5.1 Categorical and Quantitative Data Displays and Frequency Tables

Categorical and Quantitative Data Displays and Frequency Tables

OpenExamPrep provides this probabilistic and statistical reasoning review to help students master data displays, frequency distributions, and contingency tables tested on the Texas Success Initiative Assessment 2.0 (TSIA2) Mathematics section. Statistical literacy begins with the ability to classify variables, read and extract information from diverse visual representations, evaluate relationships between categorical variables, and detect intentional or unintentional graphical distortions.

---

1. Data Taxonomy: Categorical vs. Quantitative Variables

Every statistical investigation begins by identifying the nature of the data collected. The classification of a variable dictates which graphical displays, numerical summaries, and probabilistic models are mathematically valid.

                           Statistical Variables
                                     │
         ┌───────────────────────────┴───────────────────────────┐
         ▼                                                       ▼
    Categorical                                             Quantitative
   (Qualitative)                                            (Numerical)
         │                                                       │
   ┌─────┴─────┐                                           ┌─────┴─────┐
   ▼           ▼                                           ▼           ▼
Nominal     Ordinal                                     Discrete   Continuous
(Unordered) (Ranked)                                    (Counted)  (Measured)

Categorical (Qualitative) Variables

Categorical variables place individuals or observations into distinct groups or categories based on qualitative attributes or characteristics:

• Nominal Variables: Categories without any inherent quantitative order, ranking, or hierarchy. Examples include college major (e.g., Biology, Engineering, English), eye color, marital status, blood type, and city of residence.
• Ordinal Variables: Categories that possess a natural, sequential ranking or hierarchical order, but the numerical distance between successive categories cannot be measured or assumed equal. Examples include survey satisfaction ratings (e.g., Highly Dissatisfied, Neutral, Satisfied), letter grades (A, B, C, D, F), socioeconomic status (Low, Middle, High), and class standing (Freshman, Sophomore, Junior, Senior).

Quantitative (Numerical) Variables

Quantitative variables record numerical measurements or counts for which standard arithmetic operations (such as addition, subtraction, and averaging) are mathematically meaningful:

• Discrete Variables: Values that result from a counting process, producing isolated, distinct points along the number line with gaps between them. Discrete data are typically represented by whole numbers or integers. Examples include the number of college credit hours completed (e.g., 12, 15, 18), the number of siblings in a household, and the number of customer inquiries received per hour.
• Continuous Variables: Values that result from measurement along an unbroken, continuous continuum. Between any two continuous measurements, there exist an infinite number of possible intermediate values, limited only by the precision of the measuring instrument. Examples include student commute time in minutes, infant birth weight in kilograms, room temperature in degrees Celsius, and running pace in seconds per mile.

Variable Taxonomy Comparison Table

⚠️ TSIA2 High-Frequency Trap: Numbers That Are Categorical

Do not assume that every numerical entry is quantitative! Athletic jersey numbers, area codes, social security numbers, and postal ZIP codes consist of digits, but adding them or computing their arithmetic mean is completely meaningless. A ZIP code of 78701 does not have "more location" than a ZIP code of 75001. These numbers function strictly as nominal categorical labels.

---

2. Graphical Displays for Categorical Data

Categorical data are summarized by counting how many observations fall into each category (frequency) or by calculating the fraction of the total sample in each category (relative frequency).

Standard Bar Charts

A bar chart plots categorical classes along one axis (usually the horizontal axis) and frequencies or relative frequencies along the other axis (usually the vertical axis):

• Distinct Separation: Unlike histograms, adjacent bars in a bar chart do not touch. The intentional physical spacing reinforces that the categories are separate, discrete qualitative classes.
• Ordering Flexibility: In nominal bar charts, categories may be arranged alphabetically or in order of decreasing frequency (creating a Pareto chart). In ordinal bar charts, categories should always follow their natural hierarchical order.

