Choosing Appropriate Displays: Complete Guide with Solved Exercises

Master choosing the right data display for quantitative data: histograms, box plots, dot plots, stem-and-leaf plots, and scatter plots through 5 detailed exercises with comprehensive solutions.

Solution: Exercises 1 to 3
1 Choosing Between Histogram and Box Plot
Exercise 1
You have a dataset of 150 students' test scores ranging from 45 to 98. Which display would be most appropriate to show the distribution? Explain your reasoning.
Definition:

Choosing appropriate displays: Selecting the most effective visualization method based on data characteristics, sample size, and intended analysis goals.

Decision Framework:
  1. Consider sample size (n)
  2. Identify analysis goals (shape, center, spread, outliers)
  3. Evaluate data characteristics (range, distribution type)
  4. Select display that best addresses primary objectives
Step 1: Analyze data characteristics

Sample size: n = 150 (large sample)

Data type: Continuous quantitative (test scores)

Range: 45 to 98 (wide range)

Objective: Show overall distribution shape

Step 2: Evaluate display options

Histogram: Excellent for large samples, shows shape clearly, handles continuous data well

Box plot: Good for center/spread, but doesn't show shape details

Dot plot: Not suitable for 150 data points (too crowded)

Step 3: Consider analysis goals

Primary goal: Show distribution shape and overall pattern

Histogram excels at showing shape (symmetric, skewed, bimodal)

Box plot focuses on five-number summary rather than shape

Step 4: Make final decision

Histogram is most appropriate for this scenario

Large sample size (150) makes histogram ideal

Will clearly show the distribution shape and pattern

Reasoning: With 150 data points, a histogram will effectively show the distribution shape, which is the primary goal when displaying test scores. The large sample size allows for meaningful binning that reveals patterns in the data.

Histogram is most appropriate for large sample (n=150) showing distribution shape
Final Answer:

Histogram is the most appropriate display for this dataset. With 150 test scores, a histogram will effectively show the distribution shape, identify any skewness or unusual patterns, and provide insight into the overall performance distribution. The large sample size makes the histogram ideal for revealing meaningful patterns in the data.

Applied Rules:

• Sample size rule: Histograms for large samples (n > 30), dot plots for small samples

• Shape visibility: Histograms show distribution shape better than box plots

• Continuous data: Histograms handle continuous quantitative data effectively

2 Small Dataset Display Choice
Exercise 2
You have a dataset of 12 students' project grades: 87, 89, 92, 94, 95, 96, 97, 98, 99, 100, 101, 102. Which display would be most appropriate? Justify your choice.
Statistical Considerations:

Small sample considerations: For small datasets (n < 20), individual data points become important and detailed visualization methods are preferred.

Step 1: Analyze data characteristics

Sample size: n = 12 (small sample)

Data type: Quantitative (grades)

Range: 87 to 102 (relatively narrow)

Goal: Show all individual values clearly

Step 2: Evaluate display options

Dot plot: Perfect for small samples, shows each value individually

Stem-and-leaf: Also good for small samples, preserves exact values

Histogram: Not suitable (would have few bars with low resolution)

Box plot: Loses individual value information

Step 3: Consider advantages of each option

Dot plot: Easy to create, shows all values, reveals clusters/gaps

Stem-and-leaf: Preserves exact values, shows shape, allows reconstruction

Step 4: Make decision

Both dot plot and stem-and-leaf are appropriate

Dot plot is simplest and clearest for this purpose

Step 5: Consider alternative if comparison needed

If comparing to another class, box plot might be better for comparison

For standalone display, dot plot is optimal

Reasoning: With only 12 data points, showing each value individually is important. A dot plot will clearly display each grade while preserving the exact values and revealing any patterns in the distribution.

Dot plot is most appropriate for small sample (n=12) showing individual values
Final Answer:

Dot plot is the most appropriate display for this dataset. With only 12 data points, a dot plot will show each individual grade clearly, making it easy to see the distribution, identify any outliers, and observe the exact values. This is especially important for small samples where each data point carries more weight in the overall pattern.

Applied Rules:

• Small sample rule: Dot plots and stem-and-leaf plots for n < 20

• Value preservation: Small samples benefit from displays showing exact values

• Detail visibility: Individual points are important in small samples

3 Comparison Scenario Display Choice
Exercise 3
You need to compare test scores between two classes: Class A (n=25) and Class B (n=28). Which display would be most appropriate for comparing the two distributions? Explain your reasoning.
Comparison Display Choice:

Multi-group comparison: When comparing multiple distributions, displays that allow for easy visual comparison of center, spread, and shape are preferred.

