Solved Exercises on Choosing Appropriate Displays for Quantitative Data

Master data visualization: learn to select the best display for quantitative data through these 5 detailed exercises covering dot plots, histograms, box plots, and more.

Solution: Exercises 1 to 3
1 Dot Plot vs Histogram Decision
Exercise 1
You have the following data on the number of pets owned by 15 students: 0, 1, 1, 2, 2, 2, 2, 3, 3, 3, 4, 4, 5, 5, 6. Which display would be most appropriate - a dot plot or a histogram? Explain your reasoning.
Definition:

Dot plot: A graphical display that shows each data point as a dot above a number line. Best for small datasets with repeated values.

Histogram: A graphical display that uses bars to show the frequency of data within intervals. Best for larger datasets or continuous data.

Decision criteria:
  1. Consider the number of data points (sample size)
  2. Assess the range and distribution of values
  3. Evaluate whether individual values are important
  4. Determine if data is discrete or continuous
Step 1: Analyze the data characteristics

Sample size: n = 15 (relatively small)

Range: 0 to 6 pets (small range)

Type: Discrete data (number of pets)

Repetition: Some values appear multiple times (2 appears 4 times)

Step 2: Evaluate dot plot advantages

• Shows individual data points clearly

• Preserves exact values

• Good for small datasets (n ≤ 30)

• Easy to identify modes and clusters

Step 3: Evaluate histogram disadvantages for this data

• Would require grouping data into intervals

• Would lose information about individual values

• With only 7 unique values, grouping is unnecessary

Step 4: Make the decision

Dot plot is more appropriate because:

• Sample size is small (15 data points)

• Data has repeated values that can be easily visualized

• Exact values are important to preserve

Step 5: Conclusion

A dot plot will clearly show that 2 pets is the mode and allow easy identification of all data points.

Dot plot is most appropriate for this data
Final answer:

A dot plot would be most appropriate for this data. The sample size is small (n=15), the range is limited (0-6), and individual values are important to preserve. A dot plot will clearly show the frequency of each value and make the mode (2 pets) immediately apparent.

Applied rules:

• Small datasets: Use dot plots for n ≤ 30

• Discrete data: Dot plots preserve exact values

• Individual values: Use when preserving each data point matters

2 Histogram vs Box Plot Comparison
Exercise 2
You have test scores for 200 students ranging from 45 to 100. Which display would be better - a histogram or a box plot? What are the advantages of each for this scenario?
Definition:

Histogram: Shows the distribution of data using bars, revealing shape, center, spread, and potential outliers.

Box plot: Displays the five-number summary (min, Q1, median, Q3, max) and outliers using a box and whisker format.

Step 1: Analyze the data characteristics

Sample size: n = 200 (large dataset)

Range: 45 to 100 (wide range)

Type: Continuous data (test scores)

Purpose: Understanding overall distribution

Step 2: Evaluate histogram advantages

• Shows the shape of the distribution (symmetric, skewed)

• Reveals patterns and clusters in the data

• Good for large datasets

• Shows modality (unimodal, bimodal)

Step 3: Evaluate box plot advantages

• Shows the five-number summary concisely

• Excellent for comparing distributions

• Clearly identifies outliers

• Takes up less space

Step 4: Consider the primary goal

For understanding the overall distribution shape and identifying patterns, a histogram is better.

For comparing to other groups or focusing on summary statistics, a box plot is better.

Step 5: Make recommendation

For this scenario, a histogram is recommended to show the distribution shape and patterns in 200 test scores.

Histogram is better for distribution shape; box plot for summary stats
Final answer:

A histogram would be better for this scenario because it shows the distribution shape, reveals patterns in the data, and works well with large datasets. While a box plot is excellent for comparing groups and showing outliers, a histogram provides more detailed information about the distribution of 200 test scores.

Applied rules:

• Large datasets: Histograms work better than dot plots

• Distribution shape: Use histograms to show skewness and modality

• Comparison: Use box plots when comparing multiple groups

3 Stem-and-Leaf Plot Analysis
Exercise 3
For the following data representing commute times (minutes): 23, 27, 31, 32, 34, 36, 38, 40, 42, 45, 47, 51, 53, 56, 62, 65, 68, 71, 75, 78, evaluate whether a stem-and-leaf plot would be appropriate. Create the plot if appropriate.
Definition:

Stem-and-leaf plot: A graphical display that organizes data by separating each value into a "stem" (leading digit(s)) and "leaf" (final digit). Preserves original data values while showing distribution.