Segmented (Stacked) Bar Charts

A segmented bar chart compares the composition of multiple sub-groups across categorical attributes:

• Each bar represents a primary group and has a standardized total height representing 100% of that group.
• The bar is partitioned into stacked, shaded segments whose heights correspond to the relative frequencies of a secondary categorical variable.
• Segmented bar charts enable direct visual comparison of conditional proportions across groups of unequal absolute sizes.

Pie Charts (Circle Graphs)

A pie chart divides a circle (360°) into circular wedges whose central angles and surface areas are directly proportional to each category's share of the whole.

To calculate the central angle θ in degrees for a given category:

Central Angle θ = (Category Frequency ÷ Total Frequency) × 360° = Relative Frequency × 360°

Worked Example: Calculating Sector Angle
A community college surveys 800 enrolled students regarding their primary commute method:
  - Personal Automobile: 440 students
  - Public Transit: 200 students
  - Bicycle / Walking: 120 students
  - Carpool: 40 students

What is the central angle for Public Transit in a pie chart representing this cohort?

Step 1: Calculate the relative frequency (proportion):
  Proportion = 200 ÷ 800 = 0.25 (25.0%)

Step 2: Multiply by 360°:
  Central Angle = 0.25 × 360° = 90.0°

Check: Automobile = (440/800) × 360° = 198°; Bicycle/Walk = (120/800) × 360° = 54°;
Carpool = (40/800) × 360° = 18°. Sum = 198° + 90° + 54° + 18° = 360°.

---

3. Graphical Displays for Quantitative Data

Quantitative displays reveal the shape, center, spread, and unusual features of a numerical distribution.

Histograms

A histogram partitions continuous or discrete numerical data into non-overlapping, contiguous intervals called bins (or classes):

• The horizontal axis displays continuous numerical intervals of equal width.
• The vertical axis displays the frequency (count) or relative frequency of observations falling within each bin.
• Bars touch continuously: Adjacent bars must touch without gaps to indicate numerical continuity, unless a bin has a frequency of zero (which represents a true empty gap in the data).

Stem-and-Leaf Displays

A stem-and-leaf plot organizes quantitative data while retaining the exact original numerical values:

• Each data value is split into a stem (the leading digit or digits) and a leaf (the single trailing digit).
• Stems are aligned in a vertical column, separated by a vertical bar from the leaves, which are arranged horizontally in ascending numerical order.
• Every stem-and-leaf display must provide a key to define place value.

Stem-and-Leaf Plot: Daily Nursing Unit Patient Admissions (18 Days)
Key: 2 | 4 represents 24 patients

Stem │ Leaves
  1  │ 2  5  8  9
  2  │ 1  3  4  4  7  8  8
  3  │ 0  2  5  6  9
  4  │ 1  4

Data Interpretation:
- Sample size n = 18 days.
- Minimum = 12 patients; Maximum = 44 patients.
- Median is the average of the 9th and 10th values: (27 + 28) / 2 = 27.5 patients.

Dot Plots

A dot plot displays individual observations as dots stacked vertically above a number line. Dot plots are optimal for small to moderate data sets, displaying exact data points, clusters, gaps, and potential outliers without losing individual values.

Scatter Plots & Bivariate Quantitative Data

A scatter plot visualizes the relationship between two quantitative variables measured on the same individuals:

• Explanatory Variable (x-axis): The variable hypothesized to influence or predict the outcome.
• Response Variable (y-axis): The outcome variable being measured.

Pattern Interpretation in Scatter Plots:
- Positive Linear Association: As x increases, y tends to increase (points cluster along an upward-sloping line).
- Negative Linear Association: As x increases, y tends to decrease (points cluster along a downward-sloping line).
- Non-Linear Association: Points follow a curved trajectory (quadratic, exponential, or logarithmic) rather than a line.
- No Association: Points appear randomly dispersed across the grid with no discernible directional slope.