Comparison Analysis:
  1. Identify the need for comparison
  2. Consider displays that work well side-by-side
  3. Evaluate which display best shows differences
  4. Consider space efficiency and clarity
Step 1: Analyze the comparison needs

Groups to compare: 2 classes

Sample sizes: n=25 and n=28 (moderate)

Goal: Compare centers, spreads, and shapes

Step 2: Evaluate comparison options

Side-by-side box plots: Excellent for comparing centers and spreads

Back-to-back stem-and-leaf: Good for moderate samples, shows exact values

Overlapping histograms: Good for shape comparison but can be cluttered

Side-by-side dot plots: Good but takes more space

Step 3: Consider advantages

Box plots: Compact, clear comparison of key statistics, easy to see differences

Stem-and-leaf: Preserves values but more complex to create

Step 4: Consider practical aspects

Box plots are more commonly used in comparative analysis

Easier to create and interpret for most audiences

Space-efficient for reports and presentations

Step 5: Make final recommendation

Side-by-side box plots are optimal for this comparison

Will clearly show differences in medians, IQRs, and ranges

Reasoning: Side-by-side box plots are ideal for comparing two distributions. They take minimal space, clearly show differences in center and spread, and make it easy to identify outliers or unusual features in each group.

Side-by-side box plots most appropriate for comparing two distributions
Final Answer:

Side-by-side box plots are the most appropriate display for comparing the two classes. This format allows for easy visual comparison of centers (medians), spreads (IQRs and ranges), and shapes. The box plots will clearly reveal any differences in typical performance and variability between the classes while remaining concise and easy to interpret.

Applied Rules:

• Comparison rule: Box plots excel at multi-group comparisons

• Space efficiency: Box plots are compact for multiple groups

• Key statistics: Box plots highlight medians and IQRs for comparison

Choosing Appropriate Displays: Decision Framework
Display = f(n, data_type, goal)
Selection Function
Sample Size
n < 20: Dot/Stem, n > 30: Histogram
Size determines display choice
Comparison
Multiple groups: Box plots
For comparing distributions
Analysis Goal
Shape: Histogram, Values: Dot/Stem
Goal drives selection
Key Definitions:

Quantitative Data: Numerical data that can be measured and subjected to mathematical operations (heights, weights, scores, etc.).

Display Selection Criteria: Factors that influence the choice of visualization method including sample size, analysis goals, and data characteristics.

Shape Visibility: The ability of a display to show the overall pattern of the distribution (symmetric, skewed, bimodal, etc.).

Value Preservation: Whether a display maintains the exact values of the data points or groups them into categories.

Complete Selection Process:
  1. Assess sample size: Small (n < 20), moderate (20 ≤ n ≤ 100), large (n > 100)
  2. Identify analysis goals: Shape, center, spread, outliers, comparisons
  3. Consider data characteristics: Continuous vs discrete, range, distribution type
  4. Evaluate audience: Technical vs general audience considerations
  5. Select optimal display: Based on previous assessments
Tip 1: For n < 20, use dot plots or stem-and-leaf plots to show individual values.
Tip 2: For n > 30, histograms are generally preferred for showing distribution shape.
Tip 3: For comparisons between groups, side-by-side box plots are usually optimal.
Tip 4: When exact values matter, choose stem-and-leaf or dot plots over histograms.
Advantages: Proper display choice enhances understanding, reveals important patterns, supports effective communication of data insights.
Considerations: Always match display to analysis goals; consider audience; be aware of limitations of each display type.
Essential Rules:

• Sample size rule: n < 20 → dot/stem, 20 ≤ n ≤ 100 → box/histogram, n > 100 → histogram

• Goal alignment: Match display to primary analysis objective

• Comparison principle: Use side-by-side displays for multi-group comparisons

• Shape visibility: Histograms best show distribution shape

• Value preservation: Dot/stem plots maintain exact values

Solution: Exercises 4 to 5
4 Outlier Detection Display Choice
Exercise 4
You have a dataset of house prices in a neighborhood with potential outliers. Which display would be most appropriate to identify and show outliers? Justify your choice.
Outlier Detection:

Outlier identification: Some displays are specifically designed to highlight unusual data points that deviate significantly from the overall pattern.

Step 1: Analyze the outlier detection need

Goal: Identify and clearly show outliers

Data type: Quantitative (house prices)

Concern: Unusual values that may affect analysis

Step 2: Evaluate outlier detection capabilities

Box plot: Explicitly designed to identify outliers using 1.5×IQR rule

Histogram: Outliers may be visible but not clearly marked

Dot plot: Outliers visible but not systematically identified

Stem-and-leaf: Outliers visible but not systematically identified

Step 3: Consider the systematic approach

Box plots use the standard 1.5×IQR rule for outlier detection

Provides objective criteria for identifying outliers

Clearly marks outliers separately from main distribution

Step 4: Evaluate visual clarity

Box plots explicitly show whiskers and mark outliers as separate points

Easy to distinguish outliers from regular data points

Provides immediate visual identification

Step 5: Consider additional benefits

Box plots also show center, spread, and shape while highlighting outliers

Provide comprehensive view of distribution characteristics

Reasoning: Box plots are specifically designed with outlier detection in mind. They use the 1.5×IQR rule to systematically identify and visually separate outliers from the main body of data, making them ideal for this purpose.

