Description
This was my first project for the "Introduction to Data Visualization" course taught by Professor Alberto Cairo. This project was specifically designed to practice the various techniques and skills I have learned in the course for presenting information and data in a visual and engaging manner. During each iteration of the project, I received feedback from Professor Cairo, which allowed me to grasp a deeper understanding on data visualization and create a finer end product.
Timeline
- Week 1: first draft
- Week 3: second draft
- Week 4: third draft
- Week 5: final draft
Requirements
- Client: The Economist
- Theme: Homelessness
- Type of infographic:
- Tabloid-size poster oriented horizontally
- At least 6-7 graphics
Topic Selection
I chose to explore the rising cost of living in Miami-Dade County in relation to homelessness as it aligns with the theme of "homelessness." The increase in the cost of living in this specific area is a critical factor that contributes to the homelessness issue. Understanding how the cost of housing has escalated and its direct impact on homelessness provides valuable insights into the complex problem of people losing their homes and the challenges they face. I have a desire to understand this issue in the county where I currently reside.
Project Hypothesis / Research Question
Does the escalating cost of living in Miami-Dade County contribute to the increase in homelessness in the county?
Data Sources
Data Analysis Process
Once I gathered the necessary data for this infographic, I conducted EDA to search for connections, identify patterns, outliers, missing data, and evaluate the the initial encodings’ effectiveness. The data analysis process for this infographic utilized data on rental price trends, household income, overcrowded homes, and apartment purchase prices dynamics. The analysis focused on identifying key indicators, such as rising rental costs and the percentage of household income spent on housing expenses, as primary drivers of housing affordability challenges, which are in turn linked to the issue of homelessness. The tools used during this process were R, Excel, and Flourish.
Final Graphics Selection
Choropleth map
The choropleth map used in the infographic conveys essential information about homelessness in Florida. Choropleth maps are particularly well-suited for this purpose as they allow for the representation of data variations across geographic regions. In this case, the map effectively illustrates the rates of homelessness in Florida by county, highlighting areas with a higher concentration of homeless individuals. This chart emphasizes the homelessness rates across the counties in Florida. By using color gradients and shading to denote varying rates of homelessness, the choropleth map provides viewers with a quick and clear understanding of the severity of the issue in different counties. The magnifying glass added to the choropleth map directs the viewers’ focus towards Miami-Dade county.
Time-series chart
For the presentation of data on the cost of renting a home and the cost of purchasing an apartment over time, the time-series charts are chosen as they visualize trends and changes over time. These charts provide a concise and effective means of illustrating how housing costs have changed, helping viewers to understand the dynamic nature of housing affordability in Miami-Dade County. For both of these charts, colors are used to differentiate downtown and non-downtown properties.
Similarly, the time-series chart comparing the Miami homeless population to the Florida homeless population illustrates the relative scale of homelessness within the county over time.
Bar chart
The bar chart in this infographic is an addition to the choropleth map. It displays sixteen Florida counties with the most number of homeless individuals. A bar chart is effective for showing and comparing discrete data points, making it a practical choice to rank counties based on the number of homeless individuals. Miami-Dade’s bar is the sole red bar in the chart to highlight its high number of homeless individuals and to give emphasis on the county since the infographic is about Miami-Dade.
Dot matrix chart
Dot matrix charts are excellent for showing the distribution and comparison of categorical data, making them ideal for presenting the clear contrast between sheltered and unsheltered individuals in Miami-Dade county. By using dots to represent 100 homeless individuals, the chart allows viewers to discern the relative proportions of sheltered and unsheltered individuals in a visually intuitive manner. Additionally, the side-by-side presentation of data for the two years, 2021 and 2022, enables viewers to immediately identify any shifts or changes in the homeless population's housing status at that time period. Colors are used to differentiate the categories, sheltered and unsheltered individuals.
Slope chart
Slope charts are particularly effective in visually depicting trends and changes over time, making them a suitable choice to show the percentage change in overcrowded homes during the specified period. By using lines that connect the two data points, the chart provides a clear and direct representation of the direction and slope of change, allowing viewers to assess the progression or regression in overcrowding rates over these years. Out of the three categories (renters, owners, and overall) presented in the slope chart, overall is colored red to create emphasis in that category.
Pie chart
Pie charts are known for their capability to illustrate part-to-whole relationships, making them an ideal selection to portray the groups of households based on their rent expenditure as a percentage of income. This chart conveys the proportion of households facing a significant housing cost burden (spending over 30% of their income on housing expenses) compared to those with more affordable housing costs (spending less than 30% of their income on housing expenses). Moreover, the use of distinct colors to differentiate the two categories helps to distinguish the segments.