---

4. Frequency Tables & Two-Way Contingency Tables

When two categorical variables are recorded for a sample, data are organized into a two-way contingency table (cross-tabulation).

Structure of a Two-Way Table

Consider a college health center study investigating flu vaccination status and infection incidence among 500 college students:

The Three Types of Relative Frequencies

1. Joint Relative Frequency: The ratio of a single interior cell count to the grand total (all individuals in the entire table):

   Joint Relative Frequency = Cell Count ÷ Grand Total
   

   Example: What proportion of all surveyed students were vaccinated AND contracted the flu? Joint Frequency = 20 ÷ 500 = 0.04 (4.0%)

2. Marginal Relative Frequency: The ratio of a row total or column total to the grand total:

   Marginal Relative Frequency = Marginal Total ÷ Grand Total
   

   Example 1: What proportion of all surveyed students were vaccinated? Marginal Frequency (Vaccinated) = 200 ÷ 500 = 0.40 (40.0%) Example 2: What proportion of all surveyed students contracted the flu? Marginal Frequency (Contracted Flu) = 110 ÷ 500 = 0.22 (22.0%)

3. Conditional Relative Frequency: The proportion of observations within a restricted row or column (the given condition). The denominator is strictly the subtotal of that specific row or column, NOT the grand total:

   Conditional Relative Frequency = Cell Count ÷ Specific Row or Column Total
   

   Example A: Among the vaccinated students, what proportion contracted the flu?

   • Condition: Vaccinated row only (Row Total = 200).
   • Computation: 20 ÷ 200 = 0.10 (10.0%) Example B: Among the unvaccinated students, what proportion contracted the flu?
   • Condition: Unvaccinated row only (Row Total = 300).
   • Computation: 90 ÷ 300 = 0.30 (30.0%)

🔍 Statistical Insight: Comparing conditional relative frequencies across groups reveals association. Because 30% of unvaccinated students contracted the flu compared to only 10% of vaccinated students, an association exists between vaccination status and flu incidence.

---

5. Deceptive and Misleading Graph Tactics

Standardized tests regularly assess the ability to identify misleading graphical tactics that distort data:

1. Truncated Vertical Axis (Axis Not Starting at Zero): When the vertical axis of a bar chart begins at a non-zero value, visual height differences are wildly exaggerated. If Company A earns $98 million and Company B earns $100 million on an axis starting at $95 million, Company B's bar appears 5 units tall compared to Company A's 3 units tall (a 67% visual height difference), even though actual revenue differs by only 2%.
2. Non-Uniform or Inconsistent Scale Intervals: Unequal spacing between tick marks along either axis compresses or artificially stretches visual trends, creating false impressions of rapid growth or stability.
3. Disproportionate Pictographs: When pictures or icons replace standard bars, scaling an icon's height to represent double the quantity often causes the artist to double its width as well. Because surface area scales quadratically (Area = Width × Height = 2 × 2 = 4), the icon occupies four times the area, visually misleading the observer into perceiving a fourfold increase.
4. Three-Dimensional Perspective Distortion: Tilting bar charts or pie charts into 3D perspective causes wedges or bars in the foreground to appear substantially larger than equal or larger quantities positioned in the background.

---

6. TSIA2 Exam Traps & Strategic Checkpoints

• Trap 1: The Denominator in Conditional Probability/Frequency: Always read the condition phrasing carefully. "What fraction of women preferred Option A?" restricts the denominator to all women. "What fraction of Option A supporters were women?" restricts the denominator to all Option A supporters.
• Trap 2: Bar Spacing Diagnostic: If adjacent vertical bars touch, the graphic is a histogram depicting quantitative bins. If gaps exist between bars, the graphic is a bar chart depicting categorical groups.
• Trap 3: Pie Chart Sector Validation: The sum of percentages in any valid pie chart must equal 100% (or very close to 100% due to rounding), and total central angles must sum to 360°. If categories allow multiple responses per person, a pie chart is invalid.