7.1 Data Representation & Graph Interpretation

Introduction to Data Representation & Categorical Displays

Data analysis begins with collecting, organizing, and displaying information so that patterns, trends, and comparisons become clear. On the Praxis 5003 exam, elementary educators are tested on their ability to interpret various graph types, convert raw data into structured visual representations, and identify flaws in graphical design. Data is broadly divided into two main categories:

1. Categorical (Qualitative) Data: Data separated into distinct groups or categories based on non-numerical characteristics (e.g., favorite subject, eye color, mode of transportation).
2. Numerical (Quantitative) Data: Data consisting of counts or measurable quantities (e.g., student test scores, heights, number of pets owned).

Tally Charts & Frequency Tables

Before building complex graphs, raw data is often compiled into a tally chart or a frequency table. A tally mark represents a single observation, grouped in bundles of five (four vertical lines with a diagonal strike through them). A frequency table lists each category alongside its corresponding frequency (total count) and, optionally, its relative frequency (the fraction or percentage of the total count).

Bar Graphs & Double Bar Graphs

A bar graph uses rectangular bars of equal width to represent categorical data. The height or length of each bar corresponds to its frequency. Key properties of bar graphs include:

• Gaps Between Bars: Because the categories are discrete and distinct, there MUST be spaces between adjacent bars.
• Axis Labels & Uniform Scales: The categorical axis lists discrete labels, while the frequency axis must begin at zero and maintain uniform intervals.
• Double Bar Graphs: Used to compare two distinct groups across the same categories simultaneously (e.g., comparing Grade 4 vs. Grade 5 preferred lunch options). A clear key or legend is mandatory to distinguish the two sets of bars.

Circle Graphs (Pie Charts)

A circle graph (pie chart) illustrates how a total dataset (representing $100%$) is divided into proportional fractional parts or percentages. Each slice of the pie is a sector.

On the Praxis exam, you will frequently be required to calculate either the percentage of a category or the size of its central angle in degrees. Because a full circle measures $360^\circ$, the central angle formula is:

$\text{Central Angle} = \text{Relative Frequency (as a decimal)} \times 360^\circ = \frac{\text{Category Frequency}}{\text{Total Frequency}} \times 360^\circ$

Worked Example: In a class of 24 elementary students, 6 choose Soccer as their favorite sport. What central angle should represent Soccer in a circle graph?

1. Calculate the relative frequency fraction: $\frac{6}{24} = \frac{1}{4} = 0.25$ ($25%$).
2. Multiply by $360^\circ$: $0.25 \times 360^\circ = 90^\circ$.
3. Conclusion: The Soccer sector requires a right angle ($90^\circ$), occupying exactly one-quarter of the circle graph.

Numerical Data Visualizations: Line Plots, Stem-and-Leaf Plots, and Histograms

When working with quantitative numerical data, elementary math curricula emphasize displays that preserve individual data points or group data into continuous range intervals.

Line Plots (Dot Plots)

A line plot (also called a dot plot) displays individual data points as dots or "X" marks stacked vertically above a horizontal number line. Line plots provide an immediate visual depiction of the dataset's shape, spread, and central tendencies.

Number of Books Read During Summer Reading Challenge:
          X
          X     X
    X     X     X     X
    X  X  X     X     X     X
----+--+--+--+--+--+--+--+--+--+--
    1  2  3  4  5  6  7  8  9  10

From the line plot above, candidates can quickly extract statistical information:

• Mode (Most Frequent): 3 books (tallest stack with 4 marks).
• Range: $9 - 1 = 8$ books.
• Clusters & Gaps: A cluster occurs between 1 and 3 books, and another between 5 and 7 books. A gap exists at 4 books.
• Outliers: 9 books represents an isolated data point higher than the rest of the distribution.

Stem-and-Leaf Plots

A stem-and-leaf plot organizes numerical data by splitting each data value into two components based on place value:

• Stem: The leading digit(s) (e.g., tens digit or hundreds digit).
• Leaf: the trailing final digit (the ones digit).

Leaves are listed horizontally in ascending numerical order next to their respective stem. A stem-and-leaf plot retains all original raw data values while presenting a histogram-like visual profile.

Example Stem-and-Leaf Plot: Fourth-Grade Math Quiz Scores ($n = 15$)

Stem | Leaf
  6  | 2  5  8
  7  | 0  4  4  7  9
  8  | 1  3  5  8
  9  | 0  2  6
Key: 7 | 4 = 74 points

• Extracting Data: The full list of scores is $62, 65, 68, 70, 74, 74, 77, 79, 81, 83, 85, 88, 90, 92, 96$.
• Finding the Median: For $n = 15$, the median is the 8th score in order. Counting 8 leaves from the top yields $79$.
• Identifying the Mode: The leaf '4' appears twice on stem '7', indicating $74$ is the mode.

Histograms vs. Bar Graphs

A histogram looks visually similar to a bar graph, but it displays continuous numerical data divided into adjacent, non-overlapping intervals called bins or classes (e.g., 0–9, 10–19, 20–29).

Boxplots (Box-and-Whisker Plots)

A boxplot summarizes a distribution with five-number information: minimum, first quartile ($Q_1$), median, third quartile ($Q_3$), and maximum (whiskers may stop at adjacent values when outliers are marked separately). The box spans the interquartile range ($IQR = Q_3 - Q_1$). Use boxplots to compare centers and spreads of two groups quickly—Praxis may ask which class has the greater median or the wider IQR.

Scatterplots

A scatterplot displays paired quantitative data as points $(x, y)$. Look for overall pattern:

• Positive association (as $x$ increases, $y$ tends to increase) versus negative association
• Roughly linear versus curved (nonlinear)
• Clusters and outliers that break the pattern

ETS discussion prompts ask you to describe relationships shown by data on a scatterplot using this language.

Analyzing Distributions & Identifying Misleading Graphs

Distribution Shapes

Analyzing the overall shape of a graph is a core standard in data analysis:

• Symmetric Distribution: The left and right halves of the graph mirror each other around a central peak.
• Skewed Right (Positively Skewed): The majority of the data is concentrated on the left side, with a long tail stretching toward higher values on the right.
• Skewed Left (Negatively Skewed): The majority of the data is concentrated on the right side, with a long tail stretching toward lower values on the left.
• Uniform Distribution: All data intervals have approximately equal frequencies, forming a flat profile.
• Bimodal Distribution: The graph displays two distinct peaks, indicating two different common clusters.

Spotting Misleading Graphs (Praxis Exam Traps)

Graphics are sometimes constructed—either accidentally or intentionally—to distort data interpretation. On Praxis 5003, you will be asked to identify why a specific graph is misleading.

1. Truncated Vertical Axis (Broken Y-Axis): Starting the vertical frequency axis at a non-zero value (e.g., starting at 90 instead of 0) visually exaggerates minor differences between bars. A bar with a value of 95 will look five times taller than a bar with a value of 91.
2. Unequal Interval Scaling: Spacing numerical intervals unevenly along the axis (e.g., 0, 5, 10, 20, 50, 100) distorts the slope of line graphs and the area of histogram bars.
3. Pictograph & 3D Distortion: Scaling pictograph icons or 3D bars in both width and height simultaneously quadruples the visual surface area when the data value only doubles. 3D perspective tilts also make front bars appear disproportionately larger than rear bars.
4. Inappropriate Graph Selection: Utilizing a line graph to connect discrete, unrelated categorical data (e.g., connecting apple prices to orange prices with a continuous trend line) wrongly implies a continuous time trend.