this dashboard by identifying data sources that could provide both current distribution and historical trends of rent-stabilized housing across New York City. We gathered data from the NYC Rent Guidelines Board's annual Income and Affordability Study reports, which provide comprehensive borough-level statistics on rent-stabilized units from 2002 to present. While these reports are published as PDFs rather than CSV files, we were able to extract the data tables to create temporal datasets. We supplemented this with building-level data to create geographic visualizations showing where rent-stabilized housing is concentrated across neighborhoods.

After we established our data sources, we did a deep dive into the available information and determined what data stories we wanted to tell. We recognized that understanding rent stabilization requires answering multiple questions: Where do rent-stabilized units still exist? How much has the stock declined over time? What types of regulated housing exist and where? How do gains compare to losses each year? We developed five visualizations that work together to answer these questions from both geographic and temporal perspectives. We wanted this dashboard to convey that rent stabilization is both a disappearing resource and an unevenly distributed one, with some neighborhoods retaining strong protections while others have lost nearly all their stabilized housing.

Our dashboard incorporates five distinct visualizations to provide a comprehensive view of rent-stabilized housing distribution and trends across NYC:

1. Choropleth Map (Density Per 1,000 Households): The first visualization we chose was a choropleth map showing rent-stabilized units per 1,000 households by neighborhood. We wanted to show not just where stabilized housing exists in absolute numbers, but where it's most available relative to population size. Normalizing by population is crucial because a neighborhood with 5,000 stabilized units might sound impressive, but if it has 50,000 households, only 10% are protected. The per-capita metric is ratio data with a meaningful zero point, and we used a sequential blue color scale from light to dark. We chose blue rather than the red scale used for rent burden because we wanted to distinguish between the two dashboards thematically and because blue conveys stability and protection, which aligns with what rent stabilization provides. Darker blue indicates higher density of protected housing, making the neighborhoods with the strongest protections immediately visible.

2. Stacked Bar Chart (Regulated Housing Composition): In addition to rent stabilization, we wanted to show the broader landscape of regulated housing, including rent control and non-regulated units. Therefore, we chose a stacked bar chart that shows the total housing units in each borough broken down by regulation type. This chart uses categorical data for boroughs on the x-axis and quantitative data (unit counts) on the y-axis. The stacking technique allows viewers to see both the total housing stock and the proportional breakdown simultaneously. We used three distinct colors: orange for rent-stabilized (matching our borough color for Brooklyn to maintain some consistency), red for rent control, and teal for non-regulated units. The stacked format makes it easy to compare both absolute numbers (total bar height) and proportions (segment sizes) across boroughs.

3. Line Chart (Total Units Over Time): We recognized that showing the overall trend of rent stabilization loss was critical to understanding the crisis. Therefore, we chose a line chart tracking total citywide rent-stabilized units from 2002 to 2024. Time is the continuous variable on the x-axis and unit count is the quantitative continuous variable on the y-axis. The line chart is ideal for showing change over time because the connected line creates a clear visual path that emphasizes the trajectory. The downward slope immediately communicates decline without requiring viewers to read specific numbers. We used a simple blue line matching our stabilization color theme to maintain visual consistency. The minimal design adheres to data-ink ratio principles by removing unnecessary gridlines and decorative elements.

4. Diverging Stacked Area Chart (Gains vs. Losses): In addition to showing net change, we wanted to reveal the underlying dynamics of how units are both gained and lost each year. Therefore, we chose a diverging stacked area chart that separates gains (shown in green above the zero line) from losses (shown in red below). This visualization uses the zero line as a critical reference point, with positive and negative values representing opposite outcomes. The area chart format shows cumulative patterns over time, and the diverging color scheme (green for positive, red for negative) leverages universal color associations. We overlaid a blue line showing net change, which allows viewers to see both the individual components and the overall result. This visualization type is more complex but reveals patterns that would be invisible in a simple line chart, such as years when gains increased but losses increased even more.

5. Choropleth Map (Geographic Distribution by Borough): Finally, we included a building-level map showing the spatial distribution of all rent-regulated apartments, color-coded by borough. This map uses our consistent borough color scheme (Bronx blue, Brooklyn orange, Manhattan red, Queens teal, Staten Island green) to create immediate visual recognition. Unlike the first map which uses shading intensity to show density, this map uses categorical color encoding and individual building points to show exact locations. This visualization serves a different purpose: rather than showing "how much" protection exists, it shows "where exactly" protected buildings are located. The point-based approach reveals clustering patterns and helps viewers understand which specific neighborhoods within each borough contain regulated housing. The scatter of points makes it clear that rent-stabilized housing isn't evenly distributed but concentrated in specific corridors and neighborhoods.

The dashboard layout places geographic visualizations on the left and temporal visualizations on the right, creating a spatial-temporal division that helps viewers understand both where rent stabilization exists and how it has changed. The consistent use of blue for rent stabilization across multiple visualizations creates thematic unity, while the borough color scheme maintains continuity with the other dashboards in the project.