Data Visualization Basics

Data Visualization

• Data visualization is the process of displaying data in graphical format, often charts or graphs, to aid understanding and analysis.
• Visualizations improve data comprehension and communication.
• Visualizations enhance interactivity and the ability to convey information effectively to users.

Data Visualization Process

• Data sources are the origin of the information for analysis.
• Ingestion is the process of gathering and importing data.
• Storage involves organizing and preserving the data.
• Processing involves manipulating and organizing the data to prepare for analysis and visualization.
• Analysis and Visualization focuses on gaining insights from the processed data and transforming it into graphical representations.

Questions for Data Visualization Design

• What specific analyses are necessary to gain valuable insights?
• What conclusions can be gleaned from the data?
• Which visualization method best illustrates the insights?
• Should interactive dashboards or reports be generated for user interaction?

Data Visualization Examples

• Bar Charts: Useful for displaying absolute values, suitable for comparing categories or groups (like showing sales figures across different regions).
• Scatter Plots: Demonstrate relationships between two continuous variables visualized by plotted points (plotting student scores on math vs. English to visualize correlations).
• Connected Scatter Plots: Depict the progression of connections to show trends in relationships (a chart demonstrating how customer satisfaction relates to profitability over time).
• Bubble Charts: Useful for showing the relationship between three variables (like illustrating sales figures, product types, and customer location).
• Word Cloud Charts: Highlight prominent words from a text corpus, highlighting important themes (showing keywords related to customer feedback).

Visualizing Aggregate Data

• Pie Charts: Represent portions of a whole, often depicting the percentage or proportion of different categories (illustrating the proportion of customers who prefer different products).
• Donut Pie Charts: A variation of pie charts, with a central empty area, offering a more visually appealing representation.
• Heat Maps: Visually represent data using shades to signify values, highlighting trends in data such as user activity.
• Stacked Column Charts: Ideal for comparing components or subcategories within a category (like showing the composition of different customer segments).
• Tree Maps: Display hierarchical structures through nested rectangles, useful for visualizing part-to-whole relationships, showing hierarchical categories of sales regions).

Visualizing Data Distribution

• Histograms: Illustrating variable distribution, showing the frequency of data within defined ranges (example: the distribution of ages of customers at stores).
• Box Plots: Show data quartiles (minimum, first quartile, median, third quartile, and maximum) helping to identify central tendencies and outliers (highlighting age distribution within customer cohorts).
• Violin Plots: Showcase data distribution with kernel density plots, providing a comprehensive view of data variability.
• Density Plots: Show continuous distributions using smooth curves.