9.2 Frequency Tables, Bar Graphs, Line Graphs, & Histograms

Key Takeaways

  • Data is broadly classified as categorical (qualitative labels/groups) or numerical (quantitative measurements), with numerical data further split into discrete (countable) and continuous (measurable continuum).
  • Frequency tables organize raw observations into counts, relative frequencies (proportions or percentages: f / n), and cumulative frequencies.
  • Bar graphs display discrete categorical data separated by spaces, whereas histograms display continuous numerical data partitioned into contiguous, equal-width intervals (bins) where bars touch.
  • Line graphs illustrate quantitative changes over sequential time intervals, where segment slope represents rate of change, while misleading graphs frequently exploit truncated y-axes to exaggerate minor variations.
Last updated: August 2026

9.2 Frequency Tables, Bar Graphs, Line Graphs, & Histograms

Statistical literacy on the TABE 13&14 Mathematics assessment requires the ability to extract, synthesize, and evaluate data presented across diverse graphical formats. Questions test your understanding of data types, proportional reasoning with frequency tables, comparative analysis of bar and line graphs, structural differences in histograms, and critical evaluation of misleading visual axes.


Categorical vs. Numerical Data

Before selecting or interpreting a statistical display, you must identify the nature of the underlying data:

Data Classification Hierarchy:
├── Categorical (Qualitative): Labels, attributes, or classifications (e.g., job titles, blood types, car colors)
└── Numerical (Quantitative): Quantities represented by real numbers with meaningful arithmetic
    ├── Discrete: Countable integer values with distinct gaps (e.g., number of students, inventory units, defect count)
    └── Continuous: Measurable quantities along an unbroken real continuum (e.g., weight, time, height, voltage)
FeatureCategorical (Qualitative)Numerical Discrete (Quantitative)Numerical Continuous (Quantitative)
DefinitionNon-numerical categories or namesCountable whole itemsMeasurable real intervals
ExamplesDepartment name, shift (Day/Night), genderNumber of children, traffic citations, defective boltsTime elapsed, athlete weight, room temperature
Optimal DisplayFrequency Table, Bar Graph, Circle / Pie ChartFrequency Table, Bar Graph, Dot PlotHistogram, Box Plot, Line Graph (time-series)

Frequency Tables & Relative Frequency

A frequency table summarizes raw datasets by grouping observations into categories or intervals and listing the frequency ($f$)—the number of times each value occurs.

Key Frequency Metrics

  1. Sample Size ($n$): The total sum of all individual frequencies: $n = \sum f$.
  2. Relative Frequency: The fraction, decimal, or percentage of total observations belonging to a specific category: Relative Frequency=Frequency (f)Total Observations (n)Percentage=(fn)×100%\text{Relative Frequency} = \frac{\text{Frequency } (f)}{\text{Total Observations } (n)} \qquad | \qquad \text{Percentage} = \left(\frac{f}{n}\right) \times 100\%
  3. Cumulative Frequency: The running cumulative sum of frequencies up to and including the current interval.

Worked Example: Workplace Certification Exam Results

Suppose $50$ employees take a technical safety certification exam. Scores are compiled into intervals:

Score Interval (Bins)Tally CountFrequency ($f$)Relative Frequency (Fraction)Relative Frequency (%)Cumulative Frequency
$60 - 69$$$
$70 - 79$$\
$80 - 89$$\
$90 - 100$$\
Total-$n = 50$$\frac{50}{50} = 1.00$$100%$$50$

TABE Question Style: What percentage of employees scored $80$ or higher? Sum the relative percentages for $[80-89]$ and $[90-100]$: $40% + 28% = \mathbf{68%}$ (or $\frac{20 + 14}{50} = \frac{34}{50} = 68%$).


Bar Graphs: Single, Double, & Segmented

Bar graphs visually compare frequencies across distinct categorical groups. The lengths or heights of parallel rectangular bars represent categorical counts or values.

Types of Bar Graphs

  1. Single Bar Graph: Displays one quantitative measure across multiple categories (can be vertical columns or horizontal bars).
  2. Double (Grouped) Bar Graph: Places two parallel bars side-by-side for each category to compare sub-groups directly (e.g., comparing 2024 vs. 2025 revenue across four corporate divisions).
  3. Segmented (Stacked) Bar Graph: Stacks multiple sub-group components into a single composite bar of height $100%$ (or total sum), showing relative part-to-whole proportions.
Double Bar Graph Comparison (Branch Revenue in $ thousands):

$80k |   [2024] [2025]
$60k |     █      █       [2024] [2025]
$40k |     █      █         █      █       [2024] [2025]
$20k |     █      █         █      █         █      █
 $0k +-------------------------------------------------
         North Branch     East Branch      West Branch

Analytical Reading Strategy for Bar Graphs

  • Check the scale increments on the numerical axis (e.g., does each grid line represent $5, 10, 50,$ or $1,000$ units?).
  • Read values from the top edge of each bar.
  • Calculate differences, totals, or percentage increases directly: $\text{Percentage Increase} = \frac{\text{New} - \text{Old}}{\text{Old}} \times 100%$.