Step 1: Analyze data characteristics

Sample size: n = 20 (moderate size)

Range: 23 to 78 minutes

Type: Two-digit numbers with meaningful tens and units

Preservation: Need to preserve exact values

Step 2: Determine stem and leaf separation

Stems: Tens digits (2, 3, 4, 5, 6, 7)

Leaves: Units digits

Step 3: Create the stem-and-leaf plot
Stem | Leaf
  2  | 3 7
  3  | 1 2 4 6 8
  4  | 0 2 5 7
  5  | 1 3 6
  6  | 2 5 8
  7  | 1 5 8
                
Step 4: Evaluate appropriateness

Advantages: Preserves exact values, shows distribution, compact format

Appropriate because: Moderate sample size, data naturally splits into stems/leaves

Step 5: Interpret the plot

The distribution appears roughly symmetric with most values in the 30s and 40s

No obvious outliers

Stem-and-leaf plot created successfully
Final answer:

A stem-and-leaf plot is appropriate for this data. The plot shows that commute times are mostly concentrated in the 30-50 minute range, with a roughly symmetric distribution. The plot preserves exact values while showing the overall distribution.

Applied rules:

• Two-digit data: Use stem-and-leaf for natural tens/units separation

• Moderate size: Works well for 15-50 data points

• Exact values: Use when preserving original data is important

Decision Framework for Data Displays
n \leq 30 \Rightarrow \text{Dot Plot or Stem-and-Leaf}
Small Dataset Rule
Dot Plot
n \leq 30
Discrete data
Preserve values
Shows individual points
Histogram
n > 30
Continuous data
Show distribution
Groups data into bins
Box Plot
Compare groups
Show outliers
Five-number summary
Summary statistics
Key definitions:

Quantitative data: Numerical data that can be measured or counted.

Data display: Visual representation of data to reveal patterns, trends, and relationships.

Appropriateness: Selecting the display that best serves the purpose of analysis.

Decision-making process:
  1. Assess data characteristics: Size, type, range, distribution
  2. Identify analysis goals: Shape, center, spread, outliers, comparison
  3. Consider audience: Technical vs. general audience
  4. Match display to purpose: Select most informative option
  5. Verify effectiveness: Does the display answer the intended questions?
Tip 1: Small datasets (n ≤ 30) work best with dot plots or stem-and-leaf plots.
Tip 2: Large datasets (n > 30) work better with histograms or box plots.
Tip 3: Use box plots for comparing multiple groups.
Tip 4: Preserve exact values when precision matters.
Common errors: Using inappropriate scale, choosing complex displays for simple data, ignoring outliers.
Best practices: Match complexity to audience, consider multiple displays, verify readability.
Decision guidelines:

• Sample size: n ≤ 30 → dot plot/stem-and-leaf, n > 30 → histogram/box plot

• Purpose: Distribution shape → histogram, comparison → box plot

• Preservation: Need exact values → dot plot/stem-and-leaf, summary OK → box plot

Solution: Exercises 4 to 5
4 Multiple Display Comparison
Exercise 4
For the dataset of ages of 150 employees: 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150. Compare the effectiveness of a histogram vs. a box plot for this data.
Definition:

Effectiveness: How well a display achieves its intended purpose of communicating information about the data.

Comparative analysis: Evaluating multiple display options based on their strengths and weaknesses.

Step 1: Analyze the dataset characteristics

Sample size: n = 129 (large dataset)

Range: 22 to 150 years (very wide range)

Type: Continuous-like discrete data

Special feature: Contains potentially unrealistic values (150 years)

Step 2: Evaluate histogram for this data

Advantages:

• Shows the overall distribution shape

• Reveals the presence of outliers (age 150)

• Good for large datasets

Disadvantages:

• May not clearly show summary statistics

• Could be affected by outliers

Step 3: Evaluate box plot for this data

Advantages:

• Clearly shows outliers (age 150)

• Provides five-number summary

• Resistant to extreme values

Disadvantages:

• Doesn't show the overall distribution shape

• Hides the fact that data increases sequentially

Step 4: Consider the research question

If interested in distribution shape: Histogram is better

If interested in summary statistics: Box plot is better

If interested in outliers: Box plot is better

Step 5: Make recommendation

For this dataset, a box plot is more effective because it clearly identifies the outlier (age 150), which is likely a data entry error. The box plot provides clear summary statistics that are not distorted by the extreme value.

Box plot is more effective for this dataset
Final answer:

A box plot would be more effective for this dataset because it clearly identifies the outlier (age 150), which is likely a data entry error. The box plot provides robust summary statistics that are not distorted by the extreme value, making it easier to understand the true distribution of employee ages.