Sketch
During the sketching phase, I started with visualizing my project's layout and structure. I began by exploring and selecting the most appropriate data visualization types to effectively convey the information I’ve gathered. I created rough sketches that included basic shapes and labels to represent various elements of the infographic. This process was essential and allowed me to experiment with visual layouts and organization, and map out the infographic's structure and divide it into distinct sections, ensuring that the content flowed logically.
First Draft
The first draft was created based on the initial sketch. Since it is a rough first draft, there are a lot of changes that could be made. At the beginning of the design process, I focused on creating a theme that reflects the client, The Economist. The colors I chose to fit the theme are the blue and red from The Economist’s color palette. Once that was set, I used the initial sketch as a guideline to position the charts. Since it is early in the process, I used filler texts in place of the actual text to have an idea of the space I am working with.
Feedback
Some of the feedback I received based on this iteration include the inconsistent font sizes and white spaces, the dissimilarity of the labels positioning in my infographic to The Economist’s style. In addition, the charts were not telling the story as well as it needs to.
Second Draft
The second draft was created with the feedback received from the first iteration in mind. The position of the text has been moved, filling the first column and the space below the charts in the second and third columns of the first page. The charts have been modified to tell the story better. The charts about rental costs, apartment purchase prices, and household income are kept. The household income chart has been modified from a bar chart to a tree map. Four charts with more relevant data were added as seen below. I fixed the white spaces in the infographic to make it cleaner. I changed the text for each chart to be shorter and more concise, following The Economist’s style.
Feedback
The feedback I received for this iteration is to change the type of chart used to visualize the household income, add a map into the infographic, reduce text, removing the chart about insurance, and use a time-series chart instead of a bar chart.
Third Draft
Just as the second draft, I kept the feedback received from the second iteration in mind. Based on the feedback, I added a choropleth map in the first page of the infographic and a bar chart to complement it, displaying the number of homeless individuals in each Florida county. The dot matrix chart in the previous iteration is moved to the first page and rotated. The charts displaying the cost of renting a home and the cost of owning an apartment are now stacked on top of each other since they have the same x-axis. I added rental cost vs median household income ratio and apartment purchase price vs median household income ratio bar charts. These bar charts’ purpose is to supplement the time-series charts to their left. They emphasize the affordability of renting or owning homes for the households whose income falls within the median. Below these charts, I added a time-series chart, slope chart, and pie chart to show the Miami-Dade county’s homeless population over Florida’s homeless population, change in overcrowded households from 2017 to 2021, and the percentage of households who spend over 30% of their income on rent. In addition, I took time to modify the text in the infographic to describe the discoveries from the data. I selected a different blue and red colors from The Economist’s color palette to make the infographic easier on the eyes.
Feedback
The feedback I received for this draft is a reminder that a choropleth map is used to visualize rates instead of count and to give more attention to details in order to finalize the infographic. One example of the details needed to be fixed is the x-axis labels font size for the counties with most homelessness bar chart.
Final Draft
In this final iteration, I focused on the details to make sure that texts, white spaces, etc. are uniform. I adjusted the font size of the x-axis labels of the counties with most homelessness bar chart to match the rest of the x-axis labels in the infographic. I modified the choropleth map to display rates of the homeless population in Florida by county instead of the count of homeless individuals. Furthermore, I altered the title and subtitles of the charts to better explain the data.
Conclusion
Through this project, I acquired a range of valuable skills and insights. I learned to employ various data visualization techniques, enabling me to effectively convey different aspects of data. Additionally, I developed a deeper understanding of data analysis, data-driven storytelling, and the importance of incorporating iterative feedback to enhance infographic clarity and storytelling. Thematic design and establishing visual hierarchy were key lessons, as were the ability to select appropriate charts for the data gathered. This project enhanced my communication skills, enabling me to convey intricate data concepts concisely and engagingly to viewers. My attention to detail was challenged since the details that may seem unimportant, could affect the quality of the infographic.
Future Work
- Consistency: Through this project, I've come to realize the importance of consistency in design, ensuring that colors, fonts, and styles align with the chosen theme. In future projects, I will apply this lesson by maintaining consistency of things that are alike, creating visuals that are not only informative but also visually cohesive.
- Attention to detail: The project emphasized the significance of attention to detail, from font sizes to white spaces and text clarity. This experience has increased my awareness of the subtle elements that can greatly impact an infographic's quality. In future projects, I will be more vigilant about these finer details, ensuring a polished and professional final product.
- Chart selection: This project taught me the value of selecting the most appropriate chart types for different data sets. As I move forward, I will leverage this knowledge to make more informed chart selections, tailoring them to each project's specific data and goals.