Line Graphs: Trends Over Time & Rate of Change

A line graph (time-series plot) displays data points connected by straight line segments to track how a quantitative variable changes over continuous or sequential time intervals (hours, days, months, years).

Slope as Rate of Change

The steepness and direction of each line segment reflect the rate of change between consecutive time intervals:

  • Positive Slope (Upward): Value is increasing over time.
  • Negative Slope (Downward): Value is decreasing over time.
  • Horizontal Segment (Zero Slope): Value remains unchanged (plateau / constant).
  • Steepness: A steeper segment indicates a faster rate of change than a flatter segment.

Average Rate of Change=ΔyΔx=y2y1x2x1\text{Average Rate of Change} = \frac{\Delta y}{\Delta x} = \frac{y_2 - y_1}{x_2 - x_1}

Worked Example: A regional logistics center tracks package deliveries over a 5-month period: January ($4,200$), February ($4,800$), March ($4,800$), April ($6,000$), May ($5,400$).

  • Fastest Growth Interval: From March to April: $\frac{6,000 - 4,800}{4 - 3} = \mathbf{+1,200\text{ packages/month}}$.
  • Zero Growth Interval: From January to February was $+600$, but from February to March the graph is flat (slope $= 0$, growth $= 0$).
  • Decline Interval: From April to May: $\frac{5,400 - 6,000}{5 - 4} = \mathbf{-600\text{ packages/month}}$.

Histograms vs. Bar Graphs

A histogram looks similar to a vertical bar graph, but it represents fundamentally different data. Understanding this distinction is heavily tested on TABE Levels D and A.

FeatureBar GraphHistogram
Data TypeCategorical or discrete qualitative labelsNumerical Continuous data grouped into bins
Horizontal Axis ($x$)Discrete categories (e.g., Red, Blue, Green)Continuous numerical intervals (e.g., $10–19, 20–29$)
Bar SpacingSpaces / gaps between barsBars touch (no gaps unless a bin has $0$ frequency)
Order of BarsArbitrary (can be alphabetical or by size)Strict numerical order from lowest to highest
Width of BarsArbitrary visual stylingRepresents the interval width (bin size)
Histogram Structure (Equal Bin Width = 10 units, Bars Touch Contiguously):

Frequency (f)
  12 |         [████]
   8 |     [████████]
   4 | [████████████████]
   0 +------------------------
      10-19 20-29 30-39 40-49   <-- Continuous Numerical Bins

Interpreting Histogram Shapes

  1. Symmetric (Bell-Shaped): Single central peak with frequencies tapering off equally to the left and right.
  2. Skewed Right (Positively Skewed): Main cluster of data is on the left; the tail stretches toward higher numerical values on the right.
  3. Skewed Left (Negatively Skewed): Main cluster of data is on the right; the tail stretches toward lower numerical values on the left.
  4. Bimodal: Two distinct prominent peaks separated by a trough.

Misleading Graphical Displays & Axis Distortions

Standardized tests frequently challenge candidates to spot deceptive graphical representations used in media and advertising:

  1. Truncated $y$-Axis (Broken Axis Trap): When the vertical axis starts at a non-zero number (e.g., starting at $950$ instead of $0$) without a clear zigzag break symbol (//). This exaggerates small differences between values, making a $2%$ increase look like a $400%$ surge.
  2. Inconsistent Scale Increments: Using non-uniform axis intervals (e.g., jumping from $0 \to 10 \to 20 \to 50 \to 100$) to artificially compress or stretch trends.
  3. Pictograph Area Distortions: Doubling both the width and height of an icon when value doubles causes the 2D visual area to quadruple ($2^2 = 4$), severely misleading the viewer.
Loading diagram...
Taxonomy of Statistical Data Displays
Test Your Knowledge

A civil engineer records the travel speeds of 200 motor vehicles passing through a work zone. Speeds range from 32 mph to 68 mph and are grouped into intervals of 5 mph. Why is a histogram the most appropriate graphical display for this data rather than a standard bar graph?

A
B
C
D
Test Your Knowledge

A production facility tracks weekly output over four consecutive weeks: Week 1 = 1,400 units, Week 2 = 1,750 units, Week 3 = 1,750 units, Week 4 = 2,350 units. On a time-series line graph, what was the average rate of change in production output from Week 2 to Week 4?

A
B
C
D
Test Your Knowledge

A corporate report presents a vertical column chart comparing annual revenue between Division X ($50 million) and Division Y ($52 million). The column for Division Y appears three times as tall as the column for Division X. What graphical distortion is responsible for this misleading representation?

A
B
C
D