Box plot most appropriate for systematic outlier detection and visualization
Final Answer:

Box plot is the most appropriate display for identifying outliers in house prices. Box plots are specifically designed with outlier detection in mind, using the standard 1.5×IQR rule to systematically identify values that fall outside the expected range. Outliers are clearly marked as separate points beyond the whiskers, making them immediately visible and objectively identified.

Applied Rules:

• Outlier detection: Box plots use 1.5×IQR rule for systematic identification

• Visual separation: Outliers shown as distinct points beyond whiskers

• Objective criteria: Standard rule provides consistent identification

5 Shape Analysis Display Choice
Exercise 5
You want to analyze the shape of a distribution of commute times for 200 employees to determine if it's symmetric, skewed, or bimodal. Which display would be most appropriate? Explain your reasoning.
Shape Analysis:

Distribution shape: Understanding whether data follows a symmetric, skewed, or multimodal pattern is crucial for statistical analysis and interpretation.

Step 1: Analyze the shape analysis goal

Sample size: n = 200 (large sample)

Primary goal: Identify distribution shape (symmetric, skewed, bimodal)

Data type: Quantitative (commute times)

Step 2: Evaluate shape visibility in different displays

Histogram: Excellent for showing distribution shape and modality

Box plot: Shows symmetry through quartile positions but limited shape info

Dot plot: Not suitable for n=200 (too crowded)

Stem-and-leaf: Not practical for n=200

Step 3: Consider bimodal detection

Histograms clearly show multiple peaks if bimodality exists

Box plots cannot detect bimodality effectively

Shape features like tails and peaks are visible in histograms

Step 4: Evaluate large sample suitability

Histograms work excellently with large samples

Large sample size provides stable and meaningful binning

Patterns emerge clearly with sufficient data points

Step 5: Consider analysis depth

Histograms allow detailed shape analysis including skewness degree

Can identify subtle shape features that box plots miss

Provide comprehensive visual of the entire distribution

Reasoning: For shape analysis with a large sample (n=200), a histogram is optimal because it provides the clearest visualization of the distribution's overall pattern, including any bimodality, skewness, or other shape characteristics that cannot be seen in other displays.

Histogram most appropriate for shape analysis with large sample (n=200)
Final Answer:

Histogram is the most appropriate display for analyzing the shape of commute times with 200 employees. Histograms excel at showing distribution shape, including whether it's symmetric, skewed, or bimodal. The large sample size allows for meaningful binning that reveals patterns, and histograms can clearly show multiple peaks if bimodality exists, which is crucial for understanding the underlying distribution of commute times.

Applied Rules:

• Shape visibility: Histograms best show distribution shape and modality

• Large sample advantage: Histograms work well with n > 100

• Bimodal detection: Only histograms can reliably show multiple peaks

Comprehensive Guide: Display Selection Laws, Methods, Rules, and Definitions
Optimal_Display = f(sample_size, analysis_goal, data_characteristics)
Selection Formula
Key definitions:

Quantitative Data Display: Visual representations of numerical data designed to reveal patterns, trends, and relationships in the data.

Sample Size Considerations: The number of observations influences which display type is most effective for revealing meaningful patterns.

Analysis Goals: The specific objectives of the data analysis (describing shape, comparing groups, identifying outliers, etc.) determine optimal display choice.

Data Characteristics: Properties of the data including continuity, range, distribution type, and measurement scale that influence display selection.

Complete selection methodology:
  1. Sample size assessment: Categorize as small (n < 20), moderate (20 ≤ n ≤ 100), or large (n > 100)
  2. Goal identification: Determine primary analysis objectives (shape, center, spread, outliers, comparison)
  3. Characteristic evaluation: Assess data type, range, and distribution properties
  4. Audience consideration: Consider the technical level and preferences of the intended audience
  5. Display selection: Choose based on the intersection of the above factors
  6. Validation: Ensure the chosen display effectively communicates the intended message
Tip 1: For small samples (n < 20), prioritize displays that show individual values (dot plots, stem-and-leaf).
Tip 2: For large samples (n > 100), histograms are usually optimal for showing distribution shape.
Tip 3: For comparisons between groups, box plots are typically the most efficient choice.
Tip 4: When outliers are important, box plots provide systematic identification.
Tip 5: For shape analysis, histograms provide the most detailed visualization.
Advantages: Proper display choice enhances data understanding, reveals hidden patterns, supports accurate interpretation, and improves communication effectiveness.
Limitations: Each display type has inherent limitations; some information may be lost; context-dependent effectiveness varies.
Applications: Research, business analytics, quality control, academic reporting, and exploratory data analysis.
Essential selection rules:

