Importance of Visualization

• Clarity and Understanding: Visualizing information help individuals grasp

complex concepts more easily.
• Memory Enhancement: Visual content is more memorable than text. It is
easier for the brain to memorize information visually.
• Problem Solving: It helps in problem-solving by allowing individuals to
analyze information.
• Decision Making: By presenting the data in a visually appealing format,
decision-makers can quickly identify trends, patterns, and outliers.
• Creativity Enhancement: stimulates creativity by encouraging to think
visually and explore ideas.
• Collaboration and Teamwork: Visuals helps teams communicate ideas
and work together more effectively.

Explanatory Analysis:
• It aims to communicate insights and findings to others.
• It involves explaining some interesting observations.
• Focused on communicating insights to others
• Characteristics:
→ Emphasis on visualization and storytelling
→ Simplification of complex concepts
→ Hypothesis testing and confirmation
• Example: Creating a report or presentation for stakeholders to explain the
factors influencing sales and performance over the past quarter.

Exploratory Analysis:
• It focuses on uncovering patters, trends and insights within the data.
• Creating visuals to present your findings.
• Focused on understanding the data itself.
• Characteristics:
→ Emphasis on data exploration
→ Iterative and open-ended
 → Discovery-oriented
• Example: Analyzing customer survey data to identify potential factors
influencing customer satisfaction, without necessarily aiming to
communicate the findings to others immediately.

Classification Plots:
• Scatter plot:
→ Uses different colours or shapes for each class to visualize the
distribution of the data points.
→ Identify separability between classes.
• Box plot:
→ Visualize the distribution of a continuous variable
→ Useful for understanding the spread and central tendency.
• Histogram with Density Curve:
→ Compare the distribution of features between classes.
→ Helps identify difference on the distribution of features between
classes.
• Confusion Matrix Heatmap:
→ Visualize the performance of a classification model by displaying the
matrix as a heatmap
→ Helps identify the accuracy, precision, recall, and F1-score for each
class.

Regression Plots:
• Scatter Plot with Regression Line:
→ Relationship between the independent and dependent variable.
→ Overlay a regression line or curve to visualize trend
• Residual Plot:
→ Difference between observed and predicted values.
→ Helps to diagnose the assumptions of regression models.
• Predicted vs. Actual Plot:
→ Plot the predicted values against the actual values.
→ Helps asses the performance, identify patterns and outliers.
• Line Plot with Confidence intervals:
→ Plot the mean or median response variable against the independent
variable(s) with confidence intervals.
 → Visual representation of the uncertainty associated with the
regression estimates.

Tableau:
• Used for data visualization and dashboarding.
• Allows the users to create interactive and visually appealing charts, graphs
and dashboards.
• Advantages:
→ Drag-and-Drop interface for creating visualizations without coding.
→ Wide range of visualization options and customization features.
→ Can handle large datasets and real time data sources.
• Disadvantages:
→ Limited statistical analysis capabilities compared to specialized
statistical software like SPSS or R.
→ Steeper learning curve for advanced features and complex
visualizations.

SPSS (Statistical Package for the Social Sciences):

• Comprehensive statistical analysis software used for data manipulation,
descriptive statistics, hypothesis testing, and predictive modelling.
• Advantages:
→ Extensive range of statistical procedures and tests.
→ User friendly interface and point-and-click functionality.
→ Integration with syntax-based programming.
• Disadvantages:
→ Primarily focused on statistical analysis and lacks advanced
visualization.
→ Relatively expensive compared to other.

Excel:
• Widely used spreadsheet software with basic data analysis and
visualization capabilities.

• Advantages:
 → Familiar and widely accessible interface for data entry, manipulation,
and analysis.
→ Basic statistical functions and charting options for simple analyses and
visualization.
→ Integration with MS office apps and third party add-ins.
• Disadvantages:
→ Limited scalability for handling large datasets.
→ Lack of advanced statistical procedures and modelling capabilities.
→ Prone to errors and version control issues.

R:
• It is a programming language and environment for statistical computing
and graphics.
• It is widely used for data analysis, statistical modelling, ML and data
visualization.
• Advantages:
→ Extensive collection of packages
→ Highly customizable and flexible for complex analyses.
→ Strong community support and active development.
• Disadvantages:
→ Steeper learning curve, especially for users with less programming
knowledge.
→ Requires writing of code for analysis and visualization.
→ Handling large dataset may require optimization.

