Visualization and Interpretation of Categorical and Quantitative Data

Visualizing Categorical Variables

Categorical variables, such as fields of study like Biological Science and Physics, are represented using specific graphical tools. Two primary methods for displaying this data include pie charts and bar graphs.

• Pie Charts: This visualization utilizes a circle where each category is represented as a "slice of the pie." The size of each slice corresponds to the percentage of individuals or observations within that specific category. For instance, a pie chart might show the percentage distribution of students across different majors.
• Bar Charts / Bar Graphs: These use boxes (bars) to represent categories. While the concept is similar to a pie chart, the visualization differs. In a bar graph, the height of the box (the "tallest box") indicates a higher frequency or a greater number of individuals within that category.
• Differences in Representation: When looking at sources of information used by Americans aged $34$ years, the percentages on a bar graph do not necessarily have to sum to a single fixed number (like $100\%$) depending on the survey structure.

Quantitative Data Visualization: Histograms

For quantitative variables, histograms and stem-and-leaf plots are the appropriate visualization tools. These are critical for examinations and are distinct from categorical tools; for example, a histogram cannot be used for categorical variables such as "AOD."

• Constructing a Histogram:
  • Divide the possible values into distinct classes or intervals.
  • Chop the data into specific "windows" or intervals (bins).
  • Count the number of observations that fall into each interval.
• Example Analysis: If data is divided into $9$ different intervals, the count in each interval represents the number of observations (e.g., the number of states).
• Frequency Calculation Example: To find the number of freshmen with a graduation rate between $85$ and $96$, one might add the frequencies of the corresponding bars. For example, if one bar represents $11$ and another represents $14$, the total is $11 + 14 = 25$.
• Specific Intervals: In an interval such as $85$ to $90$, the height of the bar indicates exactly how many freshmen graduation rates fall within that specific range.

Key Features of Data Distributions

When analyzing quantitative data distributions via histograms or stem plots, there are three primary characteristics to consider:

1. Shape: The overall visual pattern of the data. In this course, there are three specific types of shapes often encountered. Identifying the shape is a common exam requirement.
2. Center: The midpoint of the data. Specific measurements for the center include the $\text{mean}$ and the $\text{median}$.
3. Variability (Spread): How much the data deviates or spreads out. Measurements for variability include $\text{standard deviation}$ and the $\text{five-number summary}$.

Outliers in Data Sets

An outlier is defined as an individual observation that does not follow the general distribution of the data or the overall pattern. It sits outside the established visual trend.

• Example of Outliers: In a sample of weights containing values like $100$, $120$, and $140$, a weight of $300$ or $350$ would be considered an outlier because it falls far outside the range of the other observations.

Stem-and-Leaf Plots (Stem Plots)

Stem-and-leaf plots are a highly structured way to visualize quantitative data. They are not used for categorical variables.

• Construction Steps:
  • Determine the Range: Identify the minimum and maximum values (e.g., a range from approximately $100$ to $140$).
  • The Stem: These are the first digits of the numbers. For a range of $100$ to $140$, the stems would be $10, 11, 12, 13, 14$, representing the hundreds and tens places.
  • The Leaf: This is the last digit of the number. For a value like $102$, the stem is $10$ and the leaf is $2$. If there are two instances of $102$, two $2$s are placed next to the stem $10$. A value of $107$ would place a $7$ as a leaf next to the $10$ stem.
  • Navigating the Plot: If a stem has no corresponding data points (e.g., nothing in the $110$ range), it remains empty, and you move to the next stem (e.g., $120$).

Splitting Stems in Stem Plots

Sometimes a standard stem plot does not show the data clearly enough, and "splitting the stems" is required for better visualization.

• The Process: Each stem is listed twice.
• Leaf Distribution:
  • The first instance of the stem (e.g., the first $10$) houses leaves ranging from $0$ to $4$.
  • The second instance of the stem (e.g., the second $10$) houses leaves ranging from $5$ to $9$.
• Example: If you have the values $102$ and $107$, the $2$ goes in the first $10$ stem (because $2 < 5$), and the $7$ goes in the second $10$ stem (because $7 \geq 5$).

Course Tasks and Grading

• Assignments: Students are expected to know how to hand-draw bar graphs and pie charts, as well as perform interpretations of existing diagrams.
• Grading Structure: Points are assigned to each chapter. Assignments designated as "optional" do not carry a grade and do not contribute to the final point total.
• Required Action: It is mandatory to sign up for the chapter software to access and complete the required assignments.