• Sample size rule: n < 20 → dot/stem, 20 ≤ n ≤ 100 → box/histogram, n > 100 → histogram

• Shape detection: Histograms are superior for identifying distribution shape and modality

• Outlier identification: Box plots use systematic 1.5×IQR rule for outlier detection

• Comparison efficiency: Box plots are space-efficient for multiple group comparisons

• Value preservation: Dot plots and stem-and-leaf maintain exact data values

Detailed Summary: Choosing Appropriate Displays Mastery Guide
Display_Selection_Framework
Core Decision Process
Comprehensive Definitions:

Choosing Appropriate Displays: A systematic approach to selecting the most effective visualization method based on data characteristics, sample size, and analysis objectives. The goal is to maximize information clarity and communication effectiveness.

Display Selection Framework: A structured methodology that considers sample size, data type, analysis goals, and audience requirements to determine optimal visualization strategy.

Quantitative Data Visualization: The practice of representing numerical data through visual formats that reveal patterns, relationships, and insights that might not be apparent in raw data.

Analysis Objectives: The specific goals of data examination, including describing distribution shape, comparing groups, identifying outliers, summarizing central tendency, and assessing variability.

Complete Decision Process:
  1. Data characterization: Assess sample size, data type, range, and distribution characteristics
  2. Goal setting: Define primary and secondary analysis objectives
  3. Option evaluation: Consider how each display type serves the defined goals
  4. Constraint assessment: Evaluate audience, presentation format, and technical requirements
  5. Selection decision: Choose the display that best meets all criteria
  6. Validation check: Ensure the chosen display effectively communicates the intended message
  7. Alternative consideration: Think about supplementary displays if needed
Example 1 (Simple): For 15 student grades, use dot plot to show each value individually and reveal any patterns.
Example 2 (Advanced): For 500 test scores, use histogram to show overall shape and identify any bimodality or skewness.
Common Mistake: Using histogram for small samples (n < 20) - this loses important individual value information.
Memory Aid: "Small samples show all, big samples summarize" - small samples need individual value displays.
Key Notes: Always consider the story the data tells; different displays emphasize different aspects; sample size is a primary factor; audience sophistication matters; context drives the choice.
Takeaways: No single display is universally best; effectiveness depends on context; shape analysis requires histograms; comparison requires box plots; value preservation requires dot plots or stem-and-leaf.
Core Principles for Success:

• Match to goals: Always align display choice with primary analysis objectives

• Consider sample size: Use different approaches for small, moderate, and large samples

• Think about audience: Match complexity and format to viewer sophistication

• Preserve important features: Don't lose critical information in the visualization process

• Validate effectiveness: Ensure the chosen display actually communicates what's intended

Questions & Answers

Question: How do I decide between a histogram and a box plot when both seem appropriate for my data? What factors should I prioritize?

Answer: The choice between histogram and box plot depends on your primary objective:

  • Choose histogram when: You want to see the overall shape of the distribution, identify modality (unimodal, bimodal), or detect subtle patterns in the data
  • Choose box plot when: You want to focus on center, spread, and outliers, or when comparing multiple groups

If your sample size is large (n > 100), histogram is usually better for shape analysis. If you need to compare several groups, box plots are more space-efficient. For outlier identification, box plots provide systematic detection criteria.

Consider using both if you have space and want comprehensive analysis, but prioritize based on your main analytical goal.

Question: What's the difference between a stem-and-leaf plot and a dot plot? When should I use each one?

Answer: Both preserve individual values but serve different purposes:

  • Stem-and-leaf: Shows exact values in a tabular format, good for seeing shape while preserving numbers
  • Dot plot: Shows each value as a dot on a number line, good for seeing frequency and clustering

Use stem-and-leaf when you want to see the actual numerical values clearly and perform calculations. Use dot plot when you want to easily see how many times each value occurs and identify clusters or gaps. Both work well for small samples (n < 20), but dot plots become cluttered with larger samples.

Stem-and-leaf is better for detailed analysis, dot plot is better for quick pattern recognition.

Question: Can I use multiple displays for the same dataset? Is this redundant or helpful?

Answer: Using multiple displays is often helpful, not redundant! Each display emphasizes different aspects:

  • Histogram: Shows overall shape and modality
  • Box plot: Emphasizes center, spread, and outliers
  • Dot plot: Shows individual values and frequency

For important analyses, presenting multiple views gives a more complete picture. However, don't overwhelm your audience - use multiple displays strategically when each adds unique value. For reports or presentations, sometimes showing both a histogram and box plot provides comprehensive insight.

The key is ensuring each display serves a distinct purpose rather than repeating the same information.