Green Pills and Blue Pills

• These refer to different types of fields or measures that you can drag and
drop onto the view to create visualizations.
• Green Pills:
→ Represents continuous fields or quantitative measures.
→ When we drag a green pill onto rows or columns shelf, Tableau will
create a continuous axis for that field.
→ Examples: numerical data such as sales amounts, quantities,
temperatures, etc.
• Blue Pills:
 → Represents discrete fields or dimensional attributes.
→ When we drag the blue pill onto the rows or columns, it will create
headers for discrete values.
→ Example: categorical data such as region names, product categories,
customer segments, etc.

Row Shelf:
• Located on the left side of the Tableau workspace.
• The row shelf organizes data along the vertical axis
• Fields placed on the row shelf are typically used to break down data into
categories or groups, such as regions, product names, or time periods.
• Multiple fields can be added to row shelf to create hierarchical structure.

Column Shelf:
• Located on the top side.
• The column shelf organizes data along the horizontal axis.
• Can add multiple fields to create more complex visualizations.

“Show Me” feature:

• A tool that helps users to quickly generate visualizations based on the data
fields available in the dataset.
• Offers dynamic gallery of recommended chart types.
How it works:
• Selecting Data Fields: After loading the dataset select some fields by
dragging and dropping.
• Using the “Show Me” Button: Once the fields are placed now, click on
“Show Me” button located at the top-right corner. (looks like a bar chart
icon with drop down arrow)
• Recommended Chart Types: Now we can see a gallery of chart types
displaying most suitable options related to selected fields.
• Selecting Chart Types: Browse and find the best fit charts for your data
and analysis needs.
Tooltip:
• A feature that provides additional information about data points or
elements.
• Useful to convey detailed information without cluttering.
• Hover interaction: Whenever a user hovers their mouse cursor over a data
point a tool tip appears automatically.
• Content: The content of the tooltip can be customized by the user.
• Customization: Provides options for customizing the appearance and
content of tooltips. Can control display of fields or measures.
• Interactivity: Increases interactivity by providing users with on-demand
information.
• Accessibility: Improves accessibility of visualizations by providing
additional context and descriptions.

Dimensions vs Measures
Dimensions:
• Represents categorical or qualitative data attributes.
• Describes the characteristics of the data and provide context or grouping
criteria.
• Examples, product categories, customer segments, regions, etc.
• Used to create discrete headers, labels or groupings.
• Displayed along the row or column headers and can be used to split or
partition the data.
• Blue pills.
Measures:
• Represent numerical or quantitative data attributes.
• Represent the measurable aspects of the data, such as quantities, amount,
or metrics.
• Examples, sales revenue, profit margins, quantities sold, average
temperatures, etc.
• Often used for calculations, aggregation, and statistical analysis.
• Displayed as continuous axes, bar lengths, or data points.
• Green pills.
Pivoting: Conversion of rows into columns and vice versa
Basic plot constructions and their advantages:
Scatter Plot:
• It is used for investigating relationships between quantitative variables,
such as age and income.
• Allows you to compare the values of 2 quantitative variables that you have
plotted.
• Can detect outliers and clusters.
• Provides a visual representation of the distribution of data points.
Line Plot:
• It connects a series of data points using a line.
• Used to show changes in data over time.
• Presents sequential values to help you identify trends.
• Helps identifying patterns, fluctuations, or trends over time.
• Provide clear representation of trends with continuous lines connecting
data points.
Bar Chart:
Compares numerical values like integers and percentages.
Shows variations in categories or subcategories scaling width or height.
Can highlight the larges or smallest number in a set of data or show
relationship between values.
Histogram:
Plots quantitative data with ranges of data grouped into bins.
Shows distribution of values.
Tracks the different values found in one set of data as a series of connected
bars.
Box Plot:
Enables to study the distributional characteristics and overall patterns of a
variable.
Provides useful ways to visualize the range and other characteristics of a
variable, such as median, quartiles, and outliers.
Can be created for numerical fata only.
Pie Chart:
Useful for summarizing categorical variable or dimensions.
Pie chart represents dimensional variables and its size represents the count of
each category.
Heatmaps:
Uses colour and size to help visualize data.
It consists of one or more dimensions and 1 or 2 measures.