6.2: Distinguishing Among Chart Types Flashcards
What are the four key questions one should consider when determining the best method for visualizing results?
What business question are you trying to answer?
Who is the target audience for the visualization?
Which type of data is being visualized, categorical or numerical?
What type of analysis have you performed, exploratory or confirmatory?
Define exploratory business analytics and provide examples of activities involved in this type of analysis.
Exploratory business analytics involves initial descriptive and diagnostic analytical investigations to summarize and explain performance.
Activities in exploratory analytics include describing past performance, exploring summary performance statistics, identifying anomalies and outliers, checking assumptions, and detecting patterns using historic data.
It generates questions for further investigation and hypotheses for testing.
What is confirmatory business analytics, and what is its primary focus?
Confirmatory business analytics is conducted after the data analyst understands the relationships in the data. Its primary focus is to use statistics to judge the likelihood (probability) of a future event or outcome occurring.
This type of analysis includes predictive and prescriptive analytics.
Why is understanding the type of analysis performed (exploratory or confirmatory) crucial in choosing an appropriate data visualization method?
Understanding the type of analysis (exploratory or confirmatory) is crucial because it determines the focus and purpose of the visualization.
Exploratory analysis requires visualizations that aid in exploring patterns, anomalies, and summarizing past performance, whereas confirmatory analysis needs visualizations that support predictions and probabilities for future events or outcomes.
Choosing the right visualization method aligns with the specific analytical goals, ensuring effective communication of insights to the intended audience.
What are the three most frequently used types of charts for visualizing categorical data, and how do they represent the data?
The three most frequently used types of charts for visualizing categorical data are:
Pie charts: Circular graphs where each slice represents the category’s proportion of the whole dataset.
Bar charts: Data visualizations showing categorical data using vertical/horizontal bars with varying heights/lengths proportional to the values they represent.
Stacked bar charts: Bar charts with bars subdivided into different categories, representing proportions within each category.
Why are bar charts often considered easier to interpret than pie charts for visualizing categorical data?
Bar charts are often considered easier to interpret than pie charts because our brains are better at comparing the height of columns (or lengths of horizontal bars) than comparing the sizes of pie slices, especially when proportions are relatively similar.
Bar charts provide a clearer visual representation of the differences in proportions between categories.
In the context of categorical data visualization, how does adding percentages to a pie chart improve its interpretability?
Adding percentages to a pie chart helps improve interpretability by providing numerical context to the proportions represented by each slice.
It makes it easier for viewers to understand the exact proportions of each category relative to the whole dataset.
However, even with percentages, bar charts are often preferred as they allow for a quicker and more intuitive comparison of proportions.
What does a stacked bar chart represent, and how does it differ from a regular bar chart?
A stacked bar chart represents categorical data using bars that are subdivided into different categories.
Each bar is segmented into sections, each representing a specific category, and the total length of the bar represents the whole dataset.
Unlike a regular bar chart where each bar represents a single category, a stacked bar chart shows the composition of each category in relation to the total, allowing for a detailed comparison of proportions within categories.
What is the recommended rule of thumb regarding the number of categories for which pie charts are suitable, and why is this guideline important?
The recommended rule of thumb is to avoid using pie charts when there are more than five categories.
This guideline is important because as the number of categories increases, it becomes increasingly difficult to distinguish between the size of the slices, especially when there are many skinny slices of pie. Pie charts with numerous categories can be challenging to interpret accurately.
What does a 100 percent stacked bar chart represent, and how does it differ from a regular stacked bar chart?
A 100 percent stacked bar chart represents the proportion of each category as a percentage of the whole dataset.
It differs from a regular stacked bar chart in that the total height of each bar in a 100 percent stacked bar chart always equals 100%, making it easier to compare the relative proportions of categories across different bars.
Provide examples of situations where stacked bar charts are particularly useful for data visualization.
Stacked bar charts are particularly useful when you need to:
Compare parts of the whole over time: For example, comparing product sales by type over multiple years (e.g., 2025 vs. 2026 sales).
Analyze data across different locations: For instance, comparing product sales at different branches or stores within a company.
Visualize proportions: Showing how different categories contribute to the whole, such as the distribution of product sales among various types of products.
What is a tree map, and in what situations is it useful for data visualization?
A tree map is a visualization technique that utilizes size and color to represent the proportional size of values using physical space.
Tree maps are useful when there are many categories that need to be visualized.
They are helpful in revealing patterns and identifying outliers.
However, they are not ideal for representing precise numbers or proportions.
When is a symbol map useful, and what does it represent in the context of data visualization?
A symbol map is useful for expressing categorical data proportions across geographic areas such as states or countries.
It represents the relative proportions of categorical data in different geographic regions.
Symbols, like circles in the example, are used to visualize the data, with varying sizes indicating different levels of the data variable.
Symbol maps are effective for comparing relative proportions across different geographic regions but may lack precision for detailed analysis.
How does a word cloud work, and in what scenario is it particularly beneficial for data analysis?
A word cloud is created based on the frequency of words mentioned in a text data set.
The higher the frequency of a word, the larger and bolder it appears in the word cloud.
Word clouds are useful for identifying commonly used words in text data, making them beneficial for analyzing open-ended responses in surveys or textual content.
They help quickly identify prevailing themes or sentiments within the text. Common English words can be filtered out to focus on significant terms.
What is a heat map, and how does it represent categorical data? Provide an example scenario where a heat map is advantageous.
A heat map is a graphical representation of categorical data that uses different colors or shades of the same color to indicate different values.
Darker colors represent lower values, while lighter colors represent higher values.
Heat maps are advantageous for comparing the intensity of a phenomenon across different categories.
For example, a heat map can display weekday unit sales by employee, where darker red indicates fewer units sold and darker blue indicates more units sold.
Heat maps allow for quick visual comparison of performance metrics, making them useful for data-driven decision-making.
Figure: Summery of Chart Types
