MTH130 Session 5: Data Visualization

• Session 5 of MTH130 introduces data-driven decision-making through data visualization.
• The objective is to learn about visualization, discover different types of visualizations, and understand their advantages and limitations.
• As Ben Schneiderman said, "Visualization gives you answers to questions you didn't know you had."

Recap of Statistical Concepts

• Statistics can be descriptive or inferential.
• Analysis can be performed using the population (N) or a sample (n).
• Inferential analysis involves making inferences about a population based on a selected sample.
• The sample frame refers to the accessible population.
• The sample is the actual unit under study.

Recap - Sampling Methods

• Sampling methods shown are simple random sampling and systematic sampling.
• Simple Random Sampling: Randomly selecting numbers from the population.
• Systematic Sampling: Sampling every 3rd person from the population.
• Stratified sampling is shown:
  • Dividing the population into strata.
• Cluster sampling is displayed:
  • Dividing the population into clusters.

Introduction to Data Visualization

• Data visualization aims to graphically represent information and data.
• It utilizes visual elements like charts, graphs, and maps.
• Data visualization provides accessible ways to understand trends, outliers, and data patterns.
• Complex data can be presented to non-technical audiences to avoid confusion.
• In the world of Big Data, visualization is essential for analysis of massive amounts of information and in supporting data-driven decisions.
• Key tools include charts, graphs, and mapping tools.
• Visualization enables informed decision-making and enhances communication of insights to diverse audiences

Types of Data Visualization

• These are some of the most often used Data Visualizations
• Bar Chart:
  • Uses rectangular bars to represent data values, great for comparing categories of data.
• Line Chart:
  • Displays information as a series of data points connected by straight line segments, effective for displaying trends over time.
• Pie Chart:
  • A circular statistical graphic divided into slices to illustrate numerical proportions, showcases parts of a whole or percentage breakdowns.
• Histogram:
  • A graphical display of data using bars to represent the frequency of continuous data values, useful for showing the distribution of a dataset.
• Scatter Plot:
  • Uses dots to represent values for two different numeric variables, effective for showing relationships or correlations between variables.
• Heat Map:
  • A graphical representation of data where values are depicted by colors on a matrix, suitable for displaying large volumes of data and highlighting trends.
• Venn Diagram:
  • A diagram representing sets with overlapped circles, showing relationships between the sets, Ideal for showcasing relationships or commonalities between different data sets
• Bubble Chart:
  • Represents data points in the form of bubbles, where the size of the bubble indicates a value, effective for showcasing three dimensions of data: x-axis, y-axis, and size.
• Choropleth Map:
  • A map using different shading or coloring to indicate various values in different areas, excellent for geographical data visualization.
• Sankey Diagram:
  • A visual tool that displays flows of energy, materials, or costs, perfect for illustrating the flow from one set of values to another, e.g., tracking user behavior on a website.

Historical Example: 1854 Broad Street Cholera Outbreak Map

• John Snow's map is an early example of dot map visualization.
• Early dot map visualization.
• The map depicts cholera deaths using small bar graphs on city blocks in a London neighborhood.
• It investigates the concentration and distribution of cholera deaths to identify a higher death trend in specific city blocks.
• It revealed that households with the most cholera deaths shared a common well for drinking water and highlighted the well's contamination by sewage as cholera outbreak's root cause.
• This was a groundbreaking revelation at the time as it connected cholera outbreaks in London to contaminated water wells.
• The visualization inspired proposals for constructing sewage systems and measures to protect wells from contamination.
• The early use of dot map visualization was innovative for its time.
• The Cholera map was successful in uncovering the source of the problem, illustrating the power of pushing boundaries in data visualization to discover valuable insights.