Applied rules:

• Outlier sensitivity: Box plots are more robust to extreme values

• Dataset size: Both displays work well for large datasets

• Data quality: Box plots help identify potential errors

5 Decision Tree Application
Exercise 5
Create a decision tree to determine the most appropriate display for quantitative data. Apply it to the following scenarios: A) Heights of 25 students, B) Daily temperatures over a year, C) Salaries of 500 employees.
Definition:

Decision tree: A flowchart that helps make decisions by systematically evaluating criteria at each step.

Systematic approach: Following a consistent process to select the best display.

Step 1: Create the decision tree
Start: Data characteristics
├─ Sample size ≤ 30?
│  ├─ Yes → Dot Plot or Stem-and-Leaf
│  └─ No → Continue to next question
├─ Need to preserve exact values?
│  ├─ Yes → Dot Plot or Stem-and-Leaf
│  └─ No → Continue to next question
├─ Want to show distribution shape?
│  ├─ Yes → Histogram
│  └─ No → Continue to next question
└─ Want to compare groups or show outliers?
   ├─ Yes → Box Plot
   └─ No → Histogram
                
Step 2: Apply to Scenario A (Heights of 25 students)

n = 25 ≤ 30? Yes → Dot plot or stem-and-leaf

Need exact values? Yes (precision important) → Stem-and-leaf

Best choice: Stem-and-leaf plot

Step 3: Apply to Scenario B (Daily temperatures over a year)

n = 365 > 30? Yes → Histogram or Box plot

Want to show distribution shape? Yes → Histogram

Best choice: Histogram

Step 4: Apply to Scenario C (Salaries of 500 employees)

n = 500 > 30? Yes → Histogram or Box plot

Want to show distribution shape? Yes → Histogram

Want to show outliers? Yes (salary distributions often have outliers) → Box plot

Best choice: Both (complement each other)

Step 5: Evaluate results

Scenario A: Stem-and-leaf plot preserves exact height values

Scenario B: Histogram shows seasonal temperature patterns

Scenario C: Histogram shows distribution shape, box plot shows outliers

Decision tree applied successfully to all scenarios
Final answer:

Scenario A: Stem-and-leaf plot for 25 student heights. Scenario B: Histogram for daily temperatures. Scenario C: Both histogram and box plot for salaries. The decision tree ensures appropriate display selection based on data characteristics and analysis goals.

Applied rules:

• Systematic approach: Use decision trees to ensure consistency

• Multiple displays: Sometimes more than one display is needed

• Context matters: Analysis goals influence display choice

Comprehensive Guide to Choosing Data Displays
\text{Display Choice} = f(\text{n}, \text{type}, \text{purpose}, \text{audience})
Display Selection Function
Key definitions:

Quantitative data: Numerical data that can be measured or counted, allowing for mathematical operations.

Data visualization: The representation of data through visual means to facilitate understanding and analysis.

Display appropriateness: The degree to which a visual representation effectively communicates the intended information about the data.

Effective communication: Presenting data in a way that allows the audience to quickly grasp key insights and patterns.

Systematic selection process:
  1. Characterize the data: Assess sample size, data type, range, and distribution
  2. Define the objective: Determine what insights are needed from the data
  3. Consider the audience: Match complexity and format to the intended viewers
  4. Evaluate options: Compare available display types against criteria
  5. Select and implement: Choose the most appropriate display and create it
  6. Validate effectiveness: Ensure the display achieves its intended purpose
Tip 1: Small datasets (≤30) benefit from displays that preserve individual values.
Tip 2: Large datasets (>30) work better with grouped displays like histograms.
Tip 3: Use box plots when comparing multiple groups or focusing on outliers.
Tip 4: Histograms are ideal for showing distribution shape and modality.
Tip 5: Sometimes multiple displays complement each other effectively.
Tip 6: Always consider the research question when selecting a display.
Display Comparison: Dot plots show individual values, histograms show distribution shape, box plots show summary statistics, stem-and-leaf plots preserve values while showing distribution.
Sample Size Guidelines: n ≤ 10: Dot plot, 10 < n ≤ 50: Dot plot or stem-and-leaf, n > 50: Histogram or box plot.
Analysis Goals: Shape → Histogram, Summary → Box plot, Precision → Dot plot/Stem-and-leaf, Comparison → Box plot.
Essential selection criteria:

• Sample size: n ≤ 30 → Individual displays, n > 30 → Grouped displays

• Data type: Discrete → Dot plots, Continuous → Histograms

• Analysis purpose: Distribution → Histogram, Outliers → Box plot, Precision → Stem-and-leaf

• Audience: Technical → Multiple displays, General → Simple displays

Display Selection Framework
Data Display Decision Matrix
Choose the appropriate display based on these criteria:
Small datasets (n ≤ 30): Dot plots or stem-and-leaf
Large datasets (n > 30): Histograms or box plots
Need exact values: Dot plots or stem-and-leaf
Show distribution shape: Histograms
Compare groups: Box plots
Show outliers: Box plots

Analysis: Effective data visualization matches the display type to the data characteristics and analysis goals.

