Chapter 3 Business Analytics

Data visualization

A visual presentation of data used to communicate information clearly and efficiently.

Cognitive load

The effort needed to accurately and efficiently process information communicated by a data visualization.

Preattentive attributes

Features of a visualization that help reduce the cognitive load needed to interpret it.

Main preattentive attributes

Color, size, shape, and length.

Color

A preattentive attribute that is effective for differentiating data, but overuse can become distracting.

Size

The relative amount of space an object occupies in a visualization.

Length

The horizontal, vertical, or diagonal distance of a line or bar/column; useful for showing quantitative values.

Data-ink ratio

The proportion of ink used for data compared with the total amount of ink in a table or chart.

Data-ink

The ink in a table or chart that is necessary to convey the meaning of the data.

Non-data-ink

Ink in a table or chart that serves no useful purpose in conveying the data.

Decluttering

The process of increasing the data-ink ratio by removing unnecessary visual elements.

White space

The portion of a data visualization that is devoid of markings.

When should a table be used?

When readers need specific numerical values, precise comparisons, or when values have different units or very different magnitudes.

When should a chart be used?

When the goal is to make patterns, trends, or relationships easier to see rather than focus on exact numerical values.

Chart and table together

Useful when you want readers to see patterns or changes while also being able to reference exact numerical values.

Crosstabulation

A tabular summary used to describe data for two variables.

PivotTable

An Excel crosstabulation used to summarize and analyze data.

PivotChart

A chart paired with a PivotTable that helps summarize and analyze data.

Scatter chart

A graphical presentation of the relationship between two quantitative variables.

When is a scatter chart used?

To examine the relationship between two quantitative variables.

Line chart

A chart useful for representing time-series data.

When is a line chart used?

To show changes or trends in data over time.

Bar chart

A chart using horizontal bars to display the magnitude of a quantitative variable.

Column chart

A chart using vertical bars to display the magnitude of a quantitative variable.

Bar vs. column chart

Bar charts use horizontal bars; column charts use vertical bars.

Sorted bar/column chart

A bar or column chart where categories are ordered based on their quantitative values.

Clustered chart

A bar or column chart that displays multiple categories side-by-side for comparison.

Stacked chart

A bar or column chart that stacks multiple categories on top of or beside each other to show their combined total and components.

Clustered vs. stacked charts

Clustered charts make comparing categories easier, while stacked charts emphasize totals and composition.

Limitation of clustered charts

They can become cluttered when there are more than a few variables per category.

Pie chart

A chart commonly used to compare categorical data.

Problem with pie charts

It can be difficult to compare the relative sizes of pie wedges, and color coding may be difficult for some viewers.

3-D charts

Charts using three-dimensional effects that can make comparisons more difficult and are generally poor choices for data visualization.

Trendline

A line added to a scatter chart or line chart to show the general trend in the data.

Bubble chart

A visualization that displays three variables in a two-dimensional graph.

How does a bubble chart display three variables?

Two variables are represented by the x- and y-axes, while the third variable is represented by bubble size.

Scatter-chart matrix

A visualization that allows comparison of relationships among multiple variables.

Table lens

A visualization used to examine relationships between pairs of variables.

PivotCharts and PivotTables

They can be used together to summarize and analyze data using a crosstabulation and chart.

Heat map

A two-dimensional graphical representation that uses different shades of color to indicate magnitude.

Hierarchical data

Categorical data that are further decomposed into subcategories.

Tree structure

A visualization of hierarchical data where branches lead to categories and subcategories.

Treemap

A chart using the size, color, and arrangement of rectangles to display a quantitative variable for different categories and subcategories.

Waterfall chart

A chart that displays the cumulative effect of changes on a variable of interest.

Waterfall chart columns

The magnitude of each change is represented by a column anchored at the cumulative height of previous changes.

Stock chart

A graphical display of stock prices over time.

High-Low-Close stock chart

A stock chart that displays the highest, lowest, and closing prices for a period.

Parallel-coordinates plot

A chart useful for examining data with more than two variables.

Parallel-coordinates plot lines

Each observation is represented by a line connecting the vertical axes, with the height showing the value for each variable.

Choropleth map

A geographic visualization that uses shades of color, different colors, or symbols to show quantitative or categorical geospatial data.

Choropleth map example

Colors can represent values such as temperature, with warmer colors showing higher values and cooler colors showing lower values.

Cartogram

A map-like diagram that uses geographic positioning to represent map regions without necessarily preserving land area.

Cartogram example

A state’s area may be sized according to its population rather than its actual geographic area.

Weakness of a cartogram

Distortions caused by relative sizing can make the geography difficult or meaningless to interpret.

Equal-area cartogram

A cartogram designed to provide a balanced visual representation while maintaining fidelity to relative geographic positioning.

Are cartograms available in Excel?

No.

Key performance indicator (KPI)

A metric that indicates current operating characteristics.

Data dashboard

A data visualization tool that displays multiple KPIs and automatically updates as new data becomes available.

Effective dashboard principle

Show on one screen all KPIs related to one aspect of operations.

Dashboard scrolling principle

Create multiple dashboards of related KPIs to prevent excessive scrolling.

Good dashboard KPIs

KPIs should have clear meaning and be related to decisions.

Dashboard color principle

Use color to call attention and differentiate between categorical variables.

Purpose of data dashboards

To summarize and communicate a firm's operations visually using multiple KPIs.

Preattentive attributes and cognitive load

Preattentive attributes reduce the cognitive effort needed to interpret a visualization.

High data-ink ratio

A visualization contains a high proportion of useful data-related ink compared with unnecessary visual elements.

Low data-ink ratio

A visualization contains more unnecessary visual elements relative to the information being communicated.

Table vs. chart

Tables are better for precise values and comparisons; charts are better for visually communicating patterns, trends, and relationships.

Scatter chart variables

Two quantitative variables.

Line chart data

Time-series data.

Bar/column chart purpose

Display the magnitude of a quantitative variable across categories.

Bubble chart variables

Three quantitative variables.

Heat map purpose

Show magnitude using different shades of color.

Treemap purpose

Display quantitative values across categories and subcategories using rectangle size, color, and arrangement.

Waterfall chart purpose

Show the cumulative effect of changes.

Stock chart purpose

Display stock prices over time.

Parallel-coordinates plot purpose

Examine data involving more than two variables.

Choropleth map purpose

Display geospatial data using color, shades, or symbols.

Cartogram purpose

Display geographic regions while sizing them according to another variable such as population.

Dashboard purpose

Display and monitor multiple KPIs in one visual tool.