Qualitative Data Analysis: Frequency Distributions and Data Visualization Techniques

Fundamentals of Frequency Distributions

• Definition of Frequency:

  • Frequency is the headcount or raw count of observations belonging to distinct categories within a categorical dataset.

• Three Reporting Methods for Frequency Distributions:

  • Frequency: The raw headcount of observations per category.

  • Relative Frequency: The fraction or decimal proportion of total observations represented by a category, calculated as:     $\text{Relative Frequency} = \frac{\text{Frequency}}{\text{Total Number of Observations}}$

  • Percent Frequency: The relative frequency expressed as a percentage, calculated by multiplying the relative frequency by $100$:     $\text{Percent Frequency} = \text{Relative Frequency} \times 100$

  • All three methods represent the same underlying data, presented in different formats to accommodate varying audience preferences and levels of comprehension.

• ABC Workwear Payment Method Dataset Example:

  • Total customer sample size ($N$): $200$ customers.

  • Payment categories ($4$ total): Credit Card, Debit Card, Apple Pay, PayPal.

  • Credit Card frequency count: $77$ customers.

  • Credit Card proportion: Represents approximately $\frac{1}{3}$ ($38.5\%$ ) of the customer sample population.

• Data Integrity and Self-Check Quality Controls:

  • Total Frequency Check: The sum of all individual category frequencies must strictly equal the total number of rows/observations in the dataset ($200$).

  • Total Relative Frequency Check: The sum of all relative frequencies must strictly equal $1$ (or $1.00$). Any sum other than $1$ indicates a typing or calculation error.

  • Total Percent Frequency Check: The sum of all percent frequencies must strictly equal $100\%$.

• Business Insight Derivation:

  • Identifying that Credit Card is the dominant payment method ($77$ out of $200$ customers) provides actionable intelligence for management.

  • Management can leverage this high volume to negotiate lower transaction/processing fees with credit card processors or launch targeted promotional programs.

Hand Calculations and Data Verification

• Step-by-Step Regional Payment Frequency Calculation ($N = 200$):

  • Southwest Region frequency count: $47$ customers.

  • Southwest Relative Frequency calculation:     $\text{Relative Frequency} = \frac{47}{200} = 0.235$

  • Southwest Percent Frequency calculation:     $\text{Percent Frequency} = 0.235 \times 100 = 23.5\%$

• Psychological Mechanism in Learning and Error Retention:

  • Research on educational outcomes indicates that students remember a single mistake made by an instructor during a lecture far more vividly than all the correctly presented material.

  • Human cognition inherently focuses on and retains errors; deliberately making and correcting mistakes during practice fixes concepts in long-term memory and prevents repeating those errors in application.

Basic and Horizontal Bar Charts

• Basic Bar Chart Structure:

  • X-axis (Horizontal): Displays qualitative, nominal payment categories (e.g., Credit Card, Debit Card, Apple Pay, PayPal).

  • Y-axis (Vertical): Displays frequency counts of customers (e.g., tick marks at $10$, $20$, $30$).

  • Bar Height: Directly corresponds to the headcount in each category.

• Horizontal Bar Charts:

  • Primary Use Case: Preferred over vertical bar charts when category names are long (e.g., geographic regions like Southwest, Northeast, West Coast).

  • Readability Advantage: Eliminates visual crowding and prevents readers from having to tilt their heads to read vertical or angled axis labels.

  • Information Delivery: Designed to allow viewers to grasp the key insights within a few seconds of glancing at the visual.

  • ABC Workwear Regional Example:

  • Southwest leads as the largest region with $47$ customers.

  • West Coast is the smallest region with $33$ customers.

  • Sorting Principle: Arranging horizontal bars in descending order (largest to smallest) makes ranking visual, immediate, and effortless.

• Step-by-Step Guide to Building Bar Charts in Excel:

  • Step 1: Select the data table. (In most cases, clicking any single cell inside the data range is sufficient, but if Excel fails to detect the range, manually select the full table).

  • Step 2: Click the Insert tab (located at the top-left, directly adjacent to the Home tab).

  • Step 3: Locate the Recommended Charts section in the center-right of the ribbon and select Clustered Column (or Clustered Bar).

  • Step 4: Add explicit axis titles and format/polish the chart layout for corporate presentation standards.

Pie Charts and Comparative Visualization Principles

• Pie Chart Fundamentals:

  • Purpose: Displays parts of a whole (proportional split).

  • Category Limitations: Suitable strictly for $2$ to $5$ categories ($2$ to $4$ categories is optimal). Charts with more than $5$ slices become cluttered and unintelligible.

  • ABC Workwear Customer Type Example:

  • Repeat Customers: $53\%$

  • New Customers: $47\%$

  • Strategic Business Insight: A $53\%$ repeat customer rate signals a need to implement customer retention strategies, such as customer loyalty programs, to maintain existing buyers while continuing to acquire new ones.

• Decision Framework: Bar Chart vs. Pie Chart:

  • Use a Pie Chart when highlighting proportional parts of a single whole across $2$ to $4$ categories.

  • Use a Bar Chart when:

  • Comparing relative frequencies across distinct categories.

  • Evaluating datasets with more than $4$ or $5$ categories.

  • Precise numerical comparison across categories is required (bar height differences are much easier for the human eye to judge accurately than pie slice angles).

  • Avoid Bar Charts when trying solely to illustrate simple proportions of a whole across very few categories.

  • Avoid Pie Charts when precise comparative analysis is necessary or when category counts exceed $5$.

  • Strict Avoidance Rule for $3\text{D}$ Charts: Never use $3\text{D}$ pie charts or any $3\text{D}$ charts. They distort spatial perspective, obscure actual slice dimensions, hinder visual comprehension, and fail to convey functional business information.

• Chart Selection Practice Scenarios:

  • Scenario 1: Determine which payment method is used most frequently at ABC Workwear ($4$ categories).

  • Optimal Visual: Bar Chart (best for comparing discrete category frequencies).

  • Scenario 2: Determine the share of total customers originating from each region ($5$ regions).

  • Optimal Visual: Bar Chart ($5$ categories reaches the absolute limit for pie charts; subtle slice differences such as $39$ vs $40$ vs $41$ are visually impossible to distinguish as pie slices).

  • Scenario 3: Compare the number of repeat customers against new customers ($2$ categories).

  • Optimal Visual: Pie Chart (ideal $2$-category parts-of-a-whole comparison).

  • Scenario 4: Display sales frequency across $12$ distinct product categories.

  • Optimal Visual: Bar Chart ($12$ categories is far too complex for a pie chart).

Qualitative Analysis Summary and Chart Diagnostic Rules

• Complete Qualitative Variable Summary for ABC Workwear:

  • Payment Methods ($4$ categories): Bar Chart (enables fee negotiation with top processors).

  • Regional Breakdown ($5$ regions): Sorted Horizontal Bar Chart (identifies top market geographic performance).

  • Repeat vs. New Customers ($2$ categories): Pie Chart ($53\%$ repeat vs $47\%$ new; drives loyalty program implementation).

• -