  • Dot plots: Best for small datasets with repeated values
  • Histograms: Ideal for showing distribution shape and patterns
  • Box plots: Excellent for comparing groups and showing outliers
  • Stem-and-leaf: Preserves exact values while showing distribution

Questions & Answers

Question: How do I know when to use a dot plot versus a histogram? What are the key differences that guide this decision?

Answer: The decision depends on several key factors:

  • Sample size: Use dot plots for small datasets (n ≤ 30) and histograms for larger datasets (n > 30)
  • Individual values: Dot plots preserve exact values, histograms group them into intervals
  • Data type: Dot plots work well for discrete data with repeated values, histograms for continuous or widely spread data

When to use dot plots:

  • Small datasets (10-30 observations)
  • Data with repeated values
  • When preserving exact values is important
  • Discrete data (number of children, pets, etc.)

When to use histograms:

  • Large datasets (more than 30 observations)
  • Continuous data (heights, weights, temperatures)
  • Focus on overall distribution shape
  • Need to group data for clarity

The key difference is that dot plots show individual data points, while histograms show grouped frequencies.

Question: When would I use a box plot instead of a histogram? Don't they both show the distribution of data?

Answer: While both show aspects of data distribution, they serve different purposes:

  • Histograms: Show the shape of the distribution, revealing patterns like skewness, modality (unimodal, bimodal), and clusters in the data. They display the frequency of values within intervals.
  • Box plots: Show the five-number summary (min, Q1, median, Q3, max) and clearly identify outliers. They emphasize central tendency and spread.

Use box plots when:

  • Comparing distributions across multiple groups
  • Focusing on identifying outliers
  • Showing summary statistics
  • Working with data that has extreme values

Use histograms when:

  • Understanding the overall shape of the distribution
  • Looking for patterns, clusters, or gaps
  • Showing the frequency of values in ranges
  • Presenting to audiences unfamiliar with statistical summaries

Often, both displays complement each other effectively.

Question: What's the advantage of a stem-and-leaf plot over a histogram? When would I choose one over the other?

Answer: The key difference is information preservation:

  • Stem-and-leaf plots: Preserve the original data values while showing the distribution. You can reconstruct the original dataset from the plot.
  • Histograms: Group data into bins, losing the original individual values. You cannot reconstruct the original dataset.

Choose stem-and-leaf when:

  • You need to preserve exact values
  • Working with moderate-sized datasets (15-50 observations)
  • Data naturally splits into stems and leaves (two or three digits)
  • Want to show distribution while maintaining precision

Choose histogram when:

  • Working with large datasets (more than 50 observations)
  • Focus is on overall distribution shape
  • Data doesn't naturally split into stems/leaves
  • Need a more compact visual representation

Think of stem-and-leaf plots as a hybrid between raw data and visual display, while histograms are purely visual.

Question: Can I use more than one type of display for the same dataset? When would this be beneficial?

Answer: Absolutely! Using multiple displays for the same dataset is often highly beneficial:

  • Histogram + Box plot: Histogram shows distribution shape, box plot shows outliers and summary statistics
  • Dot plot + Box plot: Dot plot shows individual values, box plot shows summary measures
  • All three together: Comprehensive view of the data distribution

Benefits of multiple displays:

  • Different aspects of the data become visible
  • Validation: patterns seen in one display confirmed by another
  • Flexibility: different audiences can focus on what's most relevant
  • Comprehensive analysis: no single aspect overlooked

Example: For salary data, a histogram shows the distribution shape (often right-skewed), while a box plot clearly shows outliers (very high salaries) that might be missed in the histogram.

The key is to choose displays that provide complementary rather than redundant information.

Question: How does the audience affect my choice of data display? Should I consider their statistical knowledge?

Answer: Audience consideration is crucial for effective data communication:

  • Technical audience: Can handle complex displays like box plots, multiple panels, or combined visualizations
  • General audience: Prefer simpler, more intuitive displays like histograms or dot plots
  • Executive audience: May prefer summary displays like box plots that quickly convey key metrics

Guidelines by audience:

  • Students/public: Start with histograms or dot plots, add box plots as needed
  • Researchers: Multiple displays showing different aspects
  • Managers: Focus on box plots for quick comparisons
  • Statisticians: Any display that best answers the question

Key principle: The best display is one that allows your specific audience to quickly understand the insights you want to communicate.

Sometimes you might create multiple versions of the same data for different audiences, each optimized for that group's needs and expertise level.